[vbox-dev] PulseAudio support rewrite [Patch]

[Frank Mehnert]

On Sunday 07 March 2010, Arthur Taylor wrote:
> Hello. Sorry for the week of delay, deadlines came and went. This may
> also be easier to do if I was closer to the timezones in Europe.

Hehe, that is surely no problem :) Your comments are always
welcome!

> On Sun, Feb 28, 2010 at 11:57 PM, Frank Mehnert <Frank.Mehnert at sun.com> 
wrote:
> > Note that I've debuggged the sequence and the guest will certainly
> > wait for 2-3 before it pauses the audio device! You are right that
> > the guest assumes that the pausing is very brief -- actually the
> > guest writes to an I/O port and the I/O operation wouldn't return
> > until the operation is done.
> >
> >
> > That fix goes into the right direction but again, note that we have
> > to drain the buffer earlier -- _before_ the guest pauses the sound device
> > because otherwise, short sound sequences will be delayed for the time
> > the guest waits until the sound device is paused. So I think we really
> > need to introduce a timer which triggers a drain after a short amount of
> > time (perhaps 50-100ms).
>
> The easy way to fix the loss of brief sounds is to make the requested
> prebuffer size small (smaller than the buffer_attr->tlength target
> buffer length value) as that would lessen the probability that data
> written would be smaller than the prebuffer, stalling playback start.
> That said, the prebuffer size is only requested and ultimately decided
> by the server and can change over the connection's lifetime. Such a
> fix would return to corking and flushing on device close. Also note
> that it is more correct to cork then flush than the other way around.
> This is the most correct way IMHO.
>
> The other way is not to use a prebuffer, which is close to what normal
> hardware does, playing silence when the buffer underflows. I can write
> a patch with that behavior for testing purposes, but I've experienced
> out-of-sync server side buffers with that code before and crashes with
> older server versions.
>
> The method of trying to trigger the playback after 50ms sounds a bit
> difficult as pulse's api doesn't provide facility for it, requiring
> something custom.

Well, for now I just applied a similar patch to your latest drain-2
patch. Let's ignore the issue with the possible loss of brief sounds
for the moment.

Kind regards,

Frank
-- 
Dr.-Ing. Frank Mehnert

Sitz der Gesellschaft:
Sun Microsystems GmbH, Sonnenallee 1, 85551 Kirchheim-Heimstetten
Amtsgericht München: HRB 161028
Geschäftsführer: Thomas Schröder, Wolfgang Engels
Vorsitzender des Aufsichtsrates: Martin Häring
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
URL: <http://www.virtualbox.org/pipermail/vbox-dev/attachments/20100308/1bdf3c39/attachment.sig>