Oilpan: account for DevToolsEmulator reference from WebSettingsImpl.

Oilpan: account for DevToolsEmulator reference from WebSettingsImpl.

WebSettingsImpl and DevToolsEmulator are both owned by WebViewImpl;
let the former keep an untracked reference to the latter.

R=haraken
BUG=553613
Committed: https://crrev.com/05f9d3fd9ce5aae3d34a57af739a58297221c946
Cr-Commit-Position: refs/heads/master@{#359854}

[sof, 2015-11-12 21:08:47]

sigbjornf@opera.com changed reviewers:
+ oilpan-reviews@chromium.org

[haraken, 2015-11-13 11:47:51]

haraken@chromium.org changed reviewers:
+ haraken@chromium.org

[haraken, 2015-11-13 11:47:51]

https://codereview.chromium.org/1437003005/diff/1/third_party/WebKit/Source/w...
File third_party/WebKit/Source/web/WebSettingsImpl.h (right):

https://codereview.chromium.org/1437003005/diff/1/third_party/WebKit/Source/w...
third_party/WebKit/Source/web/WebSettingsImpl.h:211:
RawPtrWillBeUntracedMember<DevToolsEmulator> m_devToolsEmulator;

As commented in another CL, I think our plan is to use WeakPersistent (instead
of UntracedMember) for this kind of raw pointer. (Correct me if I'm
misunderstanding the previous discussion.)

[sof, 2015-11-13 13:31:48]

https://codereview.chromium.org/1437003005/diff/1/third_party/WebKit/Source/w...
File third_party/WebKit/Source/web/WebSettingsImpl.h (right):

https://codereview.chromium.org/1437003005/diff/1/third_party/WebKit/Source/w...
third_party/WebKit/Source/web/WebSettingsImpl.h:211:
RawPtrWillBeUntracedMember<DevToolsEmulator> m_devToolsEmulator;
On 2015/11/13 11:47:51, haraken wrote:
> 
> As commented in another CL, I think our plan is to use WeakPersistent (instead
> of UntracedMember) for this kind of raw pointer. (Correct me if I'm
> misunderstanding the previous discussion.)

See the description for why I consider that is excessive in this particular
case.

[haraken, 2015-11-13 13:39:41]

https://codereview.chromium.org/1437003005/diff/1/third_party/WebKit/Source/w...
File third_party/WebKit/Source/web/WebSettingsImpl.h (right):

https://codereview.chromium.org/1437003005/diff/1/third_party/WebKit/Source/w...
third_party/WebKit/Source/web/WebSettingsImpl.h:211:
RawPtrWillBeUntracedMember<DevToolsEmulator> m_devToolsEmulator;
On 2015/11/13 13:31:48, sof wrote:
> On 2015/11/13 11:47:51, haraken wrote:
> > 
> > As commented in another CL, I think our plan is to use WeakPersistent
(instead
> > of UntracedMember) for this kind of raw pointer. (Correct me if I'm
> > misunderstanding the previous discussion.)
> 
> See the description for why I consider that is excessive in this particular
> case.

LGTM.

It's still not quite clear to me when we should prefer UntracedMember though. I
was assuming that UntracedMembers should be avoided as much as possible (i.e.,
UntracedMembers are allowed only if using WeakPersistent has a risk of hitting
performance issues).

[sof, 2015-11-16 15:34:24]

Description was changed from

==========
Oilpan: account for DevToolsEmulator reference from WebSettingsImpl.

WebSettingsImpl and DevToolsEmulator are both owned by WebViewImpl;
let the former keep an untracked reference to the latter.

R=
BUG=553613
==========

to

==========
Oilpan: account for DevToolsEmulator reference from WebSettingsImpl.

WebSettingsImpl and DevToolsEmulator are both owned by WebViewImpl;
let the former keep an untracked reference to the latter.

R=haraken
BUG=553613
==========

[sof, 2015-11-16 15:34:30]

The CQ bit was checked by sigbjornf@opera.com

[commit-bot: I haz the power, 2015-11-16 15:35:15]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1437003005/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1437003005/1

[commit-bot: I haz the power, 2015-11-16 17:02:57]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-11-16 17:03:43]

Patchset 1 (id:??) landed as
https://crrev.com/05f9d3fd9ce5aae3d34a57af739a58297221c946
Cr-Commit-Position: refs/heads/master@{#359854}