Revert of Create GPU child window on new thread.

Revert of Create GPU child window on new thread. (patchset #2 id:20001 of https://codereview.chromium.org/2208153002/ )

Reason for revert:
This is most likely causing the window received from the GPU process to be invalid.

BUG=612300
Original issue's description:
> Create GPU child window on new thread.
>
> On Windows 10 when processing touch events, the GPU process main thread
> can block waiting for the browser UI thread to process the events. This
> can cause a deadlock if the browser process has sent a synchronous IPC
> to the GPU process.
>
> We can work around this by creating all GPU process windows on separate
> threads. Then the GPU process main thread won't attach input queues with
> the browser process main thread, which reduce the risk of deadlock.
> There's still a small risk of deadlock when the window is resized, which
> causes the GPU main thread to need to wait for the window owner thread.
>
> BUG=596190
> CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
>
> Committed: https://crrev.com/6b820b8ad092a59c8315c5ed942d21f723890ca1
> Cr-Commit-Position: refs/heads/master@{#410902}

TBR=stanisc@chromium.org,piman@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=596190
Committed: https://crrev.com/e5c83eff4536ffed87306d331c885c7e5210514c
Cr-Commit-Position: refs/heads/master@{#411159}

[jbauman, 2016-08-10 22:02:24]

The CQ bit was checked by jbauman@chromium.org

[jbauman, 2016-08-10 22:02:25]

Created Revert of Create GPU child window on new thread.

[commit-bot: I haz the power, 2016-08-10 22:04:00]

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, 2016-08-10 22:05:35]

Description was changed from

==========
Revert of Create GPU child window on new thread. (patchset #2 id:20001 of
https://codereview.chromium.org/2208153002/ )

Reason for revert:
This is most likely causing the window received from the GPU process to be
invalid.

BUG=612300

Original issue's description:
> Create GPU child window on new thread.
> 
> On Windows 10 when processing touch events, the GPU process main thread
> can block waiting for the browser UI thread to process the events. This
> can cause a deadlock if the browser process has sent a synchronous IPC
> to the GPU process.
> 
> We can work around this by creating all GPU process windows on separate
> threads. Then the GPU process main thread won't attach input queues with
> the browser process main thread, which reduce the risk of deadlock.
> There's still a small risk of deadlock when the window is resized, which
> causes the GPU main thread to need to wait for the window owner thread.
> 
> BUG=596190
>
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
> 
> Committed: https://crrev.com/6b820b8ad092a59c8315c5ed942d21f723890ca1
> Cr-Commit-Position: refs/heads/master@{#410902}

TBR=stanisc@chromium.org,piman@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=596190
==========

to

==========
Revert of Create GPU child window on new thread. (patchset #2 id:20001 of
https://codereview.chromium.org/2208153002/ )

Reason for revert:
This is most likely causing the window received from the GPU process to be
invalid.

BUG=612300

Original issue's description:
> Create GPU child window on new thread.
> 
> On Windows 10 when processing touch events, the GPU process main thread
> can block waiting for the browser UI thread to process the events. This
> can cause a deadlock if the browser process has sent a synchronous IPC
> to the GPU process.
> 
> We can work around this by creating all GPU process windows on separate
> threads. Then the GPU process main thread won't attach input queues with
> the browser process main thread, which reduce the risk of deadlock.
> There's still a small risk of deadlock when the window is resized, which
> causes the GPU main thread to need to wait for the window owner thread.
> 
> BUG=596190
>
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
> 
> Committed: https://crrev.com/6b820b8ad092a59c8315c5ed942d21f723890ca1
> Cr-Commit-Position: refs/heads/master@{#410902}

TBR=stanisc@chromium.org,piman@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=596190
==========

[commit-bot: I haz the power, 2016-08-10 22:05:36]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-08-10 22:07:53]

Description was changed from

==========
Revert of Create GPU child window on new thread. (patchset #2 id:20001 of
https://codereview.chromium.org/2208153002/ )

Reason for revert:
This is most likely causing the window received from the GPU process to be
invalid.

BUG=612300

Original issue's description:
> Create GPU child window on new thread.
> 
> On Windows 10 when processing touch events, the GPU process main thread
> can block waiting for the browser UI thread to process the events. This
> can cause a deadlock if the browser process has sent a synchronous IPC
> to the GPU process.
> 
> We can work around this by creating all GPU process windows on separate
> threads. Then the GPU process main thread won't attach input queues with
> the browser process main thread, which reduce the risk of deadlock.
> There's still a small risk of deadlock when the window is resized, which
> causes the GPU main thread to need to wait for the window owner thread.
> 
> BUG=596190
>
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
> 
> Committed: https://crrev.com/6b820b8ad092a59c8315c5ed942d21f723890ca1
> Cr-Commit-Position: refs/heads/master@{#410902}

TBR=stanisc@chromium.org,piman@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=596190
==========

to

==========
Revert of Create GPU child window on new thread. (patchset #2 id:20001 of
https://codereview.chromium.org/2208153002/ )

Reason for revert:
This is most likely causing the window received from the GPU process to be
invalid.

BUG=612300

Original issue's description:
> Create GPU child window on new thread.
>
> On Windows 10 when processing touch events, the GPU process main thread
> can block waiting for the browser UI thread to process the events. This
> can cause a deadlock if the browser process has sent a synchronous IPC
> to the GPU process.
>
> We can work around this by creating all GPU process windows on separate
> threads. Then the GPU process main thread won't attach input queues with
> the browser process main thread, which reduce the risk of deadlock.
> There's still a small risk of deadlock when the window is resized, which
> causes the GPU main thread to need to wait for the window owner thread.
>
> BUG=596190
>
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
>
> Committed: https://crrev.com/6b820b8ad092a59c8315c5ed942d21f723890ca1
> Cr-Commit-Position: refs/heads/master@{#410902}

TBR=stanisc@chromium.org,piman@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=596190

Committed: https://crrev.com/e5c83eff4536ffed87306d331c885c7e5210514c
Cr-Commit-Position: refs/heads/master@{#411159}
==========

[commit-bot: I haz the power, 2016-08-10 22:07:53]

Patchset 1 (id:??) landed as
https://crrev.com/e5c83eff4536ffed87306d331c885c7e5210514c
Cr-Commit-Position: refs/heads/master@{#411159}