exo: Implement support for zcr_keyboard_configuration_v1 protocol.

exo: Implement support for zcr_keyboard_configuration_v1 protocol.

This CL allows us to notify keyboard type changes to a client
through zcr_keyboard_device_configuration_v1 interface.

BUG=670247
Committed: https://crrev.com/5d5b8cb5a204c2317e08262426b2a19c1c562a35
Cr-Commit-Position: refs/heads/master@{#437750}

[yhanada, 2016-12-06 08:27:48]

Description was changed from

==========
[WIP] Add an extension to notify what type of a keyboard is connected.

BUG=b/30757549
==========

to

==========
[WIP] exo: Implement support for zcr_keyboard_extension_v1 protocol.
    
This CL allows us to notify keyboard type changes to a client
through zcr_extended_keyboard_v1 protocol.
    
BUG=670247
==========

[yhanada, 2016-12-06 12:15:13]

The CQ bit was checked by yhanada@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-06 12:15:26]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-12-06 12:23:25]

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

[commit-bot: I haz the power, 2016-12-06 12:23:26]

Dry run: Try jobs failed on following builders:
  chromium_presubmit on master.tryserver.chromium.linux (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromium_presub...)

[yhanada, 2016-12-06 12:27:17]

Description was changed from

==========
[WIP] exo: Implement support for zcr_keyboard_extension_v1 protocol.
    
This CL allows us to notify keyboard type changes to a client
through zcr_extended_keyboard_v1 protocol.
    
BUG=670247
==========

to

==========
exo: Implement support for zcr_keyboard_extension_v1 protocol.
    
This CL allows us to notify keyboard type changes to a client
through zcr_extended_keyboard_v1 protocol.
    
BUG=670247
==========

[yhanada, 2016-12-08 23:52:55]

Description was changed from

==========
exo: Implement support for zcr_keyboard_extension_v1 protocol.
    
This CL allows us to notify keyboard type changes to a client
through zcr_extended_keyboard_v1 protocol.
    
BUG=670247
==========

to

==========
exo: Implement support for zcr_keyboard_configuration_v1 protocol.

This CL allows us to notify keyboard type changes to a client
through zcr_keyboard_device_configuration_v1 interface.
    
BUG=670247
==========

[yhanada, 2016-12-08 23:54:48]

yhanada@chromium.org changed reviewers:
+ denniskempin@chromium.org, reveman@chromium.org

[yhanada, 2016-12-08 23:54:49]

Hi denniskempin@ and reveman@, PTAL.

[reveman, 2016-12-09 00:18:37]

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard.cc
File components/exo/keyboard.cc (right):

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:77: bool HasEnabledPhysicalKeyboard() {
IsPhysicalKeyboardEnabled() ?

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:78: const auto* helper = WMHelper::GetInstance();
do we need this temporary variable and the dcheck below?

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:80: DCHECK(helper != nullptr);
DCHECK(helper) if we need to keep this

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:82: bool has_keyboard =
!helper->IsMaximizeModeWindowManagerEnabled();
why not return early here?

if (!WMHelper::GetInstance()->IsMaximizeModeWindowManagerEnabled())
  return true;

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:84: const auto* manager =
ui::InputDeviceManager::GetInstance();
do we need this temporary variable and the dcheck below?

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:86: for (const ui::InputDevice& keyboard :
manager->GetKeyboardDevices()) {
maybe use "auto& keyboard" here

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:88: (keyboard.type !=
ui::InputDeviceType::INPUT_DEVICE_INTERNAL);
return early here instead:

if (keyboard.type != ui::InputDeviceType::INPUT_DEVICE_INTERNAL)
  return true;

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:121: if (device_configuration_delegate_)
Call OnKeyboardDeviceConfigurationChanged() here instead?

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:210: if (device_configuration_delegate_)
need "{}" for multi line if-statement

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:219: if (device_configuration_delegate_)
need "{}" for multi line if-statement

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:225: if (device_configuration_delegate_)
need "{}" for multi line if-statement

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard.h
File components/exo/keyboard.h (right):

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.h:60: // The delegate instance that all events of
wl_keyboard are dispatched to.
we should not be referring to wayland interfaces in this code

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.h:63: // The delegate instance that all events of
zcr_keyboard_device_configuration
ditto. please avoid referring to wayland interfaces in this code. can you
explain this without mentioning this interface?

