Revert 106677 (caused several PPAPI test timeouts, see http://crbug.com/101154)

Revert 106677 (caused several PPAPI test timeouts, see http://crbug.com/101154)
- Remove the proxy callback tracker.

This doesn't properly delete callbacks when the corresponding resource goes
away. This can lead to leaks or crashes in the plugin when the callback is
triggered unexpectedly.

BUG=http://crbug.com/86279

Review URL: http://codereview.chromium.org/8226009

TBR=brettw@chromium.org
Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=106717

[Nico, 2011-10-21 16:16:39]

It looks like the windows test greened up after FileIO was disabled on all
platforms – so this revert can be reverted once the tree reopens.

[brettw, 2011-10-21 16:45:14]

On Fri, Oct 21, 2011 at 9:16 AM,  <thakis@chromium.org> wrote:
> It looks like the windows test greened up after FileIO was disabled on all
> platforms – so this revert can be reverted once the tree reopens.
>
> http://codereview.chromium.org/8364040/

So FileIO was disabled for everything? I'm not sure I follow the timeline here.

Brett

[Nico, 2011-10-21 16:46:22]

On Fri, Oct 21, 2011 at 9:45 AM, Brett Wilson <brettw@chromium.org> wrote:
> On Fri, Oct 21, 2011 at 9:16 AM,  <thakis@chromium.org> wrote:
>> It looks like the windows test greened up after FileIO was disabled on all
>> platforms – so this revert can be reverted once the tree reopens.
>>
>> http://codereview.chromium.org/8364040/
>
> So FileIO was disabled for everything? I'm not sure I follow the timeline
here.

Yes: http://src.chromium.org/viewvc/chrome?view=rev&revision=106716

[brettw, 2011-10-21 16:50:15]

On Fri, Oct 21, 2011 at 9:46 AM, Nico Weber <thakis@chromium.org> wrote:
> On Fri, Oct 21, 2011 at 9:45 AM, Brett Wilson <brettw@chromium.org> wrote:
>> On Fri, Oct 21, 2011 at 9:16 AM,  <thakis@chromium.org> wrote:
>>> It looks like the windows test greened up after FileIO was disabled on all
>>> platforms – so this revert can be reverted once the tree reopens.
>>>
>>> http://codereview.chromium.org/8364040/
>>
>> So FileIO was disabled for everything? I'm not sure I follow the timeline
here.
>
> Yes: http://src.chromium.org/viewvc/chrome?view=rev&revision=106716

It looks like I checked in and according to your comment, the test
started failing when I checked in. Then my patch was reverted. Now it
seems like you're saying it's OK to reland since the test is disabled.
Am I missing something?

Brett

[Nico, 2011-10-21 17:01:47]

On Fri, Oct 21, 2011 at 9:50 AM, Brett Wilson <brettw@chromium.org> wrote:
> On Fri, Oct 21, 2011 at 9:46 AM, Nico Weber <thakis@chromium.org> wrote:
>> On Fri, Oct 21, 2011 at 9:45 AM, Brett Wilson <brettw@chromium.org> wrote:
>>> On Fri, Oct 21, 2011 at 9:16 AM,  <thakis@chromium.org> wrote:
>>>> It looks like the windows test greened up after FileIO was disabled on all
>>>> platforms – so this revert can be reverted once the tree reopens.
>>>>
>>>> http://codereview.chromium.org/8364040/
>>>
>>> So FileIO was disabled for everything? I'm not sure I follow the timeline
here.
>>
>> Yes: http://src.chromium.org/viewvc/chrome?view=rev&revision=106716
>
> It looks like I checked in and according to your comment, the test
> started failing when I checked in. Then my patch was reverted. Now it
> seems like you're saying it's OK to reland since the test is disabled.
> Am I missing something?

The test was marked as FAILS_, and with your commit it started timing
out instead.

I thought that your commit impacted more than this one test.

But yes, instead of checking this CL back in you could also flip the
test back to FAILS_ and then investigate why your change made it time
out.

[brettw, 2011-10-21 18:07:48]

On Fri, Oct 21, 2011 at 9:55 AM, Nico Weber <thakis@chromium.org> wrote:
> On Fri, Oct 21, 2011 at 9:50 AM, Brett Wilson <brettw@chromium.org> wrote:
>> On Fri, Oct 21, 2011 at 9:46 AM, Nico Weber <thakis@chromium.org> wrote:
>>> On Fri, Oct 21, 2011 at 9:45 AM, Brett Wilson <brettw@chromium.org> wrote:
>>>> On Fri, Oct 21, 2011 at 9:16 AM,  <thakis@chromium.org> wrote:
>>>>> It looks like the windows test greened up after FileIO was disabled on all
>>>>> platforms – so this revert can be reverted once the tree reopens.
>>>>>
>>>>> http://codereview.chromium.org/8364040/
>>>>
>>>> So FileIO was disabled for everything? I'm not sure I follow the timeline
here.
>>>
>>> Yes: http://src.chromium.org/viewvc/chrome?view=rev&revision=106716
>>
>> It looks like I checked in and according to your comment, the test
>> started failing when I checked in. Then my patch was reverted. Now it
>> seems like you're saying it's OK to reland since the test is disabled.
>> Am I missing something?
>
> The test was marked as FAILS_, and with your commit it started timing
> out instead.

Okay, that's the key thing I missed. Thanks.

> I thought that your commit impacted more than this one test.
>
> But yes, instead of checking this CL back in you could also flip the
> test back to FAILS_ and then investigate why your change made it time
> out.

Brett