Fix imbalanced keepalive count for event pages when dealing with other views.

Fix imbalanced keepalive count for event pages when dealing with other views.

BUG=126036
TEST=see bug for repro


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=135098

[Yoyo Zhou, 2012-05-03 00:10:25]

http://codereview.chromium.org/10310005/diff/2001/chrome/browser/extensions/e...
File chrome/browser/extensions/extension_process_manager.cc (right):

http://codereview.chromium.org/10310005/diff/2001/chrome/browser/extensions/e...
chrome/browser/extensions/extension_process_manager.cc:652: // Re-register all
RenderViews for this extension. We do this to restore
Do we need to do this, or could we just not erase the background_page_data_ for
this extension if it had views open?

[Yoyo Zhou, 2012-05-03 00:10:33]

LGTM

[Matt Perry, 2012-05-03 00:18:38]

http://codereview.chromium.org/10310005/diff/2001/chrome/browser/extensions/e...
File chrome/browser/extensions/extension_process_manager.cc (right):

http://codereview.chromium.org/10310005/diff/2001/chrome/browser/extensions/e...
chrome/browser/extensions/extension_process_manager.cc:652: // Re-register all
RenderViews for this extension. We do this to restore
On 2012/05/03 00:10:25, Yoyo Zhou wrote:
> Do we need to do this, or could we just not erase the background_page_data_
for
> this extension if it had views open?

I think we still need to erase background_page_data. The reason it's erased is
that some of the Decrement signals come from the background page itself. When
the page unloads, it doesn't always get a chance to send the Decrement messages
(or they are ignored when the ExtensionHost is gone), so we reset it to 0 here.

[commit-bot: I haz the power, 2012-05-03 01:45:49]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/mpcomplete@chromium.org/10310005/2001

[commit-bot: I haz the power, 2012-05-03 05:30:20]

Change committed as 135098