[screen-orientation] Update implementation to match recent spec changes.

[screen-orientation] Update implementation to match recent spec changes.

https://github.com/w3c/screen-orientation/pull/21

BUG=162827
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=177734

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=177933

[abarth-chromium, 2014-07-07 17:29:33]

Several parts of this CL look like they have tricky memory issues.  Would you be
willing to break down this change into smaller pieces that are easier to review?
 We need to be careful not to introduce memory leaks or use-after-frees.

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
File Source/modules/screen_orientation/PartialScreenOrientation.h (right):

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
Source/modules/screen_orientation/PartialScreenOrientation.h:17: class
PartialScreenOrientation FINAL : public
NoBaseWillBeGarbageCollectedFinalized<PartialScreenOrientation>, public
WillBeHeapSupplement<Screen>, DOMWindowProperty {
PartialScreenOrientation -> ScreenScreenOrientation ?  I guess that reads kind
of odd, but that's the pattern we use to name these supplements...

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
File Source/modules/screen_orientation/ScreenOrientation.cpp (right):

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
Source/modules/screen_orientation/ScreenOrientation.cpp:171: }
You don't need this code because this object is kept alive by virtue of being
stored at window.screen.orientation.

Worse, this line of code will cause a memory leak.  If script adds an event
listener to this object, it will have pending activity forever and never be
garbage collected.

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
File Source/modules/screen_orientation/ScreenOrientation.h (right):

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
Source/modules/screen_orientation/ScreenOrientation.h:26: public ActiveDOMObject
{
You shouldn't need to be an ActiveDOMObject

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
File Source/modules/screen_orientation/ScreenOrientationController.h (right):

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
Source/modules/screen_orientation/ScreenOrientationController.h:55:
PersistentWillBeMember<ScreenOrientation> m_orientation;
This pointer is cleared in ~ScreenOrientation, which will never execute because
this point keeps m_orientation alive....

[mlamouri (slow - plz ping), 2014-07-08 14:07:59]

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
File Source/modules/screen_orientation/PartialScreenOrientation.h (right):

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
Source/modules/screen_orientation/PartialScreenOrientation.h:17: class
PartialScreenOrientation FINAL : public
NoBaseWillBeGarbageCollectedFinalized<PartialScreenOrientation>, public
WillBeHeapSupplement<Screen>, DOMWindowProperty {
On 2014/07/07 17:29:33, abarth wrote:
> PartialScreenOrientation -> ScreenScreenOrientation ?  I guess that reads kind
> of odd, but that's the pattern we use to name these supplements...

Done.

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
File Source/modules/screen_orientation/ScreenOrientation.cpp (right):

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
Source/modules/screen_orientation/ScreenOrientation.cpp:171: }
On 2014/07/07 17:29:33, abarth wrote:
> You don't need this code because this object is kept alive by virtue of being
> stored at window.screen.orientation.
> 
> Worse, this line of code will cause a memory leak.  If script adds an event
> listener to this object, it will have pending activity forever and never be
> garbage collected.

Done.

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
File Source/modules/screen_orientation/ScreenOrientation.h (right):

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
Source/modules/screen_orientation/ScreenOrientation.h:26: public ActiveDOMObject
{
On 2014/07/07 17:29:33, abarth wrote:
> You shouldn't need to be an ActiveDOMObject

So, I tried to find out when ActiveDOMObject was needed compared to
ContextLifecycleObserver and even what that later was doing. Sadly, this code
seems undocumented and miss general comments so I ended up looking at similar
code.

I wonder now why NetworkInformation uses ActiveDOMObject. It seems that
navigator.connection has a general behaviour similar to screen.orientation,
doesn't it?

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
File Source/modules/screen_orientation/ScreenOrientationController.h (right):

https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
Source/modules/screen_orientation/ScreenOrientationController.h:55:
PersistentWillBeMember<ScreenOrientation> m_orientation;
On 2014/07/07 17:29:33, abarth wrote:
> This pointer is cleared in ~ScreenOrientation, which will never execute
because
> this point keeps m_orientation alive....

I should have looked into that a bit more before uploading. I wanted to have a
raw pointer but blink-gc (clang plugin) forbids me to. I understand the general
reasons but here I know what I'm doing.

I came with another design where ScreenScreenOrientation is responsible to link
the controller and the ScreenOrientation object. It creates the link and
destroys it. I guess that should work but in practice, I still don't get a
ScreenOrientation dtor being called. I'm going to update the CL and have another
look.

