Always use the source channel layout with AudioBufferConverter.

Always use the source channel layout with AudioBufferConverter.

There are no web sites currently using this feature, but it causes
all multichannel users to waste resources unnecessarily upmixing,
generating, and transferring empty channel data.

Further this breaks fancy OSX multichannel users which have non
standard channel layouts.  Mostly because we don't want to deal
with the plethora of ways OSX wants to map channels ourselves.

Instead we can require that MSE users which want to change channel
counts, must specify the maximum number of channels in the first
initialization segment.

This also fixes use cases where clients routed multichannel streams
into WebAudio for processing instead of actual playout.

BUG=266674, 379288
TEST=Only 2 channels are created for stereo sources when a 7.1
device is connected.

R=rileya@chromium.org, wolenetz@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=274068

[DaleCurtis, 2014-05-29 23:18:08]

Tiny change, but has some impact on how MSE users will achieve channel layout
changes; so +wolenetz.

[rileya (GONE FROM CHROMIUM), 2014-05-30 00:39:46]

lgtm. I can't speak too confidently to the impact on MSE users, so I defer to
wolenetz@ there.

[DaleCurtis, 2014-05-30 18:01:40]

Note: This also fixes / unblocks hooking multi-channel streams into WebAudio on
a system with less hardware channels than the number of decoded channels.

[wolenetz, 2014-05-30 23:01:00]

lgtm % nit and similarly updating the CL title/description according to same
nit.

https://codereview.chromium.org/304233006/diff/1/media/filters/audio_renderer...
File media/filters/audio_renderer_impl.cc (right):

https://codereview.chromium.org/304233006/diff/1/media/filters/audio_renderer...
media/filters/audio_renderer_impl.cc:289: // channel layouts
(http://crbug.com/266674) only allow channel down
nit: Upmixing is still allowed, but to no higher than the channels initialized
here (which are no longer the full hw support, but whatever the first init
segment had). Adjust comment?

[DaleCurtis, 2014-05-30 23:08:56]

Description updated.

https://codereview.chromium.org/304233006/diff/1/media/filters/audio_renderer...
File media/filters/audio_renderer_impl.cc (right):

https://codereview.chromium.org/304233006/diff/1/media/filters/audio_renderer...
media/filters/audio_renderer_impl.cc:289: // channel layouts
(http://crbug.com/266674) only allow channel down
On 2014/05/30 23:01:00, wolenetz wrote:
> nit: Upmixing is still allowed, but to no higher than the channels initialized
> here (which are no longer the full hw support, but whatever the first init
> segment had). Adjust comment?

Done.

[DaleCurtis, 2014-05-30 23:09:07]

The CQ bit was checked by dalecurtis@chromium.org

[commit-bot: I haz the power, 2014-05-30 23:11:20]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dalecurtis@chromium.org/304233006/20001

[commit-bot: I haz the power, 2014-05-31 05:00:40]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  android_dbg_triggered_tests on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/android_dbg_triggered...)
  linux_chromium_chromeos_rel on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/linux_chromium_chrome...)
  win_chromium_rel on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/win_chromium_rel/buil...)

[DaleCurtis, 2014-05-31 05:56:55]

The CQ bit was checked by dalecurtis@chromium.org

[commit-bot: I haz the power, 2014-05-31 05:57:24]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dalecurtis@chromium.org/304233006/40001

[DaleCurtis, 2014-05-31 22:24:52]

Committed patchset #3 manually as r274068 (presubmit successful).