Return-Path: Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 (Mac OS X Mail 8.2 \(2104\)) Subject: Re: Intel 7260 spontaneous USB disconnect/reconnect every 23h18m07s From: Marcel Holtmann In-Reply-To: <20150825204911.GA3479@hau.nz> Date: Tue, 25 Aug 2015 14:40:33 -0700 Cc: "linux-bluetooth@vger.kernel.org" , Loic Poulain Message-Id: References: <20150819222049.15ca98f0@angwin.home> <86D221DC0B85F94D9B05B97827669A34064C4EF7@PGSMSX106.gar.corp.intel.com> <20150825165238.GA28893@hau.nz> <20150825171647.GB28893@hau.nz> <20150825191523.GA3220@hau.nz> <20150825204911.GA3479@hau.nz> To: Mark Haun Sender: linux-bluetooth-owner@vger.kernel.org List-ID: Hi Mark, >> Took me a while to figure out you have to bring down the hci0 interface and >> stop bluetoothd before running bluemoon. Otherwise I get "Failed to open >> HCI user channel." So now when I run bluemoon -T it just sits there forever >> after the welcome message. Is this expected behavior? Should I then >> restart bluetoothd in another console and leave the whole thing for a couple >> of days to catch the USB disconnect? > > Ok, I think I am seeing a bug very similar to this one reported by Loic > Poulain: > > http://marc.info/?l=linux-bluetooth&m=143220351521634&w=2 > ("bluetooth-next HCI user channel issue") > > I have the same symptom of bluemoon hanging, and I figured out that I can > recover the system without rebooting by forcing hci0 down again. "hciconfig > hci0 down" takes about ten seconds to return, and then I have a 0xfc3f > timeout in dmesg: > > [ 5563.562554] Bluetooth: hci0 command 0xfc3f tx timeout > [ 5571.557618] Bluetooth: hci0: turning off Intel device LED failed (-110) > > This is the same error Loic was getting, I assume because the commit in > question has now made it into my kernel (4.2.0-rc4). there are patches in 4.2-rc8 at least that should avoid this issue in combination with HCI User Channel. If this is still an issue, then we need to fix this. Regards Marcel