[abarth-chromium, 2014-07-08 14:43:25]

On 2014/07/08 at 14:07:59, mlamouri wrote:
>
https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
> File Source/modules/screen_orientation/ScreenOrientation.h (right):
> 
>
https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
> Source/modules/screen_orientation/ScreenOrientation.h:26: public
ActiveDOMObject {
> On 2014/07/07 17:29:33, abarth wrote:
> > You shouldn't need to be an ActiveDOMObject
> 
> So, I tried to find out when ActiveDOMObject was needed compared to
ContextLifecycleObserver and even what that later was doing. Sadly, this code
seems undocumented and miss general comments so I ended up looking at similar
code.

ActiveDOMObject gives you a superset of the notifications that
ContextLifecycleObserver gives you.  If you need a weak pointer to the context,
you probably just want ContextLifecycleObserver.  Features that need
ActiveDOMObject are those that need to manually control the lifetime of their
JavaScript wrapper or that need a notification when the document is stopping. 
Those two needs are typically paired because if you're manually controlling the
lifetime of your JavaScript wrapper, you probably need to stop keeping the
wrapper alive when the document stops.

> I wonder now why NetworkInformation uses ActiveDOMObject. It seems that
navigator.connection has a general behaviour similar to screen.orientation,
doesn't it?

NetworkInformation implements ActiveDOMObject in order to get the stop()
notification.  When the document stops, NetworkInformation stops observing the
state of network information in the underlying platform. 

void NetworkInformation::stop()
{
    m_contextStopped = true;
    stopObserving();
}

Changes flow "up" from the platform layer and trigger events in script. 
ScreenOrientation works the opposite way.  Changes flow "down" from the web
layer.  Either design works fine, but the implementation is a bit different. 
Nowadays we tend to prefer the approach taken by NetworkInformation.  If you
like, we can refactor ScreenOrientation to work that way too.  We just haven't
gone through and updated all the existing APIs to the modern approach.

>
https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
> File Source/modules/screen_orientation/ScreenOrientationController.h (right):
> 
>
https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
> Source/modules/screen_orientation/ScreenOrientationController.h:55:
PersistentWillBeMember<ScreenOrientation> m_orientation;
> On 2014/07/07 17:29:33, abarth wrote:
> > This pointer is cleared in ~ScreenOrientation, which will never execute
because
> > this point keeps m_orientation alive....
> 
> I should have looked into that a bit more before uploading. I wanted to have a
raw pointer but blink-gc (clang plugin) forbids me to. I understand the general
reasons but here I know what I'm doing.
> 
> I came with another design where ScreenScreenOrientation is responsible to
link the controller and the ScreenOrientation object. It creates the link and
destroys it. I guess that should work but in practice, I still don't get a
ScreenOrientation dtor being called. I'm going to update the CL and have another
look.

Ok.  Yeah, these lifetime issues are why we tend to use the "bottom up" design
for new features.  Then you don't have this strange coupling between the
WebFrame client objects and the DOM objects.  Instead, the DOM objects just use
resources in blink::Platform that they own.  If you like, you can take a look at
the subclasses of DeviceEventDispatcherBase.

The current design should still be workable.  It's just going to be slightly
more delicate to get right.

[mlamouri (slow - plz ping), 2014-07-08 15:10:15]

On 2014/07/08 14:43:25, abarth wrote:
> > I wonder now why NetworkInformation uses ActiveDOMObject. It seems that
> navigator.connection has a general behaviour similar to screen.orientation,
> doesn't it?
> 
> NetworkInformation implements ActiveDOMObject in order to get the stop()
> notification.  When the document stops, NetworkInformation stops observing the
> state of network information in the underlying platform. 
> 
> void NetworkInformation::stop()
> {
>     m_contextStopped = true;
>     stopObserving();
> }
> 
> Changes flow "up" from the platform layer and trigger events in script. 
> ScreenOrientation works the opposite way.  Changes flow "down" from the web
> layer.  Either design works fine, but the implementation is a bit different. 
> Nowadays we tend to prefer the approach taken by NetworkInformation.  If you
> like, we can refactor ScreenOrientation to work that way too.  We just haven't
> gone through and updated all the existing APIs to the modern approach.

Could you elaborate? I was assuming that NetInfo wanted to stop observing when
it wasn't needed. This is a problem we do not have with Screen Orientation
because there is no explicit observer code running when the API is used. Does
the difference between the API go further? (also, contrary to NetInfo and
DeviceMotion, Screen Orientation is related to the View, not the Platform).

