From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Courtier-Dutton Subject: Re: Bluetooth headset driver Date: Mon, 26 Apr 2004 22:01:40 +0100 Sender: alsa-devel-admin@lists.sourceforge.net Message-ID: <408D78B4.4000700@superbug.demon.co.uk> References: <1083001170.4dmq5ufy93r4@cybertux.akr.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1083001170.4dmq5ufy93r4@cybertux.akr.de> Errors-To: alsa-devel-admin@lists.sourceforge.net List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , List-Archive: To: Christian Ege Cc: alsa-devel@lists.sourceforge.net List-Id: alsa-devel@alsa-project.org Christian Ege wrote: > Hi anyone working on something simalar to this? > http://www.dcs.gla.ac.uk/~jp/snd-bt-sco/ > > The author of this driver won't continue his work because he switched to MAC OS > ;-( > > I think the way he tried to integrate this driver into alsa isn't the best way. > > I am not that kind of Kernel guru but if there are some other people arround > here interested in that stuff it shouldn't be too much work to integrate this > driver in alsa. > > Or isn't there any interest in bluetooth communication ;-) > > > cu > chege > I have been working on a bluetooth alsa driver, but it turns out to be a lot more complicated that one would at first think. 1) The bluetooth headset profile is a "profile" and therefore deemed to be a user space and not a kernel space program. 2) The current bluez bluetooth stack in the kernel does not provide any "playback" position feedback, so we would not be able to implement snd_pcm_delay() and therefore it would be extremely difficult to make a low latency app using it. The first use of a headset profile is for full duplex speach, which requires low latency, so that is a problem. 3) audio data path would therefore go from alsa-lib(userspace) -> alsa-driver(kernel-space) -> headset profile (userspace) -> bluez stack (kernel space). Extremely messy in my view! (What http://www.dcs.gla.ac.uk/~jp/snd-bt-sco/ currently does in fact) Summary: - Until the bluez stack provides a mechanism for us to do "snd_pcm_delay()", there is very little point advancing with the alsa side for bluetooth. Who will want to use a headset that has large uncontrollable latencies? Cheers James ------------------------------------------------------- This SF.net email is sponsored by: The Robotic Monkeys at ThinkGeek For a limited time only, get FREE Ground shipping on all orders of $35 or more. Hurry up and shop folks, this offer expires April 30th! http://www.thinkgeek.com/freeshipping/?cpg=12297