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
File components/exo/wayland/server.cc (right):

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
components/exo/wayland/server.cc:3012:
keyboard_->SetKeyboardDeviceConfigurationDelegate(this);
what if the keyboard already has a device config delegate?

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
components/exo/wayland/server.cc:3015: if (keyboard_ != nullptr)
s/keyboard_ != nullptr/keyboard/

but how can keyboard_ be null? and how are we sure it's still a valid pointer
that has not been freed?

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
components/exo/wayland/server.cc:3020: const uint32_t type =
nit: remove "const" or maybe even better, remove the temporary variable
completely

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
components/exo/wayland/server.cc:3044: // keyboard extension interface:
// keyboard configuration interface:

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
components/exo/wayland/server.cc:3046: void
keyboard_extension_get_keyboard_device_configuration(
keyboard_configuration_get_keyboard_device_configuration

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
components/exo/wayland/server.cc:3051: Keyboard* keyboard =
GetUserDataAs<Keyboard>(keyboard_resource);
move this down to where it's first used. do you need the temporary variable?

[yhanada, 2016-12-09 14:54:05]

Thank you for the quick review!

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard.cc
File components/exo/keyboard.cc (right):

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:77: bool HasEnabledPhysicalKeyboard() {
On 2016/12/09 00:18:37, reveman wrote:
> IsPhysicalKeyboardEnabled() ?

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:78: const auto* helper = WMHelper::GetInstance();
On 2016/12/09 00:18:37, reveman wrote:
> do we need this temporary variable and the dcheck below?

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:80: DCHECK(helper != nullptr);
On 2016/12/09 00:18:36, reveman wrote:
> DCHECK(helper) if we need to keep this

Acknowledged.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:82: bool has_keyboard =
!helper->IsMaximizeModeWindowManagerEnabled();
On 2016/12/09 00:18:37, reveman wrote:
> why not return early here?
> 
> if (!WMHelper::GetInstance()->IsMaximizeModeWindowManagerEnabled())
>   return true;

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:84: const auto* manager =
ui::InputDeviceManager::GetInstance();
On 2016/12/09 00:18:37, reveman wrote:
> do we need this temporary variable and the dcheck below?

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:86: for (const ui::InputDevice& keyboard :
manager->GetKeyboardDevices()) {
On 2016/12/09 00:18:37, reveman wrote:
> maybe use "auto& keyboard" here

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:88: (keyboard.type !=
ui::InputDeviceType::INPUT_DEVICE_INTERNAL);
On 2016/12/09 00:18:37, reveman wrote:
> return early here instead:
> 
> if (keyboard.type != ui::InputDeviceType::INPUT_DEVICE_INTERNAL)
>   return true;

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:121: if (device_configuration_delegate_)
On 2016/12/09 00:18:37, reveman wrote:
> Call OnKeyboardDeviceConfigurationChanged() here instead?

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:210: if (device_configuration_delegate_)
On 2016/12/09 00:18:36, reveman wrote:
> need "{}" for multi line if-statement

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:219: if (device_configuration_delegate_)
On 2016/12/09 00:18:37, reveman wrote:
> need "{}" for multi line if-statement

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.cc:225: if (device_configuration_delegate_)
On 2016/12/09 00:18:37, reveman wrote:
> need "{}" for multi line if-statement

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard.h
File components/exo/keyboard.h (right):

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.h:60: // The delegate instance that all events of
wl_keyboard are dispatched to.
On 2016/12/09 00:18:37, reveman wrote:
> we should not be referring to wayland interfaces in this code

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/keyboard...
components/exo/keyboard.h:63: // The delegate instance that all events of
zcr_keyboard_device_configuration
On 2016/12/09 00:18:37, reveman wrote:
> ditto. please avoid referring to wayland interfaces in this code. can you
> explain this without mentioning this interface?

