MediaSource must use ref-counting to handle pending activity counting.

MediaSource must use ref-counting to handle pending activity counting.

Switch MediaSource to deriving from RefCountedGarbageCollected, as
ActiveDOMObject::setPendingActivity() relies on ref-counting.

R=haraken@chromium.org
BUG=340522
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=177324

[sof, 2014-07-01 15:50:44]

Please take a look.

Some mediasource failures seen,


https://storage.googleapis.com/chromium-layout-test-archives/WebKit_Linux_Oil...

[haraken, 2014-07-01 16:02:01]

I wonder why the current oilpan build is working with
GarbageCollectedFinalized<MediaSource>. If MediaSource is using
setPendingActivity(), won't it cause a compile error if we use
GarbageCollectedFinalized<MediaSource>?

https://codereview.chromium.org/360323002/diff/1/Source/core/html/HTMLMediaEl...
File Source/core/html/HTMLMediaElement.cpp (right):

https://codereview.chromium.org/360323002/diff/1/Source/core/html/HTMLMediaEl...
Source/core/html/HTMLMediaElement.cpp:392: closeMediaSource();

This looks problematic. You're touching m_mediaSource in a destructor.

[sof, 2014-07-01 16:12:37]

On 2014/07/01 16:02:01, haraken wrote:
> I wonder why the current oilpan build is working with
> GarbageCollectedFinalized<MediaSource>. If MediaSource is using
> setPendingActivity(), won't it cause a compile error if we use
> GarbageCollectedFinalized<MediaSource>?
> 

No, the ref-forwarding on HTMLMediaSource hides the problem.

>
https://codereview.chromium.org/360323002/diff/1/Source/core/html/HTMLMediaEl...
> File Source/core/html/HTMLMediaElement.cpp (right):
> 
>
https://codereview.chromium.org/360323002/diff/1/Source/core/html/HTMLMediaEl...
> Source/core/html/HTMLMediaElement.cpp:392: closeMediaSource();
> 
> This looks problematic. You're touching m_mediaSource in a destructor.

[sof, 2014-07-01 16:14:11]

https://codereview.chromium.org/360323002/diff/1/Source/core/html/HTMLMediaEl...
File Source/core/html/HTMLMediaElement.cpp (right):

https://codereview.chromium.org/360323002/diff/1/Source/core/html/HTMLMediaEl...
Source/core/html/HTMLMediaElement.cpp:392: closeMediaSource();
On 2014/07/01 16:02:00, haraken wrote:
> 
> This looks problematic. You're touching m_mediaSource in a destructor.

I've reverted the HTMLMediaElement changes for now.

[haraken, 2014-07-01 16:21:00]

LGTM

https://codereview.chromium.org/360323002/diff/40001/Source/modules/mediasour...
File Source/modules/mediasource/MediaSource.h (right):

https://codereview.chromium.org/360323002/diff/40001/Source/modules/mediasour...
Source/modules/mediasource/MediaSource.h:134: HTMLMediaElement*
m_attachedElement;

You might want to address this FIXME when you make
HTMLMediaElement::m_mediaSource a Member :)

[sof, 2014-07-01 16:22:41]

https://codereview.chromium.org/360323002/diff/40001/Source/modules/mediasour...
File Source/modules/mediasource/MediaSource.h (right):

https://codereview.chromium.org/360323002/diff/40001/Source/modules/mediasour...
Source/modules/mediasource/MediaSource.h:134: HTMLMediaElement*
m_attachedElement;
On 2014/07/01 16:20:59, haraken wrote:
> 
> You might want to address this FIXME when you make
> HTMLMediaElement::m_mediaSource a Member :)

I'm just green'ing the tree, for now.

[sof, 2014-07-01 16:24:26]

The CQ bit was checked by sigbjornf@opera.com

[commit-bot: I haz the power, 2014-07-01 16:24:45]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/sigbjornf@opera.com/360323002/40001

[commit-bot: I haz the power, 2014-07-01 17:33:30]

Change committed as 177324