Return-Path: <cyrus@holtmann.org>
Date: Mon, 4 Nov 2013 14:15:39 +0100
From: Stevie Trujillo <stevie.trujillo@gmail.com>
To: Hui Wang <hui.wang@canonical.com>, marcel@holtmann.org,
 tedd.an@intel.com
Cc: linux-bluetooth@vger.kernel.org
Subject: Re: Intel 7260 bluetooth malfunction when it is connected to EHCI
 bus
Message-ID: <20131104141539.772e1636@localhost>
In-Reply-To: <527737C7.40208@canonical.com>
References: <526F8804.9060906@canonical.com>
	<A07820C9-C272-4AD4-AE16-49F81956C2D7@holtmann.org>
	<52707428.9070908@canonical.com>
	<6091DDC7-7B00-4457-82A0-F8CC9B4EE9DD@holtmann.org>
	<5271EAA2.1030705@canonical.com>
	<D0B840E3-36C4-4751-A757-1ADDAC0707B9@holtmann.org>
	<5273147E.4040004@canonical.com>
	<C3BBD7E9-6D91-430E-96DF-2D4885F89DCC@holtmann.org>
	<527371D4.1010305@canonical.com>
	<9DFD9618-158E-41C4-8183-134A3FC4E61F@holtmann.org>
	<527737C7.40208@canonical.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
List-ID: <linux-bluetooth.vger.kernel.org>

Hello, I've been trying to use my computer as "handsfree" with
"Hands-Free Profile" using "Intel Corporation Centrino Wireless-N
1030". I accept() the BTPROTO_SCO socket, but never recv() any
data. Your post looked similar!

When googling I found this:

http://www.netbsd.org/docs/guide/en/chap-bluetooth.html#chap-bluetooth-audio
   SCO Audio will not work with ehci(4) USB controllers, since support
   for Isochronous data over EHCI is missing in NetBSD. 

Does Linux support this? (I have no idea what it means)

I'm sorry if this mail is somewhat broken. I could only find your email
on "gmane" and it had encrypted all your email addresses!

On Mon, 04 Nov 2013 13:59:35 +0800
Hui Wang <hui.wang-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org> wrote:

> On 11/03/2013 03:05 AM, Marcel Holtmann wrote:
> > Hi Hui,
> >
> <snip>
> >>
> >> The attachment lenovo-k4450-btmon-v1.log is the log of intel 7260
> >> init procedure + scan headset + connect headset + play sound.
> > I looked through the logs and the controller init is just fine. So
> > in theory this should route the SCO audio frames over HCI. For some
> > reason it does not and I have no idea why not.
> >
> > That it makes a different between EHCI and XHCI is strange. I can
> > not explain this. Maybe Tedd can help you since I have no further
> > insights in our chips at this moment.
> 
> Thanks Marcel for your kind help and analysis.
> 
> Tedd, did you reproduce this issue on the machines with intel 7260 BT 
> module connected to the EHCI bus? Did you have any idea on this issue?