Oilpan: Enable oilpan for callback classes

Oilpan: Enable oilpan for callback classes

BUG=

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=182656

[keishi, 2014-09-18 11:18:35]

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

[sof, 2014-09-18 11:38:54]

https://codereview.chromium.org/563703002/diff/50001/Source/bindings/template...
File Source/bindings/templates/methods.cpp (right):

https://codereview.chromium.org/563703002/diff/50001/Source/bindings/template...
Source/bindings/templates/methods.cpp:92: {{argument.idl_type}}*
{{argument.name}} = nullptr;
nit: ";" at the end appears redundant.

[Jens Widell, 2014-09-18 12:13:18]

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

[Jens Widell, 2014-09-18 12:13:20]

https://codereview.chromium.org/563703002/diff/50001/Source/bindings/template...
File Source/bindings/templates/methods.cpp (right):

https://codereview.chromium.org/563703002/diff/50001/Source/bindings/template...
Source/bindings/templates/methods.cpp:92: {{argument.idl_type}}*
{{argument.name}} = nullptr;
I think we can (and should) remove the special handling for callback interfaces
here, keeping just the special case for EventListener.

The only difference would be that the explicit nullptr initialization goes away,
but it's only necessary because of other special cases for callback interfaces,
where we can add it instead.

https://codereview.chromium.org/563703002/diff/50001/Source/bindings/template...
Source/bindings/templates/methods.cpp:143: if (info.Length() >
{{argument.index}} && !isUndefinedOrNull(info[{{argument.index}}])) {
We could skip the "info.Length() > {{argument.index}} &&" part here. If the
argument is missing, "info[{{argument.index}}]" will be undefined, so the
isUndefinedOrNull() check is enough.

This matches how we handle optional arguments with default values in
generate_arguments().

https://codereview.chromium.org/563703002/diff/50001/Source/bindings/template...
Source/bindings/templates/methods.cpp:151: }
Change this to

  } else {
    {{argument.name}} = nullptr;
  }

to replace the nullptr initialization in generate_argument_var_declaration().

[tkent, 2014-09-19 00:37:33]

Should we do this now?  Making RequestAnimationFrame GCed looks to have
performance risk.

[tkent, 2014-09-19 00:44:31]

On 2014/09/19 00:37:33, tkent (overloaded) wrote:
> Should we do this now?  Making RequestAnimationFrame GCed looks to have
> performance risk.

Ah, no. Each frame of RAF doesn't create a GC object.  No performance risk.

[keishi, 2014-09-19 06:43:35]

https://codereview.chromium.org/563703002/diff/50001/Source/bindings/template...
File Source/bindings/templates/methods.cpp (right):

https://codereview.chromium.org/563703002/diff/50001/Source/bindings/template...
Source/bindings/templates/methods.cpp:92: {{argument.idl_type}}*
{{argument.name}} = nullptr;
On 2014/09/18 12:13:19, Jens Widell wrote:
> I think we can (and should) remove the special handling for callback
interfaces
> here, keeping just the special case for EventListener.
> 
> The only difference would be that the explicit nullptr initialization goes
away,
> but it's only necessary because of other special cases for callback
interfaces,
> where we can add it instead.

Done.

https://codereview.chromium.org/563703002/diff/50001/Source/bindings/template...
Source/bindings/templates/methods.cpp:151: }
On 2014/09/18 12:13:19, Jens Widell wrote:
> Change this to
> 
>   } else {
>     {{argument.name}} = nullptr;
>   }
> 
> to replace the nullptr initialization in generate_argument_var_declaration().

Done.

[keishi, 2014-09-19 06:44:15]

https://codereview.chromium.org/563703002/diff/50001/Source/bindings/template...
File Source/bindings/templates/methods.cpp (right):

https://codereview.chromium.org/563703002/diff/50001/Source/bindings/template...
Source/bindings/templates/methods.cpp:143: if (info.Length() >
{{argument.index}} && !isUndefinedOrNull(info[{{argument.index}}])) {
On 2014/09/18 12:13:19, Jens Widell wrote:
> We could skip the "info.Length() > {{argument.index}} &&" part here. If the
> argument is missing, "info[{{argument.index}}]" will be undefined, so the
> isUndefinedOrNull() check is enough.
> 
> This matches how we handle optional arguments with default values in
> generate_arguments().

