461135 - Audio does not play in <audio> or <video> elements with PulseAudio

Status: RESOLVED FIXED

(Core :: Audio/Video, defect)

Description

[cajbir (:cajbir)]

When PulseAudio is used as the sound manager, audio does not playback if another application is playing sound.

Steps to Reproduce:
1. Visit URL in bug.
2. Start mplayer (or another sound app) and play some audio.
3. Click play on page from (1).

What should happen:

Video and audio should play.

What does happen:

Video plays but there is no audio.

Comment 1

[cajbir (:cajbir)]

The problem lies with the alsa backend in libsydneyaudio. This has recently been fixed and upgrading to this fixes the issue.

Comment 2

[cajbir (:cajbir)]

Attachment: Update to latest libsydneyaudio alsa backend

Comment 3

[Reed Loden [:reed]]

Why is code being committed without any type of review at all? I understand that this code is from an outside source, but you could at least get rs= for landing it rather than just landing it without anybody else taking a peek. Mozilla's source includes a ton of third-party modules, but (hopefully) each new update to our in-tree copies has been looked at by more than one person. I don't think it's too much to ask that you get rs= (rubber-stamp) from some other person before just landing large changes like this.

Referring to:
http://hg.mozilla.org/mozilla-central/rev/ec6f908d16c0

Comment 4

[cajbir (:cajbir)]

I was told by roc that commit the media third party updates without review was fine as long as they'd been reviewed by the upstream. Roc explitly told me he didn't want to review third party code and this was the approach to take.

Comment 5

[cajbir (:cajbir)]

Reed, see bug 448636 comment 9. I confirmed later when I pushed other media libraries as I was unsure about reviews.

Comment 6

[cajbir (:cajbir)]

http://hg.mozilla.org/mozilla-central/rev/ec6f908d16c0

Comment 7

[Robert O'Callahan (:roc) (email my personal email if necessary)]

(In reply to comment #3)
> (hopefully) each
> new update to our in-tree copies has been looked at by more than one person.

Sure, but for upstream libraries, the non-Mozilla people are suitable reviewers, not some random Mozilla person.

> I
> don't think it's too much to ask that you get rs= (rubber-stamp) from some
> other person before just landing large changes like this.

Chris has my rubber-stamp. Rubber-stamp means the person didn't look at the code.

Comment 8

[Robert O'Callahan (:roc) (email my personal email if necessary)]

In the future it would be good to note rs=roc in the commit message(s).