Fix early prerender cancellation not working with PlzNavigate.

Fix early prerender cancellation not working with PlzNavigate.

With PlzNavigate the navigation starts in the browser, so LaunchAppWithUrl cancels the prerender before we get the PrerenderHostMsg_AddLinkRelPrerender IPC from the renderer. Fix this by noticing that the prerender was already cancelled when the IPC arrives.

This fixes
PlatformAppUrlRedirectorBrowserTest.PrerenderedClickInTabIntercepted
with PlzNavigate.

BUG=504347
Committed: https://crrev.com/e54c55a96737caf4a3c435dac442fe77122ff33e
Cr-Commit-Position: refs/heads/master@{#437633}

[jam, 2016-12-08 22:04:27]

The CQ bit was checked by jam@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-12-08 22:05:06]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[jam, 2016-12-08 22:59:56]

Description was changed from

==========
foo

BUG=
==========

to

==========
Fix early prerender cancellation not working with PlzNavigate.

With PlzNavigate the navigation starts in the browser, so LaunchAppWithUrl
cancels the prerender before we get the PrerenderHostMsg_AddLinkRelPrerender IPC
from the renderer. Fix this by noticing that the prerender was already cancelled
when the IPC arrives.

This fixes
PlatformAppUrlRedirectorBrowserTest.PrerenderedClickInTabIntercepted
with PlzNavigate.

BUG=504347
==========

[jam, 2016-12-08 22:59:56]

jam@chromium.org changed reviewers:
+ davidben@chromium.org

[commit-bot: I haz the power, 2016-12-08 23:06:12]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-12-08 23:06:13]

Dry run: This issue passed the CQ dry run.

[davidben, 2016-12-09 19:26:10]

lgtm. +pasko FYI.

Right, this is the ridiculous thing where prerenders are destroyed a PostTask
later, isn't it?

[jam, 2016-12-09 19:31:04]

The CQ bit was checked by jam@chromium.org

[commit-bot: I haz the power, 2016-12-09 19:32:49]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-12-09 20:41:58]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1481311864834790, "parent_rev":
"1cd75c4e87e26a88e09f011df9ce61b78a8c3305", "commit_rev":
"2acb702891c534846b7013e44b0f873a508a366c"}

[commit-bot: I haz the power, 2016-12-09 20:42:23]

Description was changed from

==========
Fix early prerender cancellation not working with PlzNavigate.

With PlzNavigate the navigation starts in the browser, so LaunchAppWithUrl
cancels the prerender before we get the PrerenderHostMsg_AddLinkRelPrerender IPC
from the renderer. Fix this by noticing that the prerender was already cancelled
when the IPC arrives.

This fixes
PlatformAppUrlRedirectorBrowserTest.PrerenderedClickInTabIntercepted
with PlzNavigate.

BUG=504347
==========

to

==========
Fix early prerender cancellation not working with PlzNavigate.

With PlzNavigate the navigation starts in the browser, so LaunchAppWithUrl
cancels the prerender before we get the PrerenderHostMsg_AddLinkRelPrerender IPC
from the renderer. Fix this by noticing that the prerender was already cancelled
when the IPC arrives.

This fixes
PlatformAppUrlRedirectorBrowserTest.PrerenderedClickInTabIntercepted
with PlzNavigate.

BUG=504347

Review-Url: https://codereview.chromium.org/2554573012
==========

[commit-bot: I haz the power, 2016-12-09 20:42:25]

Committed patchset #1 (id:1)

[pasko, 2016-12-12 13:46:30]

lgtm

On 2016/12/09 19:26:10, davidben wrote:
> lgtm. +pasko FYI.
> 
> Right, this is the ridiculous thing where prerenders are destroyed a PostTask
> later, isn't it?

Yes, the multi-stage process of killing prerenders is not fully transparent to
me. I guess it is remnants of trying to keep prerenders for longer in various
old experiments.

[commit-bot: I haz the power, 2016-12-12 14:40:09]

Description was changed from

==========
Fix early prerender cancellation not working with PlzNavigate.

With PlzNavigate the navigation starts in the browser, so LaunchAppWithUrl
cancels the prerender before we get the PrerenderHostMsg_AddLinkRelPrerender IPC
from the renderer. Fix this by noticing that the prerender was already cancelled
when the IPC arrives.

This fixes
PlatformAppUrlRedirectorBrowserTest.PrerenderedClickInTabIntercepted
with PlzNavigate.

BUG=504347

Review-Url: https://codereview.chromium.org/2554573012
==========

to

==========
Fix early prerender cancellation not working with PlzNavigate.

With PlzNavigate the navigation starts in the browser, so LaunchAppWithUrl
cancels the prerender before we get the PrerenderHostMsg_AddLinkRelPrerender IPC
from the renderer. Fix this by noticing that the prerender was already cancelled
when the IPC arrives.

This fixes
PlatformAppUrlRedirectorBrowserTest.PrerenderedClickInTabIntercepted
with PlzNavigate.

BUG=504347

Committed: https://crrev.com/e54c55a96737caf4a3c435dac442fe77122ff33e
Cr-Commit-Position: refs/heads/master@{#437633}
==========

[commit-bot: I haz the power, 2016-12-12 14:40:10]

Patchset 1 (id:??) landed as
https://crrev.com/e54c55a96737caf4a3c435dac442fe77122ff33e
Cr-Commit-Position: refs/heads/master@{#437633}

[davidben, 2016-12-12 16:45:28]

On 2016/12/12 13:46:30, pasko wrote:
> lgtm
> 
> On 2016/12/09 19:26:10, davidben wrote:
> > lgtm. +pasko FYI.
> > 
> > Right, this is the ridiculous thing where prerenders are destroyed a
PostTask
> > later, isn't it?
> 
> Yes, the multi-stage process of killing prerenders is not fully transparent to
> me. I guess it is remnants of trying to keep prerenders for longer in various
> old experiments.

I think it's also to hack around some reentrancy bugs. I hope the plan is to
still remove all of this code?