HTML Imports: Fix race conditions on HTMLImportsChild

HTML Imports: Fix race conditions on HTMLImportsChild

There are two race conditions in HTMLImportChild.

1. ensureLoader() is called in stateDidChange(), which is triggered via a timer.
   That means there is a timing in which m_customElementMicrotaskStep isn't created but
   accompanying <link rel=import> is in tree. This allows some microtask can slip away
   from the blocking import.

2. HTMLImport::isDone() can return true when m_customElementMicrotaskStep is YET to be
   created, that can happen the call on ensureLoader().

This change covers these two cases. It also fixes a trivial flakiness in
import-custom-element-dup-resolve.html

TEST=import-custom-element-dup-resolve.html
R=dglazkov@chromium.org, dominicc@chromium.org
BUG=371654
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=173794

[Hajime Morrita, 2014-05-09 21:30:33]

Dimitri, could you rubberstamp this? I want to have this
before branch :-/

[dglazkov, 2014-05-09 21:51:26]

rslgtm.

[Hajime Morrita, 2014-05-09 21:59:43]

The CQ bit was checked by morrita@chromium.org

[commit-bot: I haz the power, 2014-05-09 22:00:37]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/morrita@chromium.org/272243004/1

[commit-bot: I haz the power, 2014-05-09 22:48:01]

FYI, CQ is re-trying this CL (attempt #1).
Please consider checking whether the failures are real,
and report flakes to chrome-troopers@google.com.
The failing builders are:
  linux_blink_dbg on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/linux_blink_dbg/builds/7127)

[commit-bot: I haz the power, 2014-05-10 00:14:00]

Change committed as 173794

[dominicc (has gone to gerrit), 2014-05-12 00:42:18]

https://codereview.chromium.org/272243004/diff/1/Source/core/dom/custom/Custo...
File Source/core/dom/custom/CustomElementMicrotaskQueue.cpp (right):

https://codereview.chromium.org/272243004/diff/1/Source/core/dom/custom/Custo...
Source/core/dom/custom/CustomElementMicrotaskQueue.cpp:106: bool
CustomElementMicrotaskQueue::needsProcessOrStop() const
I think the complexity of Imports has degraded this a bit... I think you should
try moving this back to imports (it resembles a lot of Imports code and is only
triggered by imports) and slay the complexity dragon there.

https://codereview.chromium.org/272243004/diff/1/Source/core/dom/custom/Custo...
File Source/core/dom/custom/CustomElementMicrotaskQueue.h (right):

https://codereview.chromium.org/272243004/diff/1/Source/core/dom/custom/Custo...
Source/core/dom/custom/CustomElementMicrotaskQueue.h:55: bool
needsProcessOrStop() const;
What is this? Why do imports leak into this abstraction?

[Hajime Morrita, 2014-05-14 00:24:55]

On 2014/05/12 00:42:18, dominicc wrote:
>
https://codereview.chromium.org/272243004/diff/1/Source/core/dom/custom/Custo...
> File Source/core/dom/custom/CustomElementMicrotaskQueue.cpp (right):
> 
>
https://codereview.chromium.org/272243004/diff/1/Source/core/dom/custom/Custo...
> Source/core/dom/custom/CustomElementMicrotaskQueue.cpp:106: bool
> CustomElementMicrotaskQueue::needsProcessOrStop() const
> I think the complexity of Imports has degraded this a bit... I think you
should
> try moving this back to imports (it resembles a lot of Imports code and is
only
> triggered by imports) and slay the complexity dragon there.
> 
>
https://codereview.chromium.org/272243004/diff/1/Source/core/dom/custom/Custo...
> File Source/core/dom/custom/CustomElementMicrotaskQueue.h (right):
> 
>
https://codereview.chromium.org/272243004/diff/1/Source/core/dom/custom/Custo...
> Source/core/dom/custom/CustomElementMicrotaskQueue.h:55: bool
> needsProcessOrStop() const;
> What is this? Why do imports leak into this abstraction?

This is fair point. I tried to keep the responsibility inside imports side but
failed is this first cut.
Let me think more.