Allow sample rate pass through on Linux.

Allow sample rate pass through on Linux.

Requiring the native output sample rate to avoid glitching appears
to have just been an issue with a poor tlength choice.  xians and
I resolved this prior to turning PulseAudio on by default, but left
the native sample rate requirement in at the time.

Since "native sample rate" can change on the fly with PulseAudio
our approach of specifying FIX_RATE was incorrect.  Allowing Pulse
to handle resampling lets us remove the FIX_RATE flag.  It also
improves CPU usage in cases where Pulse can configure the output
device to match the requested sample rate.

BUG=229918
TEST=extensive manual checks for glitching.
R=xians@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=202233

[DaleCurtis, 2013-05-24 00:40:12]

xians: Can you patch this in locally and see if you hear any problems?  Since we
adjusted the tlength in https://chromiumcodereview.appspot.com/12328097/ I don't
believe we need the FIX_RATE anymore.

[no longer working on chromium, 2013-05-24 08:43:03]

On 2013/05/24 00:40:12, DaleCurtis wrote:
> xians: Can you patch this in locally and see if you hear any problems?  Since
we
> adjusted the tlength in https://chromiumcodereview.appspot.com/12328097/ I
don't
> believe we need the FIX_RATE anymore.

Sure, I will try it today and come back to you. Actually I did test the pulse
sound without resampler, and could not hear glitch at that time, but I just
thought it might be because of our powerful machine.

[no longer working on chromium, 2013-05-24 13:59:15]

On 2013/05/24 08:43:03, xians1 wrote:
> On 2013/05/24 00:40:12, DaleCurtis wrote:
> > xians: Can you patch this in locally and see if you hear any problems? 
Since
> we
> > adjusted the tlength in https://chromiumcodereview.appspot.com/12328097/ I
> don't
> > believe we need the FIX_RATE anymore.
> 
> Sure, I will try it today and come back to you. Actually I did test the pulse
> sound without resampler, and could not hear glitch at that time, but I just
> thought it might be because of our powerful machine.

I did some simple tests, it sounds OK and no glitch on my machine.
lgtm

[commit-bot: I haz the power, 2013-05-24 20:32:26]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/dalecurtis@chromium.org/15957002/1

[DaleCurtis, 2013-05-24 23:54:52]

Committed patchset #1 manually as r202233 (presubmit successful).