Done.

[Mads Ager (chromium), 2014-09-19 08:11:51]

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

[Mads Ager (chromium), 2014-09-19 08:11:52]

LGTM

[Jens Widell, 2014-09-19 09:07:56]

bindings/ LGTM, with one nit.

https://codereview.chromium.org/563703002/diff/70001/Source/bindings/modules/...
File Source/bindings/modules/v8/custom/V8SQLTransactionCustom.cpp (right):

https://codereview.chromium.org/563703002/diff/70001/Source/bindings/modules/...
Source/bindings/modules/v8/custom/V8SQLTransactionCustom.cpp:94:
SQLStatementCallback* callback = nullptr;
Nit: Could you align this custom code to what the code generator would have
generated? E.g., move nullptr assignment and drop info.Length() check.

[keishi, 2014-09-19 10:30:07]

https://codereview.chromium.org/563703002/diff/70001/Source/bindings/modules/...
File Source/bindings/modules/v8/custom/V8SQLTransactionCustom.cpp (right):

https://codereview.chromium.org/563703002/diff/70001/Source/bindings/modules/...
Source/bindings/modules/v8/custom/V8SQLTransactionCustom.cpp:94:
SQLStatementCallback* callback = nullptr;
On 2014/09/19 09:07:55, Jens Widell wrote:
> Nit: Could you align this custom code to what the code generator would have
> generated? E.g., move nullptr assignment and drop info.Length() check.

Done.

[keishi, 2014-09-19 10:30:22]

Thanks.

[keishi, 2014-09-19 10:30:30]

The CQ bit was checked by keishi@chromium.org

[commit-bot: I haz the power, 2014-09-19 10:30:51]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patchset/563703002/90001

[Jens Widell, 2014-09-19 10:34:06]

https://codereview.chromium.org/563703002/diff/90001/Source/bindings/modules/...
File Source/bindings/modules/v8/custom/V8SQLTransactionCustom.cpp (right):

https://codereview.chromium.org/563703002/diff/90001/Source/bindings/modules/...
Source/bindings/modules/v8/custom/V8SQLTransactionCustom.cpp:106:
SQLStatementErrorCallback* errorCallback = nullptr;
Follow-up nit: Align this block too?

[commit-bot: I haz the power, 2014-09-19 10:42:59]

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

[commit-bot: I haz the power, 2014-09-19 10:43:02]

Try jobs failed on following builders:
  blink_presubmit on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/blink_presubmit/builds/1...)

[haraken, 2014-09-20 09:00:47]

LGTM

[tkent, 2014-09-22 02:04:06]

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
File Source/modules/filesystem/DirectoryReader.cpp (right):

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/DirectoryReader.cpp:60: // FIXME: This Persistent
keeps the reader alive until all of the readDirectory results are received.
crbug.com/350285
Will this FIXME be resolved?

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/DirectoryReader.cpp:141: ErrorCallback* errorCallback
= m_errorCallback;
Are you sure we don't need .release()?

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
File Source/modules/filesystem/FileSystemCallbacks.cpp (right):

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:82:
handleEventOrScheduleCallback(m_errorCallback.get(),
FileError::create(static_cast<FileError::ErrorCode>(code)));
Ditto.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:156:
handleEventOrScheduleCallback(m_successCallback.get(),
DirectoryEntry::create(m_fileSystem, m_expectedPath));
Ditto.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:158:
handleEventOrScheduleCallback(m_successCallback.get(),
FileEntry::create(m_fileSystem, m_expectedPath));
Ditto.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:217:
handleEventOrScheduleCallback(m_successCallback.get(),
DOMFileSystem::create(m_executionContext.get(), name, m_type, rootURL));
Ditto.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:240:
handleEventOrScheduleCallback(m_errorCallback.get(),
FileError::create(FileError::INVALID_MODIFICATION_ERR));
Ditto.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:245:
handleEventOrScheduleCallback(m_successCallback.get(),
DirectoryEntry::create(filesystem, absolutePath));
Ditto.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:247:
handleEventOrScheduleCallback(m_successCallback.get(),
FileEntry::create(filesystem, absolutePath));
Ditto.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:266:
handleEventOrScheduleCallback(m_successCallback.get(),
Metadata::create(metadata));
Ditto.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:287:
handleEventOrScheduleCallback(m_successCallback.get(), m_fileWriter.release());
Ditto.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:315:
handleEventOrScheduleCallback(m_successCallback.get(),
DOMFileSystemBase::createFile(metadata, m_url, m_fileSystem->type(), m_name));
Ditto.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:334:
handleEventOrScheduleCallback(m_successCallback.get());
Ditto.

