Re-reland: Add MemoryCoordinator

Re-reland: Add MemoryCoordinator

This is fourth attempt... Pass InterfacePtr instead of InterfaceRegistry
to avoid dependency from components/memory_coordinator to shell/services

Original description:

This CL adds MemoryCoordinator mojo interface and
its implementation. When --enable-features=MemoryCoordinator
is specified the browser and renderers create their
MemoryCoordinators. MemoryCoordinatorImpl (which lives in
the browser process) creates a mojo message pipe when a
renderer is launched.

This CL doesn't add actual handling of memory events, which
will be implemented in follow-up CLs.

BUG=617492
TBR=chrisha@chromium.org,haraken@chromium.org,dcheng@chromium.org,jam@chromium.org

Committed: https://crrev.com/77f19b3affd0f29a77d1db1eebb6cc00e5a45898
Cr-Commit-Position: refs/heads/master@{#406247}

[bashi, 2016-07-15 06:38:43]

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

[commit-bot: I haz the power, 2016-07-15 06:38:55]

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, 2016-07-15 08:13:42]

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

[commit-bot: I haz the power, 2016-07-15 08:13:44]

Dry run: This issue passed the CQ dry run.

[bashi, 2016-07-19 02:09:58]

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

[commit-bot: I haz the power, 2016-07-19 02:10:43]

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

[dcheng, 2016-07-19 02:14:23]

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
File components/memory_coordinator/child/child_memory_coordinator_impl.cc
(right):

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
components/memory_coordinator/child/child_memory_coordinator_impl.cc:12: if
(parent_.is_bound()) {
(Btw, why would this not be bound?)

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
File components/memory_coordinator/child/child_memory_coordinator_impl.h
(right):

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
components/memory_coordinator/child/child_memory_coordinator_impl.h:24:
ChildMemoryCoordinatorImpl(mojom::MemoryCoordinatorHandlePtr parent);
Nit: explicit

[bashi, 2016-07-19 02:23:49]

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

[commit-bot: I haz the power, 2016-07-19 02:24:25]

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

[bashi, 2016-07-19 02:24:49]

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
File components/memory_coordinator/child/child_memory_coordinator_impl.cc
(right):

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
components/memory_coordinator/child/child_memory_coordinator_impl.cc:12: if
(parent_.is_bound()) {
On 2016/07/19 02:14:23, dcheng wrote:
> (Btw, why would this not be bound?)

In unittest I don't pass a bound pointer.

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
File components/memory_coordinator/child/child_memory_coordinator_impl.h
(right):

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
components/memory_coordinator/child/child_memory_coordinator_impl.h:24:
ChildMemoryCoordinatorImpl(mojom::MemoryCoordinatorHandlePtr parent);
On 2016/07/19 02:14:23, dcheng wrote:
> Nit: explicit

Thanks. Done.

[commit-bot: I haz the power, 2016-07-19 03:51:26]

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

[commit-bot: I haz the power, 2016-07-19 03:51:26]

Dry run: Try jobs failed on following builders:
  win_chromium_rel_ng on master.tryserver.chromium.win (JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_...)

[dcheng, 2016-07-19 04:54:07]

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
File components/memory_coordinator/child/child_memory_coordinator_impl.cc
(right):

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
components/memory_coordinator/child/child_memory_coordinator_impl.cc:12: if
(parent_.is_bound()) {
On 2016/07/19 02:24:49, bashi1 wrote:
> On 2016/07/19 02:14:23, dcheng wrote:
> > (Btw, why would this not be bound?)
> 
> In unittest I don't pass a bound pointer.

Is it possible to make the unit test pass a bound pointer, or is that too hard?
It would be nice to not have test only code if possible.

[bashi, 2016-07-19 06:02:08]

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
File components/memory_coordinator/child/child_memory_coordinator_impl.cc
(right):

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
components/memory_coordinator/child/child_memory_coordinator_impl.cc:12: if
(parent_.is_bound()) {
On 2016/07/19 04:54:07, dcheng wrote:
> On 2016/07/19 02:24:49, bashi1 wrote:
> > On 2016/07/19 02:14:23, dcheng wrote:
> > > (Btw, why would this not be bound?)
> > 
> > In unittest I don't pass a bound pointer.
> 
> Is it possible to make the unit test pass a bound pointer, or is that too
hard?
> It would be nice to not have test only code if possible.