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
File components/exo/wayland/server.cc (right):

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
components/exo/wayland/server.cc:3012:
keyboard_->SetKeyboardDeviceConfigurationDelegate(this);
On 2016/12/09 00:18:37, reveman wrote:
> what if the keyboard already has a device config delegate?

I added a call of |OnKeyboardDestroying| in
|Keyboard::SetDeviceConfigurationDelegate|.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
components/exo/wayland/server.cc:3015: if (keyboard_ != nullptr)
On 2016/12/09 00:18:37, reveman wrote:
> s/keyboard_ != nullptr/keyboard/
> 
> but how can keyboard_ be null? and how are we sure it's still a valid pointer
> that has not been freed? 

In order to avoid referencing a invalid pointer, I added |OnKeyboardDestroying|
to |KeyboardDeviceConfigurationDelegate| just like |PointerStylusDelegate|

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
components/exo/wayland/server.cc:3020: const uint32_t type =
On 2016/12/09 00:18:37, reveman wrote:
> nit: remove "const" or maybe even better, remove the temporary variable
> completely

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
components/exo/wayland/server.cc:3044: // keyboard extension interface:
On 2016/12/09 00:18:37, reveman wrote:
> // keyboard configuration interface:

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
components/exo/wayland/server.cc:3046: void
keyboard_extension_get_keyboard_device_configuration(
On 2016/12/09 00:18:37, reveman wrote:
> keyboard_configuration_get_keyboard_device_configuration

Done.

https://codereview.chromium.org/2506263002/diff/60001/components/exo/wayland/...
components/exo/wayland/server.cc:3051: Keyboard* keyboard =
GetUserDataAs<Keyboard>(keyboard_resource);
On 2016/12/09 00:18:37, reveman wrote:
> move this down to where it's first used. do you need the temporary variable?

Done.

[yhanada, 2016-12-09 14:54:11]

The CQ bit was checked by yhanada@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-09 14:54:29]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[yhanada, 2016-12-09 15:37:45]

The CQ bit was checked by yhanada@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-09 15:38:08]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[reveman, 2016-12-09 16:15:03]

https://codereview.chromium.org/2506263002/diff/120001/components/exo/keyboar...
File components/exo/keyboard.cc (right):

https://codereview.chromium.org/2506263002/diff/120001/components/exo/keyboar...
components/exo/keyboard.cc:120:
device_configuration_delegate_->OnKeyboardDestroying(this);
This is surprising and inconsistent with SetStylusDelegate. Please remove it
from here and make sure the caller is handling the case where a stylus delegate
already exists properly.

https://codereview.chromium.org/2506263002/diff/120001/components/exo/keyboar...
components/exo/keyboard.cc:219: if (device_configuration_delegate_) {
maybe we can just call OnKeyboardDeviceConfigurationChanged() here instead to
avoid duplicating the if statement?

https://codereview.chromium.org/2506263002/diff/120001/components/exo/keyboar...
components/exo/keyboard.cc:226: if (device_configuration_delegate_) {
call OnKeyboardDeviceConfigurationChanged() here?

https://codereview.chromium.org/2506263002/diff/120001/components/exo/keyboard.h
File components/exo/keyboard.h (right):

https://codereview.chromium.org/2506263002/diff/120001/components/exo/keyboar...
components/exo/keyboard.h:34: 
nit: no need for this blankline. ctor/dtor can be grouped

https://codereview.chromium.org/2506263002/diff/120001/components/exo/wayland...
File components/exo/wayland/server.cc (right):

https://codereview.chromium.org/2506263002/diff/120001/components/exo/wayland...
components/exo/wayland/server.cc:3029: if (is_physical) {
you can use ...send_type_change(resource_, is_physical ?
..KEYBOARD_TYPE_PHYSICAL : KEYBOARD_TYPE_VIRTUAL) here if you prefer. your call

https://codereview.chromium.org/2506263002/diff/120001/components/exo/wayland...
components/exo/wayland/server.cc:3057: // keyboard configuration interface:
nit: s/keyboard configuration/keyboard_configuration/

https://codereview.chromium.org/2506263002/diff/120001/components/exo/wayland...
components/exo/wayland/server.cc:3059: void
keyboard_configuration_get_keyboard_device_configuration(
I think there's an issue here if the client requests a keyboard device config
interface twice but stylus has the same issue so I having it be consistent with
stylus for now and fix both in a follow up patch is better. the best way to
handle that is likely to produce an error if a delegate already exists.

