Crash fix: Remove a NOTREACHED that is wrong.

Crash fix: Remove a NOTREACHED that is wrong.

I removed the return in a previous cleanup which allowed us to get data
on the fact that this is indeed reached in the wild.

R=mattm
BUG=105767
TEST=none

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=112718

[mattm, 2011-11-30 23:15:28]

http://codereview.chromium.org/8741019/diff/1/chrome/browser/ui/login/login_p...
File chrome/browser/ui/login/login_prompt.cc (right):

http://codereview.chromium.org/8741019/diff/1/chrome/browser/ui/login/login_p...
chrome/browser/ui/login/login_prompt.cc:452: return;
So if we think this is an expected behavior, should it also have a CancelAuth
like above?

[James Hawkins, 2011-12-01 21:37:15]

http://codereview.chromium.org/8741019/diff/1/chrome/browser/ui/login/login_p...
File chrome/browser/ui/login/login_prompt.cc (right):

http://codereview.chromium.org/8741019/diff/1/chrome/browser/ui/login/login_p...
chrome/browser/ui/login/login_prompt.cc:452: return;
On 2011/11/30 23:15:28, mattm wrote:
> So if we think this is an expected behavior, should it also have a CancelAuth
> like above?

Done.

[asanka, 2011-12-01 21:46:39]

http://codereview.chromium.org/8741019/diff/5001/chrome/browser/ui/login/logi...
File chrome/browser/ui/login/login_prompt.cc (right):

http://codereview.chromium.org/8741019/diff/5001/chrome/browser/ui/login/logi...
chrome/browser/ui/login/login_prompt.cc:450: if (!wrapper) {
I'm curious how this happens in practice.  Can you add a comment explaining why?

[James Hawkins, 2011-12-01 21:48:22]

http://codereview.chromium.org/8741019/diff/5001/chrome/browser/ui/login/logi...
File chrome/browser/ui/login/login_prompt.cc (right):

http://codereview.chromium.org/8741019/diff/5001/chrome/browser/ui/login/logi...
chrome/browser/ui/login/login_prompt.cc:450: if (!wrapper) {
On 2011/12/01 21:46:39, asanka wrote:
> I'm curious how this happens in practice.  Can you add a comment explaining
why?

I don't know how it happens in the wild; we've gotten crash reports for it
though.

[mattm, 2011-12-01 21:49:45]

On 2011/12/01 21:48:22, James Hawkins wrote:
>
http://codereview.chromium.org/8741019/diff/5001/chrome/browser/ui/login/logi...
> File chrome/browser/ui/login/login_prompt.cc (right):
> 
>
http://codereview.chromium.org/8741019/diff/5001/chrome/browser/ui/login/logi...
> chrome/browser/ui/login/login_prompt.cc:450: if (!wrapper) {
> On 2011/12/01 21:46:39, asanka wrote:
> > I'm curious how this happens in practice.  Can you add a comment explaining
> why?
> 
> I don't know how it happens in the wild; we've gotten crash reports for it
> though.

There's a very similar crash in the SSL client cert selector that just came up
recently too, that was caused by background tabs from extensions.
http://code.google.com/p/chromium/issues/detail?id=104757 (comment 15)

[James Hawkins, 2011-12-02 02:12:02]

Do we need anything else here?

[mattm, 2011-12-02 04:26:22]

On 2011/12/02 02:12:02, James Hawkins wrote:
> Do we need anything else here?

I guess if you've already logged into the site, the cached login values will be
used and this code doesn't even get run?  I guess there isn't really anything
else to be done.

lgtm