Tried in PS#5. ChildMemoryCoordinatorImplTest becomes a bit complicated. Does it
make sense to you?

[dcheng, 2016-07-19 07:19:05]

LGTM

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
File components/memory_coordinator/child/child_memory_coordinator_impl.cc
(right):

https://codereview.chromium.org/2149413002/diff/20001/components/memory_coord...
components/memory_coordinator/child/child_memory_coordinator_impl.cc:12: if
(parent_.is_bound()) {
On 2016/07/19 06:02:08, bashi1 wrote:
> On 2016/07/19 04:54:07, dcheng wrote:
> > On 2016/07/19 02:24:49, bashi1 wrote:
> > > On 2016/07/19 02:14:23, dcheng wrote:
> > > > (Btw, why would this not be bound?)
> > > 
> > > In unittest I don't pass a bound pointer.
> > 
> > Is it possible to make the unit test pass a bound pointer, or is that too
> hard?
> > It would be nice to not have test only code if possible.
> 
> Tried in PS#5. ChildMemoryCoordinatorImplTest becomes a bit complicated. Does
it
> make sense to you?

Yeah, it's a bit more complicated but I think it's a bit nicer to put the burden
on tests rather than on non-testing code.

[bashi, 2016-07-19 08:15:45]

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

[commit-bot: I haz the power, 2016-07-19 08:16:07]

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, 2016-07-19 10:23:51]

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

[commit-bot: I haz the power, 2016-07-19 10:23:52]

Dry run: This issue passed the CQ dry run.

[bashi, 2016-07-19 11:14:41]

tasak@ and tzik@ helped me to build chrome with this CL on win/gyp and it seems
there is no error. Let me try to land this again.

[bashi, 2016-07-19 11:14:51]

The CQ bit was checked by bashi@chromium.org

[commit-bot: I haz the power, 2016-07-19 11:15:06]

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, 2016-07-19 11:20:02]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2016-07-19 11:20:06]

CQ bit was unchecked.

[commit-bot: I haz the power, 2016-07-19 11:22:57]

Description was changed from

==========
Re-reland: Add MemoryCoordinator

This is fourth attempt... Pass InterfacePtr instead of InterfaceRegistry
to avoid dependency from components/memory_coordinator to shell/services

Original description:

This CL adds MemoryCoordinator mojo interface and
its implementation. When --enable-features=MemoryCoordinator
is specified the browser and renderers create their
MemoryCoordinators. MemoryCoordinatorImpl (which lives in
the browser process) creates a mojo message pipe when a
renderer is launched.

This CL doesn't add actual handling of memory events, which
will be implemented in follow-up CLs.

BUG=617492
TBR=chrisha@chromium.org,haraken@chromium.org,dcheng@chromium.org,jam@chromiu...
==========

to

==========
Re-reland: Add MemoryCoordinator

This is fourth attempt... Pass InterfacePtr instead of InterfaceRegistry
to avoid dependency from components/memory_coordinator to shell/services

Original description:

This CL adds MemoryCoordinator mojo interface and
its implementation. When --enable-features=MemoryCoordinator
is specified the browser and renderers create their
MemoryCoordinators. MemoryCoordinatorImpl (which lives in
the browser process) creates a mojo message pipe when a
renderer is launched.

This CL doesn't add actual handling of memory events, which
will be implemented in follow-up CLs.

BUG=617492
TBR=chrisha@chromium.org,haraken@chromium.org,dcheng@chromium.org,jam@chromiu...

Committed: https://crrev.com/77f19b3affd0f29a77d1db1eebb6cc00e5a45898
Cr-Commit-Position: refs/heads/master@{#406247}
==========

[commit-bot: I haz the power, 2016-07-19 11:22:58]

Patchset 5 (id:??) landed as
https://crrev.com/77f19b3affd0f29a77d1db1eebb6cc00e5a45898
Cr-Commit-Position: refs/heads/master@{#406247}