Get rid of ChannelInit::SetSingleProcessIOTaskRunner()

Get rid of ChannelInit::SetSingleProcessIOTaskRunner()

The globally-set TaskRunner doesn't work because some child
thread can run inside the browser process even when other
child threads run in their own processes.
For example, there is a configuration where GpuChildThread
being in-process while RenderThreadImpl isn't.

It is troublesome once non-renderer child processes adopt
ChannelMojo.

This CL eliminates the global in-process IO task runner
and instead, passes an IO runnner for each thread
if it is configured as an in-process mode.

Note that InProcessChildThreadParams object is introuced to
clarify that the parameter is for in-process mode.

This is a spin-off from https://codereview.chromium.org/960693003/.

R=jamesr@chromium.org, rockot@chromium.org, jam@chromium.org
BUG=377980
Committed: https://crrev.com/c6238ab6b40c8f0a291f75dbceb58535cf942eac
Cr-Commit-Position: refs/heads/master@{#321047}

[jam, 2015-03-09 15:40:52]

Which parts do you want each of us to review?

[Hajime Morrita, 2015-03-09 17:05:54]

On 2015/03/09 15:40:52, jam wrote:
> Which parts do you want each of us to review?
It should've been clearer. Sorry about that.

- jam: content/
- jamesr: content/**/mojo
- rockot: channel_init.*

[Ken Rockot(use gerrit already), 2015-03-09 17:16:49]

On 2015/03/09 17:05:54, morrita wrote:
> On 2015/03/09 15:40:52, jam wrote:
> > Which parts do you want each of us to review?
> It should've been clearer. Sorry about that.
> 
> - jam: content/
> - jamesr: content/**/mojo
> - rockot: channel_init.*

Just a quick observation: It looks like there are some major problems with
browser tests in this CL. Have you been able to repro locally and find the
cause?

[jamesr, 2015-03-09 17:49:12]

error message is:

[24192:24282:0306/172301:ERROR:ipc_mojo_bootstrap.cc(105)] Got inconsistent
message from client.

[Hajime Morrita, 2015-03-09 18:21:39]

On 2015/03/09 17:49:12, jamesr wrote:
> error message is:
> 
> [24192:24282:0306/172301:ERROR:ipc_mojo_bootstrap.cc(105)] Got inconsistent
> message from client.

Hmm, looking... I'll make them green and ping you folks back. Sorry for
disturbing.

[Hajime Morrita, 2015-03-12 19:42:35]

OK, now it's green. PTAL?

- jam: content/
- jamesr: content/**/mojo
- rockot: channel_init.*

[jamesr, 2015-03-12 19:47:21]

lgtm, although I'm not an OWNER here

[Ken Rockot(use gerrit already), 2015-03-12 20:06:37]

also not an owner, but i've looked over everything here and it all lgtm.

https://codereview.chromium.org/987693005/diff/20001/content/child/child_thre...
File content/child/child_thread_impl.cc (right):

https://codereview.chromium.org/987693005/diff/20001/content/child/child_thre...
content/child/child_thread_impl.cc:53: #include
"content/common/mojo/channel_init.h"
nit: I believe you can kill this include now.

https://codereview.chromium.org/987693005/diff/20001/content/common/in_proces...
File content/common/in_process_child_thread_params.h (right):

https://codereview.chromium.org/987693005/diff/20001/content/common/in_proces...
content/common/in_process_child_thread_params.h:16: class CONTENT_EXPORT
InProcessChildThreadParams {
nit: Please document this class briefly

https://codereview.chromium.org/987693005/diff/20001/content/common/in_proces...
content/common/in_process_child_thread_params.h:31: };
nit: DISALLOW_COPY_AND_ASSIGN

[Hajime Morrita, 2015-03-12 22:08:19]

Thanks for the review folks!

John: PTAL for overall change under content/ ?
Mojo-specific bits seems good now.

https://codereview.chromium.org/987693005/diff/20001/content/child/child_thre...
File content/child/child_thread_impl.cc (right):

https://codereview.chromium.org/987693005/diff/20001/content/child/child_thre...
content/child/child_thread_impl.cc:53: #include
"content/common/mojo/channel_init.h"
On 2015/03/12 20:06:37, Ken Rockot wrote:
> nit: I believe you can kill this include now.

Good catch! Removed.

https://codereview.chromium.org/987693005/diff/20001/content/common/in_proces...
File content/common/in_process_child_thread_params.h (right):

https://codereview.chromium.org/987693005/diff/20001/content/common/in_proces...
content/common/in_process_child_thread_params.h:16: class CONTENT_EXPORT
InProcessChildThreadParams {
On 2015/03/12 20:06:37, Ken Rockot wrote:
> nit: Please document this class briefly

Done.

https://codereview.chromium.org/987693005/diff/20001/content/common/in_proces...
content/common/in_process_child_thread_params.h:31: };
On 2015/03/12 20:06:37, Ken Rockot wrote:
> nit: DISALLOW_COPY_AND_ASSIGN

This is copy-able.

[Hajime Morrita, 2015-03-16 17:39:54]

@jam: PTAL as a content/ OWNER?

[Hajime Morrita, 2015-03-17 17:23:34]

morrita@chromium.org changed reviewers:
+ nasko@chromium.org

[Hajime Morrita, 2015-03-17 17:23:35]

Nasko: Could you take a took as a content/ OWNER?
I was told that John is ooo this week and this CL blocks
next Mojo roll.