[commit-bot: I haz the power, 2016-12-09 16:38:54]

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

[commit-bot: I haz the power, 2016-12-09 16:38:55]

Dry run: This issue passed the CQ dry run.

[yhanada, 2016-12-10 00:09:11]

PTAL.

https://codereview.chromium.org/2506263002/diff/120001/components/exo/keyboar...
File components/exo/keyboard.cc (right):

https://codereview.chromium.org/2506263002/diff/120001/components/exo/keyboar...
components/exo/keyboard.cc:120:
device_configuration_delegate_->OnKeyboardDestroying(this);
On 2016/12/09 16:15:02, reveman wrote:
> This is surprising and inconsistent with SetStylusDelegate. Please remove it
> from here and make sure the caller is handling the case where a stylus
delegate
> already exists properly.

Done.

https://codereview.chromium.org/2506263002/diff/120001/components/exo/keyboar...
components/exo/keyboard.cc:219: if (device_configuration_delegate_) {
On 2016/12/09 16:15:02, reveman wrote:
> maybe we can just call OnKeyboardDeviceConfigurationChanged() here instead to
> avoid duplicating the if statement?

Done.

https://codereview.chromium.org/2506263002/diff/120001/components/exo/keyboar...
components/exo/keyboard.cc:226: if (device_configuration_delegate_) {
On 2016/12/09 16:15:02, reveman wrote:
> call OnKeyboardDeviceConfigurationChanged() here?

Done.

https://codereview.chromium.org/2506263002/diff/120001/components/exo/keyboard.h
File components/exo/keyboard.h (right):

https://codereview.chromium.org/2506263002/diff/120001/components/exo/keyboar...
components/exo/keyboard.h:34: 
On 2016/12/09 16:15:02, reveman wrote:
> nit: no need for this blankline. ctor/dtor can be grouped

Done.

https://codereview.chromium.org/2506263002/diff/120001/components/exo/wayland...
File components/exo/wayland/server.cc (right):

https://codereview.chromium.org/2506263002/diff/120001/components/exo/wayland...
components/exo/wayland/server.cc:3029: if (is_physical) {
On 2016/12/09 16:15:02, reveman wrote:
> you can use ...send_type_change(resource_, is_physical ?
> ..KEYBOARD_TYPE_PHYSICAL : KEYBOARD_TYPE_VIRTUAL) here if you prefer. your
call

Done.

https://codereview.chromium.org/2506263002/diff/120001/components/exo/wayland...
components/exo/wayland/server.cc:3057: // keyboard configuration interface:
On 2016/12/09 16:15:02, reveman wrote:
> nit: s/keyboard configuration/keyboard_configuration/

Done.

https://codereview.chromium.org/2506263002/diff/120001/components/exo/wayland...
components/exo/wayland/server.cc:3059: void
keyboard_configuration_get_keyboard_device_configuration(
On 2016/12/09 16:15:02, reveman wrote:
> I think there's an issue here if the client requests a keyboard device config
> interface twice but stylus has the same issue so I having it be consistent
with
> stylus for now and fix both in a follow up patch is better. the best way to
> handle that is likely to produce an error if a delegate already exists.

I see. I added TODO here. Should I added TODO in |stylus_get_pointer_stylus|?

