Check for NULL frame tree root.

Check for NULL frame tree root.

BUG=240295
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=204096

[nasko, 2013-06-03 23:40:47]

Charlie,
I think this might be useful DCHECK to add to help understand why we have a case
of NULL frame tree root node.
What do you think?
Nasko

[Charlie Reis, 2013-06-04 00:00:55]

I'm not sure I see how this would help diagnose the problem.  Maybe we can chat
more about it tomorrow.

https://codereview.chromium.org/16325021/diff/1/content/browser/web_contents/...
File content/browser/web_contents/web_contents_impl.cc (right):

https://codereview.chromium.org/16325021/diff/1/content/browser/web_contents/...
content/browser/web_contents/web_contents_impl.cc:2489:
DCHECK(frame_tree_root_.get()) <<
DCHECKs only get hit during debug builds, so I don't expect many people are
going to hit this.

https://codereview.chromium.org/16325021/diff/1/content/browser/web_contents/...
content/browser/web_contents/web_contents_impl.cc:2490: GetLastCommittedURL() <<
"|" << GetActiveURL();
These wouldn't show up in a crash dump, so at best we would ask for someone to
paste them into a bug report from the console.

I'm also not sure the URL tells us much.  The URL seems orthogonal to how we get
into a state with a null root-- wouldn't it be more about timing of how the tab
is opened and closed?

[nasko, 2013-06-04 17:48:24]

As we chatted, I'm just doing a check to ensure we don't crash, since this logic
will change in the near future.

[Charlie Reis, 2013-06-04 17:51:56]

LGTM, but please update the CL title and description before you submit it. 
Thanks!

[commit-bot: I haz the power, 2013-06-04 18:14:51]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/nasko@chromium.org/16325021/5001

[commit-bot: I haz the power, 2013-06-05 00:16:02]

Change committed as 204096