[WebsiteSettings] Set view to null on manager to ensure no stale access.

[WebsiteSettings] Set view to null on manager to ensure no stale access.

R=leng@chromium.org, markusheintz@chromium.org
BUG=343234
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=251392

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=251919

[Greg Billock, 2014-02-13 04:39:07]

On 2014/02/13 04:36:44, Greg Billock wrote:

The crash is this:

Thread 0 CRASHED [EXCEPTION_ACCESS_VIOLATION_READ @ 18446744073550408666] MAGIC
SIGNATURE THREAD
0x62d2b887	[chrome.dll -bubble_delegate.cc:209
]	views::BubbleDelegateView::StartFade(bool)
0x61da57d0	[chrome.dll -permissions_bubble_view.cc:259
]	PermissionBubbleViewViews::Hide()
0x61da5357	[chrome.dll -permission_bubble_manager.cc:47
]	PermissionBubbleManager::SetView(PermissionBubbleView *)
0x628bef4c	[chrome.dll -browser_view.cc:1496
]	BrowserView::TabDeactivated(content::WebContents *)

Which looks like a stale access. No-one's figured out how to repro the bug, but
the theory is that there's some escape hatch which is clearing the view without
notifying the permission bubble manager. This should cure all such bugs and
futureproof the view/manager connection.

[leng, 2014-02-13 17:58:35]

lgtm

[markusheintz_, 2014-02-13 18:01:57]

LGTM

[Greg Billock, 2014-02-13 18:23:24]

The CQ bit was checked by gbillock@chromium.org

[commit-bot: I haz the power, 2014-02-13 18:26:00]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/gbillock@chromium.org/162713003/40001

[commit-bot: I haz the power, 2014-02-13 19:37:08]

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

[commit-bot: I haz the power, 2014-02-13 19:37:08]

Retried try job too often on android_clang_dbg for step(s) slave_steps
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=android_cl...

[Greg Billock, 2014-02-13 19:44:21]

The CQ bit was checked by gbillock@chromium.org

[Greg Billock, 2014-02-13 19:46:27]

The CQ bit was unchecked by gbillock@chromium.org

[Greg Billock, 2014-02-13 20:06:32]

The CQ bit was checked by gbillock@chromium.org

[commit-bot: I haz the power, 2014-02-13 20:10:14]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/gbillock@chromium.org/162713003/360001

[commit-bot: I haz the power, 2014-02-14 00:07:43]

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

[commit-bot: I haz the power, 2014-02-14 00:07:43]

Retried try job too often on linux_chromeos for step(s) browser_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=linux_chro...

[Greg Billock, 2014-02-14 18:40:20]

The CQ bit was checked by gbillock@chromium.org

[commit-bot: I haz the power, 2014-02-14 18:40:44]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/gbillock@chromium.org/162713003/360001

[commit-bot: I haz the power, 2014-02-14 19:33:00]

Change committed as 251392

[hubbe, 2014-02-14 20:17:56]

A revert of this CL has been created in
https://codereview.chromium.org/167613002/ by hubbe@chromium.org.

The reason for reverting is:
http://build.chromium.org/p/chromium.memory/builders/Linux%20ASAN%20Tests%20%....

[Greg Billock, 2014-02-18 18:24:21]

On 2014/02/14 20:17:56, hubbe wrote:
> A revert of this CL has been created in
> https://codereview.chromium.org/167613002/ by mailto:hubbe@chromium.org.
> 
> The reason for reverting is:
>
http://build.chromium.org/p/chromium.memory/builders/Linux%2520ASAN%2520Tests....

I was thinking the WebContents would outlive the bubble view, but this isn't
always the case. Need to connect up the other way as well.

[Greg Billock, 2014-02-18 19:21:15]

On 2014/02/18 18:24:21, Greg Billock wrote:
> On 2014/02/14 20:17:56, hubbe wrote:
> > A revert of this CL has been created in
> > https://codereview.chromium.org/167613002/ by mailto:hubbe@chromium.org.
> > 
> > The reason for reverting is:
> >
>
http://build.chromium.org/p/chromium.memory/builders/Linux%252520ASAN%252520T....
> 
> I was thinking the WebContents would outlive the bubble view, but this isn't
> always the case. Need to connect up the other way as well.

Markus, want another look at this?

[Greg Billock, 2014-02-18 23:07:37]

The CQ bit was checked by gbillock@chromium.org

[commit-bot: I haz the power, 2014-02-18 23:08:26]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/gbillock@chromium.org/162713003/800001

[commit-bot: I haz the power, 2014-02-19 02:05:12]

Change committed as 251919