[keishi, 2014-09-22 05:44:02]

https://codereview.chromium.org/563703002/diff/90001/Source/bindings/modules/...
File Source/bindings/modules/v8/custom/V8SQLTransactionCustom.cpp (right):

https://codereview.chromium.org/563703002/diff/90001/Source/bindings/modules/...
Source/bindings/modules/v8/custom/V8SQLTransactionCustom.cpp:106:
SQLStatementErrorCallback* errorCallback = nullptr;
On 2014/09/19 10:34:06, Jens Widell wrote:
> Follow-up nit: Align this block too?

Done.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
File Source/modules/filesystem/DirectoryReader.cpp (right):

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/DirectoryReader.cpp:60: // FIXME: This Persistent
keeps the reader alive until all of the readDirectory results are received.
crbug.com/350285
I don't think it is resolved, we still need this to keep the reader alive.
Changed comment from Persistent to Member.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/DirectoryReader.cpp:141: ErrorCallback* errorCallback
= m_errorCallback;
On 2014/09/22 02:04:05, tkent (high review load) wrote:
> Are you sure we don't need .release()?
> 

Added back.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
File Source/modules/filesystem/FileSystemCallbacks.cpp (right):

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:82:
handleEventOrScheduleCallback(m_errorCallback.get(),
FileError::create(static_cast<FileError::ErrorCode>(code)));
On 2014/09/22 02:04:06, tkent (high review load) wrote:
> Ditto.

Done.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:156:
handleEventOrScheduleCallback(m_successCallback.get(),
DirectoryEntry::create(m_fileSystem, m_expectedPath));
On 2014/09/22 02:04:06, tkent (high review load) wrote:
> Ditto.

Done.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:158:
handleEventOrScheduleCallback(m_successCallback.get(),
FileEntry::create(m_fileSystem, m_expectedPath));
On 2014/09/22 02:04:05, tkent (high review load) wrote:
> Ditto.

Done.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:217:
handleEventOrScheduleCallback(m_successCallback.get(),
DOMFileSystem::create(m_executionContext.get(), name, m_type, rootURL));
On 2014/09/22 02:04:06, tkent (high review load) wrote:
> Ditto.

Done.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:240:
handleEventOrScheduleCallback(m_errorCallback.get(),
FileError::create(FileError::INVALID_MODIFICATION_ERR));
On 2014/09/22 02:04:06, tkent (high review load) wrote:
> Ditto.

Done.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:245:
handleEventOrScheduleCallback(m_successCallback.get(),
DirectoryEntry::create(filesystem, absolutePath));
On 2014/09/22 02:04:06, tkent (high review load) wrote:
> Ditto.

Done.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:247:
handleEventOrScheduleCallback(m_successCallback.get(),
FileEntry::create(filesystem, absolutePath));
On 2014/09/22 02:04:06, tkent (high review load) wrote:
> Ditto.

Done.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:266:
handleEventOrScheduleCallback(m_successCallback.get(),
Metadata::create(metadata));
On 2014/09/22 02:04:06, tkent (high review load) wrote:
> Ditto.

Done.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:287:
handleEventOrScheduleCallback(m_successCallback.get(), m_fileWriter.release());
On 2014/09/22 02:04:05, tkent (high review load) wrote:
> Ditto.

Done.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:315:
handleEventOrScheduleCallback(m_successCallback.get(),
DOMFileSystemBase::createFile(metadata, m_url, m_fileSystem->type(), m_name));
On 2014/09/22 02:04:06, tkent (high review load) wrote:
> Ditto.