[yhanada, 2016-12-10 00:09:20]

The CQ bit was checked by yhanada@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-10 00:10:14]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[reveman, 2016-12-10 00:13:55]

lgtm

https://codereview.chromium.org/2506263002/diff/120001/components/exo/wayland...
File components/exo/wayland/server.cc (right):

https://codereview.chromium.org/2506263002/diff/120001/components/exo/wayland...
components/exo/wayland/server.cc:3059: void
keyboard_configuration_get_keyboard_device_configuration(
On 2016/12/10 at 00:09:11, yhanada wrote:
> On 2016/12/09 16:15:02, reveman wrote:
> > I think there's an issue here if the client requests a keyboard device
config
> > interface twice but stylus has the same issue so I having it be consistent
with
> > stylus for now and fix both in a follow up patch is better. the best way to
> > handle that is likely to produce an error if a delegate already exists.
> 
> I see. I added TODO here. Should I added TODO in |stylus_get_pointer_stylus|?

Yes, adding a TODO for the stylus too would be nice. Thanks.

[yhanada, 2016-12-10 00:47:28]

Thank you for reviewing!

https://codereview.chromium.org/2506263002/diff/120001/components/exo/wayland...
File components/exo/wayland/server.cc (right):

https://codereview.chromium.org/2506263002/diff/120001/components/exo/wayland...
components/exo/wayland/server.cc:3059: void
keyboard_configuration_get_keyboard_device_configuration(
On 2016/12/10 00:13:55, reveman wrote:
> On 2016/12/10 at 00:09:11, yhanada wrote:
> > On 2016/12/09 16:15:02, reveman wrote:
> > > I think there's an issue here if the client requests a keyboard device
> config
> > > interface twice but stylus has the same issue so I having it be consistent
> with
> > > stylus for now and fix both in a follow up patch is better. the best way
to
> > > handle that is likely to produce an error if a delegate already exists.
> > 
> > I see. I added TODO here. Should I added TODO in
|stylus_get_pointer_stylus|?
> 
> Yes, adding a TODO for the stylus too would be nice. Thanks. 

Done.

[yhanada, 2016-12-10 00:47:46]

The CQ bit was checked by yhanada@chromium.org

[yhanada, 2016-12-10 00:47:47]

The patchset sent to the CQ was uploaded after l-g-t-m from reveman@chromium.org
Link to the patchset: https://codereview.chromium.org/2506263002/#ps160001
(title: "add one more TODO")

[commit-bot: I haz the power, 2016-12-10 00:48:45]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-12-10 04:16:07]

CQ is committing da patch.

Bot data: {"patchset_id": 160001, "attempt_start_ts": 1481330866928820,
"parent_rev": "4ba21546871f1c467bd8a0d7e473233641eb39a7", "commit_rev":
"cb2bf293d00bcba56f4b88e9d0aa0aff973e9043"}

[commit-bot: I haz the power, 2016-12-10 04:17:26]

Description was changed from

==========
exo: Implement support for zcr_keyboard_configuration_v1 protocol.

This CL allows us to notify keyboard type changes to a client
through zcr_keyboard_device_configuration_v1 interface.
    
BUG=670247
==========

to

==========
exo: Implement support for zcr_keyboard_configuration_v1 protocol.

This CL allows us to notify keyboard type changes to a client
through zcr_keyboard_device_configuration_v1 interface.
    
BUG=670247

Review-Url: https://codereview.chromium.org/2506263002
==========

[commit-bot: I haz the power, 2016-12-10 04:17:27]

Committed patchset #9 (id:160001)

[commit-bot: I haz the power, 2016-12-12 15:06:20]

Description was changed from

==========
exo: Implement support for zcr_keyboard_configuration_v1 protocol.

This CL allows us to notify keyboard type changes to a client
through zcr_keyboard_device_configuration_v1 interface.
    
BUG=670247

Review-Url: https://codereview.chromium.org/2506263002
==========

to

==========
exo: Implement support for zcr_keyboard_configuration_v1 protocol.

This CL allows us to notify keyboard type changes to a client
through zcr_keyboard_device_configuration_v1 interface.

BUG=670247

Committed: https://crrev.com/5d5b8cb5a204c2317e08262426b2a19c1c562a35
Cr-Commit-Position: refs/heads/master@{#437750}
==========

[commit-bot: I haz the power, 2016-12-12 15:06:21]

Patchset 9 (id:??) landed as
https://crrev.com/5d5b8cb5a204c2317e08262426b2a19c1c562a35
Cr-Commit-Position: refs/heads/master@{#437750}