android: Limit bindToCaller check to webview

android: Limit bindToCaller check to webview

There is evidence the BindToCaller check is causing to GPU hangs
although not sure about the exact mechanism. This change guards the
BindToCaller check behind a new setting in ChildProcessCreationParams
and then only enable it in webview.

Note also moved a few lines out of lock protection in
ChildProcessServiceImpl setupConnection.

This is intended to be merged back to m58.

BUG=690118
Review-Url: https://codereview.chromium.org/2796453003
Cr-Commit-Position: refs/heads/master@{#461883}
Committed: https://chromium.googlesource.com/chromium/src/+/2e90ee35449665137ff4bd8ff457d656b67ffa4c

[boliu, 2017-04-01 04:45:55]

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

[commit-bot: I haz the power, 2017-04-01 04:46:02]

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, 2017-04-01 05:34:24]

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

[commit-bot: I haz the power, 2017-04-01 05:34:24]

Dry run: This issue passed the CQ dry run.

[boliu, 2017-04-03 12:58:39]

Description was changed from

==========
android: Limit bindToCaller check to webview

BUG=690118
==========

to

==========
android: Limit bindToCaller check to webview

There is evidence the BindToCaller check is causing to GPU hangs
although not sure about the exact mechanism. This change guards the
BindToCaller check behind a new setting in ChildProcessCreationParams
and then only enable it in webview.

Note also moved a few lines out of lock protection in
ChildProcessServiceImpl setupConnection.

BUG=690118
==========

[boliu, 2017-04-03 12:59:14]

Description was changed from

==========
android: Limit bindToCaller check to webview

There is evidence the BindToCaller check is causing to GPU hangs
although not sure about the exact mechanism. This change guards the
BindToCaller check behind a new setting in ChildProcessCreationParams
and then only enable it in webview.

Note also moved a few lines out of lock protection in
ChildProcessServiceImpl setupConnection.

BUG=690118
==========

to

==========
android: Limit bindToCaller check to webview

There is evidence the BindToCaller check is causing to GPU hangs
although not sure about the exact mechanism. This change guards the
BindToCaller check behind a new setting in ChildProcessCreationParams
and then only enable it in webview.

Note also moved a few lines out of lock protection in
ChildProcessServiceImpl setupConnection.

This is intended to be merged back to m58.

BUG=690118
==========

[boliu, 2017-04-03 12:59:24]

boliu@chromium.org changed reviewers:
+ rsesek@chromium.org

[boliu, 2017-04-03 12:59:25]

ptal

[Robert Sesek, 2017-04-03 19:05:16]

LGTM

[boliu, 2017-04-03 19:10:46]

boliu@chromium.org changed reviewers:
+ mariakhomenko@chromium.org

[boliu, 2017-04-03 19:10:46]

+maria for chrome/

[boliu, 2017-04-04 16:07:11]

boliu@chromium.org changed reviewers:
+ nyquist@chromium.org

[boliu, 2017-04-04 16:07:12]

maria is busy today, +nyquist to for chrome/android?

[nyquist, 2017-04-04 18:30:33]

lgtm

https://codereview.chromium.org/2796453003/diff/20001/chrome/android/java/src...
File
chrome/android/java/src/org/chromium/chrome/browser/MonochromeApplication.java
(right):

https://codereview.chromium.org/2796453003/diff/20001/chrome/android/java/src...
chrome/android/java/src/org/chromium/chrome/browser/MonochromeApplication.java:27:
final boolean bindToCaller = false;
Nit: Why are all these final? We usually don't do that unless it's necessary in
//chrome.

[boliu, 2017-04-04 20:00:07]

https://codereview.chromium.org/2796453003/diff/20001/chrome/android/java/src...
File
chrome/android/java/src/org/chromium/chrome/browser/MonochromeApplication.java
(right):

https://codereview.chromium.org/2796453003/diff/20001/chrome/android/java/src...
chrome/android/java/src/org/chromium/chrome/browser/MonochromeApplication.java:27:
final boolean bindToCaller = false;
On 2017/04/04 18:30:33, nyquist wrote:
> Nit: Why are all these final? We usually don't do that unless it's necessary
in
> //chrome.

umm, I guess I'm carrying over from c++. these are more or less constants, and
shouldn't be reused for other things in the method

[boliu, 2017-04-04 21:21:55]

https://codereview.chromium.org/2796453003/diff/20001/chrome/android/java/src...
File
chrome/android/java/src/org/chromium/chrome/browser/MonochromeApplication.java
(right):

https://codereview.chromium.org/2796453003/diff/20001/chrome/android/java/src...
chrome/android/java/src/org/chromium/chrome/browser/MonochromeApplication.java:27:
final boolean bindToCaller = false;
On 2017/04/04 20:00:07, boliu wrote:
> On 2017/04/04 18:30:33, nyquist wrote:
> > Nit: Why are all these final? We usually don't do that unless it's necessary
> in
> > //chrome.
> 
> umm, I guess I'm carrying over from c++. these are more or less constants, and
> shouldn't be reused for other things in the method

removed final in chrome

[boliu, 2017-04-04 21:22:24]

The CQ bit was checked by boliu@chromium.org

[boliu, 2017-04-04 21:22:25]

The patchset sent to the CQ was uploaded after l-g-t-m from rsesek@chromium.org,
nyquist@chromium.org
Link to the patchset: https://codereview.chromium.org/2796453003/#ps40001
(title: "rebase, removed final from chrome")

[commit-bot: I haz the power, 2017-04-04 21:23:13]

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, 2017-04-04 23:05:41]

CQ is committing da patch.

Bot data: {"patchset_id": 40001, "attempt_start_ts": 1491340944841540,
"parent_rev": "72d9c9daf661d93d0d55a94071ad5567cd5adf6f", "commit_rev":
"2e90ee35449665137ff4bd8ff457d656b67ffa4c"}

[commit-bot: I haz the power, 2017-04-04 23:06:36]

Description was changed from

==========
android: Limit bindToCaller check to webview

There is evidence the BindToCaller check is causing to GPU hangs
although not sure about the exact mechanism. This change guards the
BindToCaller check behind a new setting in ChildProcessCreationParams
and then only enable it in webview.

Note also moved a few lines out of lock protection in
ChildProcessServiceImpl setupConnection.

This is intended to be merged back to m58.

BUG=690118
==========

to

==========
android: Limit bindToCaller check to webview

There is evidence the BindToCaller check is causing to GPU hangs
although not sure about the exact mechanism. This change guards the
BindToCaller check behind a new setting in ChildProcessCreationParams
and then only enable it in webview.

Note also moved a few lines out of lock protection in
ChildProcessServiceImpl setupConnection.

This is intended to be merged back to m58.

BUG=690118

Review-Url: https://codereview.chromium.org/2796453003
Cr-Commit-Position: refs/heads/master@{#461883}
Committed:
https://chromium.googlesource.com/chromium/src/+/2e90ee35449665137ff4bd8ff457...
==========

[commit-bot: I haz the power, 2017-04-04 23:06:37]

Committed patchset #3 (id:40001) as
https://chromium.googlesource.com/chromium/src/+/2e90ee35449665137ff4bd8ff457...