Oilpan: Remove adoptRefCountedGarbageCollected

Oilpan: Remove adoptRefCountedGarbageCollected

adoptRefCountedGarbageCollected is no longer needed. When creating a RefCountedGarbageCollected object, we can just use 'new RefCountedGarbageCollectedObj()'. Even if a GC is triggered between when the object is created and when the object is adopted, there is no issue because our conservative scanning finds the object and traces it.

This CL also removes ThreadSafeRefCountedGarbageCollected because it is now unused (WebAudio was the only user of ThreadSafeRefCountedGarbageCollected, but we already shipped Oilpan for WebAudio and removed the need).

BUG=340522
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=183582

[haraken, 2014-10-02 07:48:07]

haraken@chromium.org changed reviewers:
+ ager@chromium.org, oilpan-reviews@chromium.org

[haraken, 2014-10-02 07:48:34]

PTAL

[tkent, 2014-10-02 07:59:35]

tkent@chromium.org changed reviewers:
+ tkent@chromium.org

[tkent, 2014-10-02 07:59:36]

lgtm

https://codereview.chromium.org/614373007/diff/1/Source/platform/heap/Handle.h
File Source/platform/heap/Handle.h (right):

https://codereview.chromium.org/614373007/diff/1/Source/platform/heap/Handle....
Source/platform/heap/Handle.h:1059: template<typename T> RawPtr<T>
adoptRefWillBeNoop(T* ptr)
RawPtr<T> -> T* ?

https://codereview.chromium.org/614373007/diff/1/Source/platform/heap/Handle....
Source/platform/heap/Handle.h:1063: return RawPtr<T>(ptr);
RawPTR<T>(ptr) -> ptr?

https://codereview.chromium.org/614373007/diff/1/Source/platform/heap/Handle....
Source/platform/heap/Handle.h:1066: template<typename T> RawPtr<T>
adoptPtrWillBeNoop(T* ptr)
Ditto.

https://codereview.chromium.org/614373007/diff/1/Source/platform/heap/Handle....
Source/platform/heap/Handle.h:1070: return RawPtr<T>(ptr);
Ditto.

[haraken, 2014-10-02 09:42:08]

> https://codereview.chromium.org/614373007/diff/1/Source/platform/heap/Handle.h
> File Source/platform/heap/Handle.h (right):
> 
>
https://codereview.chromium.org/614373007/diff/1/Source/platform/heap/Handle....
> Source/platform/heap/Handle.h:1059: template<typename T> RawPtr<T>
> adoptRefWillBeNoop(T* ptr)
> RawPtr<T> -> T* ?
> 
>
https://codereview.chromium.org/614373007/diff/1/Source/platform/heap/Handle....
> Source/platform/heap/Handle.h:1063: return RawPtr<T>(ptr);
> RawPTR<T>(ptr) -> ptr?
> 
>
https://codereview.chromium.org/614373007/diff/1/Source/platform/heap/Handle....
> Source/platform/heap/Handle.h:1066: template<typename T> RawPtr<T>
> adoptPtrWillBeNoop(T* ptr)
> Ditto.
> 
>
https://codereview.chromium.org/614373007/diff/1/Source/platform/heap/Handle....
> Source/platform/heap/Handle.h:1070: return RawPtr<T>(ptr);
> Ditto.

Thanks for review! Done.

[haraken, 2014-10-02 09:42:13]

The CQ bit was checked by haraken@chromium.org

[commit-bot: I haz the power, 2014-10-02 09:42:59]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/614373007/20001

[commit-bot: I haz the power, 2014-10-02 10:10:55]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-10-02 10:10:56]

Try jobs failed on following builders:
  linux_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/linux_blink_rel/builds/2...)

[haraken, 2014-10-12 15:00:15]

On 2014/10/02 10:10:56, I haz the power (commit-bot) wrote:
> Try jobs failed on following builders:
>   linux_blink_rel on tryserver.blink
>
(http://build.chromium.org/p/tryserver.blink/builders/linux_blink_rel/builds/2...)

I need to land this...

[haraken, 2014-10-12 15:00:34]

The CQ bit was checked by haraken@chromium.org

[commit-bot: I haz the power, 2014-10-12 15:01:25]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/614373007/330001

[commit-bot: I haz the power, 2014-10-12 15:17:04]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-10-12 15:17:05]

Try jobs failed on following builders:
  linux_chromium_gn_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/linux_chromium_gn_rel/bu...)

[haraken, 2014-10-12 16:18:44]

The CQ bit was checked by haraken@chromium.org

[commit-bot: I haz the power, 2014-10-12 16:19:26]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/614373007/460001

[commit-bot: I haz the power, 2014-10-12 19:19:20]

Committed patchset #5 (id:460001) as 183582

[sof, 2014-10-13 06:34:40]

sigbjornf@opera.com changed reviewers:
+ sigbjornf@opera.com

[sof, 2014-10-13 06:34:41]

This leads to a larger amount of assert failures (layout + unit tests) on the
Oilpan bots,


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

i.e., the SVG properties have a base object (SVGPropertyBase) which is
RefCountedGarbageCollected, but the derived objects are still create()d using
adoptRef() and handled via (Pass)RefPtr.

The use of an initial m_refCount = 0 for RefCountedGarbageCollected here creates
an imbalance.

IOW, more work is needed here. I suggest we revert for now.

[sof, 2014-10-13 06:49:24]

On 2014/10/13 06:34:41, sof wrote:
> This leads to a larger amount of assert failures (layout + unit tests) on the
> Oilpan bots,
> 
> 
>
https://storage.googleapis.com/chromium-layout-test-archives/WebKit_Linux_Oil...
> 
> i.e., the SVG properties have a base object (SVGPropertyBase) which is
> RefCountedGarbageCollected, but the derived objects are still create()d using
> adoptRef() and handled via (Pass)RefPtr.
> 
> The use of an initial m_refCount = 0 for RefCountedGarbageCollected here
creates
> an imbalance.
> 
> IOW, more work is needed here. I suggest we revert for now.

Looks like there is a relatively tidy way to address it; let's try that first.

[sof, 2014-10-13 07:22:19]

On 2014/10/13 06:49:24, sof wrote:
> On 2014/10/13 06:34:41, sof wrote:
> > This leads to a larger amount of assert failures (layout + unit tests) on
the
> > Oilpan bots,
> > 
> > 
> >
>
https://storage.googleapis.com/chromium-layout-test-archives/WebKit_Linux_Oil...
> > 
> > i.e., the SVG properties have a base object (SVGPropertyBase) which is
> > RefCountedGarbageCollected, but the derived objects are still create()d
using
> > adoptRef() and handled via (Pass)RefPtr.
> > 
> > The use of an initial m_refCount = 0 for RefCountedGarbageCollected here
> creates
> > an imbalance.
> > 
> > IOW, more work is needed here. I suggest we revert for now.
> 
> Looks like there is a relatively tidy way to address it; let's try that first.

https://codereview.chromium.org/650033002/