Done.

https://codereview.chromium.org/563703002/diff/90001/Source/modules/filesyste...
Source/modules/filesystem/FileSystemCallbacks.cpp:334:
handleEventOrScheduleCallback(m_successCallback.get());
On 2014/09/22 02:04:06, tkent (high review load) wrote:
> Ditto.

Done.

[keishi, 2014-09-22 05:44:15]

The CQ bit was checked by keishi@chromium.org

[keishi, 2014-09-22 05:44:23]

The CQ bit was unchecked by keishi@chromium.org

[tkent, 2014-09-22 07:29:47]

lgtm

[keishi, 2014-09-22 07:39:28]

The CQ bit was checked by keishi@chromium.org

[commit-bot: I haz the power, 2014-09-22 07:39:50]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/563703002/110001

[commit-bot: I haz the power, 2014-09-22 07:40:44]

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

[commit-bot: I haz the power, 2014-09-22 07:40:46]

Failed to apply patch for Source/modules/webdatabase/Database.cpp:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  patching file Source/modules/webdatabase/Database.cpp
  Hunk #1 FAILED at 112.
  1 out of 1 hunk FAILED -- saving rejects to file
Source/modules/webdatabase/Database.cpp.rej

Patch:       Source/modules/webdatabase/Database.cpp
Index: Source/modules/webdatabase/Database.cpp
diff --git a/Source/modules/webdatabase/Database.cpp
b/Source/modules/webdatabase/Database.cpp
index
7cb734e3d12960b2449868802a5de7159b356622..780d1cd16bc2687f0d2a919d1c9d600cba04094d
100644
--- a/Source/modules/webdatabase/Database.cpp
+++ b/Source/modules/webdatabase/Database.cpp
@@ -112,45 +112,45 @@ void Database::closeImmediately()
     }
 }
 
-void Database::changeVersion(const String& oldVersion, const String&
newVersion, PassOwnPtrWillBeRawPtr<SQLTransactionCallback> callback,
PassOwnPtrWillBeRawPtr<SQLTransactionErrorCallback> errorCallback,
PassOwnPtrWillBeRawPtr<VoidCallback> successCallback)
+void Database::changeVersion(const String& oldVersion, const String&
newVersion, SQLTransactionCallback* callback, SQLTransactionErrorCallback*
errorCallback, VoidCallback* successCallback)
 {
     ChangeVersionData data(oldVersion, newVersion);
     runTransaction(callback, errorCallback, successCallback, false, &data);
 }
 
-void Database::transaction(PassOwnPtrWillBeRawPtr<SQLTransactionCallback>
callback, PassOwnPtrWillBeRawPtr<SQLTransactionErrorCallback> errorCallback,
PassOwnPtrWillBeRawPtr<VoidCallback> successCallback)
+void Database::transaction(SQLTransactionCallback* callback,
SQLTransactionErrorCallback* errorCallback, VoidCallback* successCallback)
 {
     runTransaction(callback, errorCallback, successCallback, false);
 }
 
-void Database::readTransaction(PassOwnPtrWillBeRawPtr<SQLTransactionCallback>
callback, PassOwnPtrWillBeRawPtr<SQLTransactionErrorCallback> errorCallback,
PassOwnPtrWillBeRawPtr<VoidCallback> successCallback)
+void Database::readTransaction(SQLTransactionCallback* callback,
SQLTransactionErrorCallback* errorCallback, VoidCallback* successCallback)
 {
     runTransaction(callback, errorCallback, successCallback, true);
 }
 
-static void callTransactionErrorCallback(ExecutionContext*,
PassOwnPtrWillBeRawPtr<SQLTransactionErrorCallback> callback,
PassOwnPtr<SQLErrorData> errorData)
+static void callTransactionErrorCallback(ExecutionContext*,
SQLTransactionErrorCallback* callback, PassOwnPtr<SQLErrorData> errorData)
 {
     RefPtrWillBeRawPtr<SQLError> error = SQLError::create(*errorData);
     callback->handleEvent(error.get());
 }
 