The essential part was already reviewed by relevant people.
What remaining is kind of integrity check.

[nasko, 2015-03-18 00:12:45]

LGTM with a few nits

https://codereview.chromium.org/987693005/diff/80001/content/common/in_proces...
File content/common/in_process_child_thread_params.cc (right):

https://codereview.chromium.org/987693005/diff/80001/content/common/in_proces...
content/common/in_process_child_thread_params.cc:1: // Copyright (c) 2012 The
Chromium Authors. All rights reserved.
nit: No need for "(c)" and we recently entered the year 2015 : )

https://codereview.chromium.org/987693005/diff/80001/content/common/in_proces...
File content/common/in_process_child_thread_params.h (right):

https://codereview.chromium.org/987693005/diff/80001/content/common/in_proces...
content/common/in_process_child_thread_params.h:1: // Copyright (c) 2015 The
Chromium Authors. All rights reserved.
nit: no "(c)"

https://codereview.chromium.org/987693005/diff/80001/content/gpu/gpu_child_th...
File content/gpu/gpu_child_thread.h (left):

https://codereview.chromium.org/987693005/diff/80001/content/gpu/gpu_child_th...
content/gpu/gpu_child_thread.h:45: // For single-process mode.
Is this comment no longer valid?

https://codereview.chromium.org/987693005/diff/80001/content/gpu/in_process_g...
File content/gpu/in_process_gpu_thread.h (right):

https://codereview.chromium.org/987693005/diff/80001/content/gpu/in_process_g...
content/gpu/in_process_gpu_thread.h:30: // Deleted in CleanUp() on the gpu
thread, so don't use smart pointers.
nit: Blank line above the comment.

https://codereview.chromium.org/987693005/diff/80001/content/renderer/render_...
File content/renderer/render_thread_impl.cc (right):

https://codereview.chromium.org/987693005/diff/80001/content/renderer/render_...
content/renderer/render_thread_impl.cc:448: 
nit: no need for the new empty line.

[Hajime Morrita, 2015-03-18 00:38:40]

Thanks much for the quick review! Landing with amendments.

https://codereview.chromium.org/987693005/diff/80001/content/common/in_proces...
File content/common/in_process_child_thread_params.cc (right):

https://codereview.chromium.org/987693005/diff/80001/content/common/in_proces...
content/common/in_process_child_thread_params.cc:1: // Copyright (c) 2012 The
Chromium Authors. All rights reserved.
On 2015/03/18 00:12:45, nasko wrote:
> nit: No need for "(c)" and we recently entered the year 2015 : )
:-/

https://codereview.chromium.org/987693005/diff/80001/content/common/in_proces...
content/common/in_process_child_thread_params.cc:1: // Copyright (c) 2012 The
Chromium Authors. All rights reserved.
On 2015/03/18 00:12:45, nasko wrote:
> nit: No need for "(c)" and we recently entered the year 2015 : )
:-/

https://codereview.chromium.org/987693005/diff/80001/content/common/in_proces...
File content/common/in_process_child_thread_params.h (right):

https://codereview.chromium.org/987693005/diff/80001/content/common/in_proces...
content/common/in_process_child_thread_params.h:1: // Copyright (c) 2015 The
Chromium Authors. All rights reserved.
On 2015/03/18 00:12:45, nasko wrote:
> nit: no "(c)"

Done.

https://codereview.chromium.org/987693005/diff/80001/content/gpu/gpu_child_th...
File content/gpu/gpu_child_thread.h (left):

https://codereview.chromium.org/987693005/diff/80001/content/gpu/gpu_child_th...
content/gpu/gpu_child_thread.h:45: // For single-process mode.
On 2015/03/18 00:12:45, nasko wrote:
> Is this comment no longer valid?
This is valid, but I think the parameter name explains
it well enough.

https://codereview.chromium.org/987693005/diff/80001/content/gpu/in_process_g...
File content/gpu/in_process_gpu_thread.h (right):

https://codereview.chromium.org/987693005/diff/80001/content/gpu/in_process_g...
content/gpu/in_process_gpu_thread.h:30: // Deleted in CleanUp() on the gpu
thread, so don't use smart pointers.
On 2015/03/18 00:12:45, nasko wrote:
> nit: Blank line above the comment.

Done.

https://codereview.chromium.org/987693005/diff/80001/content/renderer/render_...
File content/renderer/render_thread_impl.cc (right):

https://codereview.chromium.org/987693005/diff/80001/content/renderer/render_...
content/renderer/render_thread_impl.cc:448: 
On 2015/03/18 00:12:45, nasko wrote:
> nit: no need for the new empty line.

Done.

[Hajime Morrita, 2015-03-18 00:40:44]

Landing

[Hajime Morrita, 2015-03-18 00:40:45]

The CQ bit was checked by morrita@chromium.org

[Hajime Morrita, 2015-03-18 00:40:46]

The patchset sent to the CQ was uploaded after l-g-t-m from jamesr@chromium.org,
rockot@chromium.org, nasko@chromium.org
Link to the patchset: https://codereview.chromium.org/987693005/#ps100001
(title: "Landing")

[commit-bot: I haz the power, 2015-03-18 00:41:22]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/987693005/100001

[commit-bot: I haz the power, 2015-03-18 01:48:45]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2015-03-18 01:52:34]

Patchset 6 (id:??) landed as
https://crrev.com/c6238ab6b40c8f0a291f75dbceb58535cf942eac
Cr-Commit-Position: refs/heads/master@{#321047}