Make sure to use scoped_refptr for refcounted params in audio/video/remoting

Make sure to use scoped_refptr for refcounted params in audio/video/remoting

BUG=28083
TEST=existing

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=62949

[awong, 2010-10-18 16:32:11]

Hey Will,

I don't think I 100% understand this change.  Most of our code doesn't actually
use shared ownership semantics even though our objects are refpointers. That is,
we know that the messageloop is silent before we release the last ref on the
object.

I spot checked a couple of the calls here, and the make_scoped_refptr() actually
ends up just binding to a raw pointer anyways...

-Albert

[scherkus (not reviewing), 2010-10-18 16:35:05]

Believe it's motivated by:
http://groups.google.com/a/chromium.org/group/chromium-dev/browse_thread/thre...

[willchan no longer on Chromium, 2010-10-18 17:19:27]

Nope, it's motivated by wanting to land
http://codereview.chromium.org/3549010/show to fix
http://code.google.com/p/chromium/issues/detail?id=28083.

On Mon, Oct 18, 2010 at 9:35 AM, <scherkus@chromium.org> wrote:

> Believe it's motivated by:
>
>
http://groups.google.com/a/chromium.org/group/chromium-dev/browse_thread/thre...
>
>
> http://codereview.chromium.org/3821005/show
>

[awong, 2010-10-18 17:50:43]

LGTM

Chatted offline.  In this case, we're going to cost an unnecessary ref/deref,
but it's worth it to unblock checking in Will's CL.