PluginLoader loads plugins again after last load completes

PluginLoader loads plugins again after last load completes

If the list of plugins has updated before PluginLoader gets a
request to load plugins it occasionally fails to load the entire
list. This happens when PluginLoader is serving the last plugin
load request.

To make sure PluginLoader is up-to-date with the latest plugin
list all plugin load requests are served.

BUG=117561
TEST=content_unittests --gtest_filter=PluginLoader*
R=rsesek

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=126250

[James Hawkins, 2012-03-09 21:53:21]

I'm not sure I'm the right person to review this.

[Alpha Left Google, 2012-03-09 21:54:40]

Add jam@ as reviewer.

[Robert Sesek, 2012-03-09 22:08:48]

lgtm

[commit-bot: I haz the power, 2012-03-09 22:26:40]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/hclam@chromium.org/9665017/2001

[commit-bot: I haz the power, 2012-03-09 22:26:43]

Presubmit check for 9665017-2001 failed and returned exit status 1.

Running presubmit commit checks ...

** Presubmit ERRORS **
Missing LGTM from an OWNER for files in these directories:
    content/browser

Presubmit checks took 1.1s to calculate.

[jam, 2012-03-10 00:44:09]

lgtm

[commit-bot: I haz the power, 2012-03-10 00:48:32]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/hclam@chromium.org/9665017/2001

[commit-bot: I haz the power, 2012-03-10 03:06:09]

Try job failure for 9665017-2001 (retry) on win_rel for step "browser_tests".
It's a second try, previously, step "browser_tests" failed.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_rel&nu...

[commit-bot: I haz the power, 2012-03-12 18:16:29]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/hclam@chromium.org/9665017/2001

[commit-bot: I haz the power, 2012-03-12 22:51:12]

Change committed as 126250

[Bernhard Bauer, 2012-03-19 15:08:28]

Wait, doesn't this mean that we load plug-ins from disk once for each call to
PluginLoaderPosix::LoadPlugins? That seems wrong. The whole point of queuing
requests is that we only need to load the plug-in list once.