Introduce AudioSourceProviderClientLockScope.

Introduce AudioSourceProviderClientLockScope.

It makes lock/unlock of AudioSourceProviderClient easier.
This class holds a strong reference to AudioSourceProviderClient. So we
don't need MediaElementAudioSourceNode::m_keepAliveWhileLocking.

BUG=395940
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=178999

[tkent, 2014-07-25 09:05:51]

Please review this.

[haraken, 2014-07-26 04:11:37]

https://codereview.chromium.org/416333002/diff/20001/Source/core/html/HTMLMed...
File Source/core/html/HTMLMediaElement.cpp (right):

https://codereview.chromium.org/416333002/diff/20001/Source/core/html/HTMLMed...
Source/core/html/HTMLMediaElement.cpp:175:
AudioSourceProviderClientLockScope(HTMLMediaElement& element)

Add explicit.

https://codereview.chromium.org/416333002/diff/20001/Source/core/html/HTMLMed...
Source/core/html/HTMLMediaElement.cpp:188:
RawPtrWillBeMember<AudioSourceProviderClient> m_client;

I wonder if we can just use RefPtrWillBeMember<AudioSourceProviderClient> here
and remove manual ref()/deref() in MediaElementAudioSourceNode.cpp. Since
AudioSourceProviderClientLockScope is stack-allocated, there would be no risk
for the RefPtr to produce a cycle.

[tkent, 2014-07-28 01:20:55]

https://codereview.chromium.org/416333002/diff/20001/Source/core/html/HTMLMed...
File Source/core/html/HTMLMediaElement.cpp (right):

https://codereview.chromium.org/416333002/diff/20001/Source/core/html/HTMLMed...
Source/core/html/HTMLMediaElement.cpp:175:
AudioSourceProviderClientLockScope(HTMLMediaElement& element)
On 2014/07/26 04:11:37, haraken wrote:
> 
> Add explicit.

Done.

https://codereview.chromium.org/416333002/diff/20001/Source/core/html/HTMLMed...
Source/core/html/HTMLMediaElement.cpp:188:
RawPtrWillBeMember<AudioSourceProviderClient> m_client;
On 2014/07/26 04:11:37, haraken wrote:
> 
> I wonder if we can just use RefPtrWillBeMember<AudioSourceProviderClient> here
> and remove manual ref()/deref() in MediaElementAudioSourceNode.cpp. Since
> AudioSourceProviderClientLockScope is stack-allocated, there would be no risk
> for the RefPtr to produce a cycle.

Done.  But do you think Patch Set 3 is better than Patch Set 2?

[haraken, 2014-07-28 01:28:43]

LGTM

> > I wonder if we can just use RefPtrWillBeMember<AudioSourceProviderClient>
here
> > and remove manual ref()/deref() in MediaElementAudioSourceNode.cpp. Since
> > AudioSourceProviderClientLockScope is stack-allocated, there would be no
risk
> > for the RefPtr to produce a cycle.
> 
> Done.  But do you think Patch Set 3 is better than Patch Set 2?

I understand that removing the manual ref/deref won't clean up things than I
expected. I'm OK with either patch set 2 or 3.

[tkent, 2014-07-28 01:38:35]

I'll land Patch Set 2 + explicit fix.

[tkent, 2014-07-28 01:38:40]

The CQ bit was checked by tkent@chromium.org

[commit-bot: I haz the power, 2014-07-28 01:38:54]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/tkent@chromium.org/416333002/60001

[commit-bot: I haz the power, 2014-07-28 02:41:35]

Change committed as 178999