> >
>
https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
> > File Source/modules/screen_orientation/ScreenOrientationController.h
(right):
> > 
> >
>
https://codereview.chromium.org/374623002/diff/80001/Source/modules/screen_or...
> > Source/modules/screen_orientation/ScreenOrientationController.h:55:
> PersistentWillBeMember<ScreenOrientation> m_orientation;
> > On 2014/07/07 17:29:33, abarth wrote:
> > > This pointer is cleared in ~ScreenOrientation, which will never execute
> because
> > > this point keeps m_orientation alive....
> > 
> > I should have looked into that a bit more before uploading. I wanted to have
a
> raw pointer but blink-gc (clang plugin) forbids me to. I understand the
general
> reasons but here I know what I'm doing.
> > 
> > I came with another design where ScreenScreenOrientation is responsible to
> link the controller and the ScreenOrientation object. It creates the link and
> destroys it. I guess that should work but in practice, I still don't get a
> ScreenOrientation dtor being called. I'm going to update the CL and have
another
> look.
> 
> Ok.  Yeah, these lifetime issues are why we tend to use the "bottom up" design
> for new features.  Then you don't have this strange coupling between the
> WebFrame client objects and the DOM objects.  Instead, the DOM objects just
use
> resources in blink::Platform that they own.  If you like, you can take a look
at
> the subclasses of DeviceEventDispatcherBase.

Screen Orientation implementation is no longer using blink::Platform on purpose.

> The current design should still be workable.  It's just going to be slightly
> more delicate to get right.

I'm not sure what's the difficulties I should expect here except understanding
why the ScreenOrientation object seems to be kept alive. Is that a direct
consequence from that design choice?

[mlamouri (slow - plz ping), 2014-07-08 15:42:09]

So, it seems that when ScreenScreenOrientation is being destroyed, m_orientation
has no more reference. It's not clear why ScreenOrientation isn't being
destroped but I wonder if that could be for a Oilpan-specific reason I'm
missing.

ager@, could you have a look?

[mlamouri (slow - plz ping), 2014-07-08 16:41:24]

I've updated the patch to make ScreenScreenOrientation no longer a
DOMWindowProperty but make ScreenORientation one instead. Also,
ScreenOrientaiton no longer expect ScreenOrientationController to be alive. I've
changed the ownership to have ScreenScreenOrientation and
ScreenOrientationController to keep ScreenOrientation alive. If there is a new
ScreenOrientation created while ScreenOrientationController is alive, it will
kill update its internal state. That means that we might have
ScreenOrientationController keeping a ScreenOrientation alive but until
ScreenOrientationController dies. IOW, there shouldn't be a cycle here.

PTAL

[abarth-chromium, 2014-07-09 06:04:27]

LGTM!

[abarth-chromium, 2014-07-09 06:04:33]

The CQ bit was checked by abarth@chromium.org

[commit-bot: I haz the power, 2014-07-09 06:05:24]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/mlamouri@chromium.org/374623002/140001

[sof, 2014-07-09 07:02:53]

oilpan lgtm.

https://codereview.chromium.org/374623002/diff/140001/Source/modules/screen_o...
File Source/modules/screen_orientation/ScreenOrientationController.cpp (right):

https://codereview.chromium.org/374623002/diff/140001/Source/modules/screen_o...
Source/modules/screen_orientation/ScreenOrientationController.cpp:135: if
(!ScreenOrientationController::from(*childFrames[i]))
Would be tidier as

    ScreenOrientationController *controller =
ScreenOrientationController::from(*childFrames[i]);
    if (controller)
        controller->notifyOrientationChanged();

or "if (ScreenOrientationController* controller ..)"

https://codereview.chromium.org/374623002/diff/140001/Source/modules/screen_o...
File Source/modules/screen_orientation/ScreenScreenOrientation.cpp (right):

https://codereview.chromium.org/374623002/diff/140001/Source/modules/screen_o...
Source/modules/screen_orientation/ScreenScreenOrientation.cpp:33: if
(!screen.frame()) {
Superfluous braces.

https://codereview.chromium.org/374623002/diff/140001/Source/modules/screen_o...
Source/modules/screen_orientation/ScreenScreenOrientation.cpp:37: if
(!self.m_orientation) {
Superfluous braces.

https://codereview.chromium.org/374623002/diff/140001/Source/web/WebLocalFram...
File Source/web/WebLocalFrameImpl.cpp (right):

https://codereview.chromium.org/374623002/diff/140001/Source/web/WebLocalFram...
Source/web/WebLocalFrameImpl.cpp:1381: if
(ScreenOrientationController::from(*frame()))
Would be preferable (to my eyes) to create a binding for
ScreenOrientationController and use it.

