On Tue, Apr 17, 2012 at 04:29:01PM +0200, David Henningsson wrote: > On 04/17/2012 10:08 AM, Tanu Kaskinen wrote: > >In my mind a port is an "atomic routing unit", and a port > >that can be used for both input and output is not atomic (it > >defines two routes). An example of practical problems having > >two routes in one port: you can't set a property for output > >without setting it also for input. > Mark, do you have any opinion on this? IIRC you've been advocating > for keeping e g headset jacks as one entity, but my memory might > fail me. It doesn't really matter what PulseAudio does internally so long as it comes out with the right answer at the UI level. The thing that concerns me with bundling things together is that the kernel needs to provide enough information to userspace to let it do things like figure out that there's a headset jack rather than distinct headphone and microphone jacks or whatever as this can to make a difference to user expecations. > Also does it have any implication on the pending UCM patches? I'd intuitively expect it to make no difference, UCM does separate out the input and output sides in its data. > For desktops / HDA, we only have single-direction ports so far, and > I don't see that changing in the foreseeable future, so not much to > worry about there. This is definitely *not* true - as I have repeatedly pointed out all the Apple hardware has headset jacks that are compatible with the iDevice headsets. See the recent thread from Kevin Hillman on alsa-devel... -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 836 bytes Desc: Digital signature URL: <http://lists.freedesktop.org/archives/pulseaudio-discuss/attachments/20120417/91d2a663/attachment.pgp>