-void Database::runTransaction(PassOwnPtrWillBeRawPtr<SQLTransactionCallback>
callback, PassOwnPtrWillBeRawPtr<SQLTransactionErrorCallback> errorCallback,
-    PassOwnPtrWillBeRawPtr<VoidCallback> successCallback, bool readOnly, const
ChangeVersionData* changeVersionData)
+void Database::runTransaction(SQLTransactionCallback* callback,
SQLTransactionErrorCallback* errorCallback,
+    VoidCallback* successCallback, bool readOnly, const ChangeVersionData*
changeVersionData)
 {
     // FIXME: Rather than passing errorCallback to SQLTransaction and then
sometimes firing it ourselves,
     // this code should probably be pushed down into DatabaseBackend so that we
only create the SQLTransaction
     // if we're actually going to run it.
 #if ENABLE(ASSERT)
-    SQLTransactionErrorCallback* originalErrorCallback = errorCallback.get();
+    SQLTransactionErrorCallback* originalErrorCallback = errorCallback;
 #endif
     RefPtrWillBeRawPtr<SQLTransaction> transaction =
SQLTransaction::create(this, callback, successCallback, errorCallback,
readOnly);
     RefPtrWillBeRawPtr<SQLTransactionBackend> transactionBackend =
backend()->runTransaction(transaction, readOnly, changeVersionData);
     if (!transactionBackend) {
-        OwnPtrWillBeRawPtr<SQLTransactionErrorCallback> callback =
transaction->releaseErrorCallback();
+        SQLTransactionErrorCallback* callback =
transaction->releaseErrorCallback();
         ASSERT(callback == originalErrorCallback);
         if (callback) {
             OwnPtr<SQLErrorData> error =
SQLErrorData::create(SQLError::UNKNOWN_ERR, "database has been closed");
-           
executionContext()->postTask(createCrossThreadTask(&callTransactionErrorCallback,
callback.release(), error.release()));
+           
executionContext()->postTask(createCrossThreadTask(&callTransactionErrorCallback,
callback, error.release()));
         }
     }
 }

[keishi, 2014-09-24 05:56:10]

The CQ bit was checked by keishi@chromium.org

[commit-bot: I haz the power, 2014-09-24 05:57:10]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/563703002/130001

[commit-bot: I haz the power, 2014-09-24 06:15:46]

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

[commit-bot: I haz the power, 2014-09-24 06:15:49]

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...)

[keishi, 2014-09-25 03:51:19]

The CQ bit was checked by keishi@chromium.org

[commit-bot: I haz the power, 2014-09-25 03:52:07]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/563703002/150001

[commit-bot: I haz the power, 2014-09-25 05:04:30]

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

[commit-bot: I haz the power, 2014-09-25 05:04:33]

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...)

[tkent, 2014-09-25 06:49:51]

https://codereview.chromium.org/563703002/diff/150001/Source/core/html/HTMLSe...
File Source/core/html/HTMLSelectElement.cpp (right):

https://codereview.chromium.org/563703002/diff/150001/Source/core/html/HTMLSe...
Source/core/html/HTMLSelectElement.cpp:384: return
HTMLFormControlElementWithState::createRenderer(style);
Looks an accidental change.

[keishi, 2014-09-25 07:02:56]

https://codereview.chromium.org/563703002/diff/150001/Source/core/html/HTMLSe...
File Source/core/html/HTMLSelectElement.cpp (right):

https://codereview.chromium.org/563703002/diff/150001/Source/core/html/HTMLSe...
Source/core/html/HTMLSelectElement.cpp:384: return
HTMLFormControlElementWithState::createRenderer(style);
On 2014/09/25 06:49:51, tkent (high review load) wrote:
> Looks an accidental change.

Sorry, thanks.

[keishi, 2014-09-25 07:03:10]

The CQ bit was checked by keishi@chromium.org

[commit-bot: I haz the power, 2014-09-25 07:04:06]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/563703002/170001

[commit-bot: I haz the power, 2014-09-25 08:15:59]

Committed patchset #10 (id:170001) as 182656

[leviw_travelin_and_unemployed, 2014-10-14 19:35:14]

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

[leviw_travelin_and_unemployed, 2014-10-14 19:35:15]

We believe this patch may have caused memory leaks:
https://code.google.com/p/chromium/issues/detail?id=423047

I lack a lot of context on this change, but I do see a bunch of OwnPtrs being
turned into raw pointers, and I'm unsure of their lifetime...