[Mads Ager (chromium), 2014-07-09 07:09:03]

LGTM2

[commit-bot: I haz the power, 2014-07-09 07:13:32]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  linux_blink_dbg on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/linux_blink_dbg/builds/1...)
  linux_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/linux_blink_rel/builds/1...)
  mac_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/mac_blink_rel/builds/14696)
  win_blink_rel on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/win_blink_rel/builds/15992)

[commit-bot: I haz the power, 2014-07-09 07:39:37]

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

[commit-bot: I haz the power, 2014-07-09 07:39:38]

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

[mlamouri (slow - plz ping), 2014-07-09 09:17:30]

https://codereview.chromium.org/374623002/diff/140001/Source/modules/screen_o...
File Source/modules/screen_orientation/ScreenOrientationController.cpp (right):

https://codereview.chromium.org/374623002/diff/140001/Source/modules/screen_o...
Source/modules/screen_orientation/ScreenOrientationController.cpp:135: if
(!ScreenOrientationController::from(*childFrames[i]))
On 2014/07/09 07:02:53, sof wrote:
> Would be tidier as
> 
>     ScreenOrientationController *controller =
> ScreenOrientationController::from(*childFrames[i]);
>     if (controller)
>         controller->notifyOrientationChanged();
> 
> or "if (ScreenOrientationController* controller ..)"

Done.

https://codereview.chromium.org/374623002/diff/140001/Source/modules/screen_o...
File Source/modules/screen_orientation/ScreenScreenOrientation.cpp (right):

https://codereview.chromium.org/374623002/diff/140001/Source/modules/screen_o...
Source/modules/screen_orientation/ScreenScreenOrientation.cpp:33: if
(!screen.frame()) {
On 2014/07/09 07:02:53, sof wrote:
> Superfluous braces.

Done.

https://codereview.chromium.org/374623002/diff/140001/Source/modules/screen_o...
Source/modules/screen_orientation/ScreenScreenOrientation.cpp:37: if
(!self.m_orientation) {
On 2014/07/09 07:02:53, sof wrote:
> Superfluous braces.

Done.

[mlamouri (slow - plz ping), 2014-07-09 09:17:43]

The CQ bit was checked by mlamouri@chromium.org

[commit-bot: I haz the power, 2014-07-09 09:18:33]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/mlamouri@chromium.org/374623002/160001

[commit-bot: I haz the power, 2014-07-09 12:52:35]

Change committed as 177734

[Justin Novosad, 2014-07-09 18:46:01]

On 2014/07/09 12:52:35, I haz the power (commit-bot) wrote:
> Change committed as 177734

New failure:
http://build.chromium.org/p/chromium.webkit/builders/Android%20Tests%20%28dbg...

[ RUN      ] RenderWidgetTest.OnResize
ASSERTION FAILED: RuntimeEnabledFeatures::screenOrientationEnabled()

Appeared in range 177731-177738

[Justin Novosad, 2014-07-09 18:49:31]

On 2014/07/09 18:46:01, junov wrote:
> On 2014/07/09 12:52:35, I haz the power (commit-bot) wrote:
> > Change committed as 177734
> 
> New failure:
>
http://build.chromium.org/p/chromium.webkit/builders/Android%20Tests%20%28dbg...
> 
> [ RUN      ] RenderWidgetTest.OnResize
> ASSERTION FAILED: RuntimeEnabledFeatures::screenOrientationEnabled()
> 
> Appeared in range 177731-177738

Author is not online. Reverting...

[Justin Novosad, 2014-07-09 18:51:14]

A revert of this CL has been created in
https://codereview.chromium.org/381753002/ by junov@chromium.org.

The reason for reverting is: Causing content_browsertests failure on Android:
http://build.chromium.org/p/chromium.webkit/builders/Android%20Tests%20%28dbg....

[sof, 2014-07-09 19:33:38]

It'd be good if you could address this as part of relanding the CL.

https://codereview.chromium.org/374623002/diff/160001/Source/modules/screen_o...
File Source/modules/screen_orientation/ScreenScreenOrientation.cpp (right):

https://codereview.chromium.org/374623002/diff/160001/Source/modules/screen_o...
Source/modules/screen_orientation/ScreenScreenOrientation.cpp:49:
m_orientation->trace(visitor);
Oh, this needs to be visitor->trace(m_orientation) as you want to mark the
ScreenOrientation via the visitor before later on tracing the actual object (if
not already marked.)

[sof, 2014-07-10 05:06:21]

https://codereview.chromium.org/374623002/diff/160001/Source/modules/screen_o...
File Source/modules/screen_orientation/ScreenScreenOrientation.cpp (right):

https://codereview.chromium.org/374623002/diff/160001/Source/modules/screen_o...
Source/modules/screen_orientation/ScreenScreenOrientation.cpp:49:
m_orientation->trace(visitor);
On 2014/07/09 19:33:37, sof wrote:
> Oh, this needs to be visitor->trace(m_orientation) as you want to mark the
> ScreenOrientation via the visitor before later on tracing the actual object
(if
> not already marked.)

It might be worth(re?)considering if the validation of trace() bodies could only
recognize trace() method calls of the above form. There might be some exceptions
(Persistents?) that don't fit, but I suspect not many.

[Mads Ager (chromium), 2014-07-10 06:07:11]

Yes, we should update the plugin to be more precise in the checks that it
performs. As both of us just proved it is easy to miss something like this. :(

[zerny-chromium, 2014-07-10 09:21:13]

On 2014/07/10 06:07:11, Mads Ager (chromium) wrote:
> Yes, we should update the plugin to be more precise in the checks that it
> performs. As both of us just proved it is easy to miss something like this. :(

Yes, this error can result in a dangling pointer. I'll start looking at a plugin
update for that now.

[mlamouri (slow - plz ping), 2014-07-10 09:41:57]

On 2014/07/09 18:49:31, junov wrote:
> Author is not online. Reverting...

I was in a train with a very flaky connection. Unfortunately, I saw your ping
when I arrived and it was too late. Sorry about that.

This should now be fixed.

https://codereview.chromium.org/374623002/diff/160001/Source/modules/screen_o...
File Source/modules/screen_orientation/ScreenScreenOrientation.cpp (right):

https://codereview.chromium.org/374623002/diff/160001/Source/modules/screen_o...
Source/modules/screen_orientation/ScreenScreenOrientation.cpp:49:
m_orientation->trace(visitor);
On 2014/07/09 19:33:37, sof wrote:
> Oh, this needs to be visitor->trace(m_orientation) as you want to mark the
> ScreenOrientation via the visitor before later on tracing the actual object
(if
> not already marked.)

Done.

[mlamouri (slow - plz ping), 2014-07-10 09:42:13]

The CQ bit was checked by mlamouri@chromium.org

[commit-bot: I haz the power, 2014-07-10 09:42:54]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/mlamouri@chromium.org/374623002/180001

[commit-bot: I haz the power, 2014-07-10 10:49:49]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  linux_blink_dbg on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/linux_blink_dbg/builds/1...)
  mac_gpu_triggered_tests on tryserver.chromium.gpu
(http://build.chromium.org/p/tryserver.chromium.gpu/builders/mac_gpu_triggered...)

[commit-bot: I haz the power, 2014-07-10 11:58:49]

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

[commit-bot: I haz the power, 2014-07-10 11:58:50]

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

[mlamouri (slow - plz ping), 2014-07-10 17:14:39]

The CQ bit was checked by mlamouri@chromium.org

[commit-bot: I haz the power, 2014-07-10 17:15:27]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/mlamouri@chromium.org/374623002/180001

[commit-bot: I haz the power, 2014-07-10 17:19:18]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  linux_blink_dbg on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/linux_blink_dbg/builds/1...)

[commit-bot: I haz the power, 2014-07-10 17:22:21]

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

[commit-bot: I haz the power, 2014-07-10 17:22:25]

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

[mlamouri (slow - plz ping), 2014-07-11 09:06:24]

The CQ bit was checked by mlamouri@chromium.org

[commit-bot: I haz the power, 2014-07-11 09:07:22]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/mlamouri@chromium.org/374623002/200001

[commit-bot: I haz the power, 2014-07-11 10:20:15]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  linux_blink_dbg on tryserver.blink
(http://build.chromium.org/p/tryserver.blink/builders/linux_blink_dbg/builds/1...)

[commit-bot: I haz the power, 2014-07-11 10:30:04]

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

[commit-bot: I haz the power, 2014-07-11 10:30:05]

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

[mlamouri (slow - plz ping), 2014-07-11 11:10:51]

The CQ bit was checked by mlamouri@chromium.org

[commit-bot: I haz the power, 2014-07-11 11:11:56]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/mlamouri@chromium.org/374623002/200001

[commit-bot: I haz the power, 2014-07-11 12:17:34]

Change committed as 177933