Set correct viewport size for an out of process iframe.

Set correct viewport size for an out of process iframe.

This patch adds an IPC for iframe parameter initialization when it has
been swapped out for out of process rendering.

It changes where CrossProcessFrameConnector is being created because
it was being created too late; it is much more useful for it to be able
to receive messages from the swapped out frame as soon as that frame is
swapped out.

There is currently no mechanism for dynamically changing the size of an
iframe. It is not yet clear how this will be implemented in Blink.

Dependency on Blink CL: https://codereview.chromium.org/135163006/

R=creis@chromium.org, jam@chromium.org
BUG=325803
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=251986

[kenrb, 2014-02-04 18:35:56]

Charlie for review of the RenderFrameHostManager change and the swap out flow in
general.
John for review of the remainder.

[Charlie Reis, 2014-02-04 20:11:12]

Yes, I think this makes sense.  Just a few minor questions below.

https://codereview.chromium.org/132383005/diff/1/content/browser/frame_host/r...
File content/browser/frame_host/render_frame_host_manager.cc (right):

https://codereview.chromium.org/132383005/diff/1/content/browser/frame_host/r...
content/browser/frame_host/render_frame_host_manager.cc:532: // frame in its
parent's process.
I think there's something more to explain in this comment, since it's a bit
subtle: if we do a second cross-process navigation in the subframe, we don't
reset the connector.  Only the first cross-process subframe navigation needs the
connector, because that's when it leaves the parent's process.  (Hence the
conditional on line 534.)

https://codereview.chromium.org/132383005/diff/1/content/browser/frame_host/r...
content/browser/frame_host/render_frame_host_manager.cc:534: if
(!cross_process_frame_connector_)
Style nit: Needs braces.

https://codereview.chromium.org/132383005/diff/1/content/renderer/render_fram...
File content/renderer/render_frame_impl.cc (right):

https://codereview.chromium.org/132383005/diff/1/content/renderer/render_fram...
content/renderer/render_frame_impl.cc:441: frame_->setIsRemote(true);
Do we need to set this back to false when swapping it back in?  (Or is this
still only called for subframes and we don't yet support swapping them back in?)

https://codereview.chromium.org/132383005/diff/1/content/renderer/render_fram...
content/renderer/render_frame_impl.cc:1634: void
RenderFrameImpl::initializeChildFrame(const blink::WebRect& frame_rect,
Just to be sure I follow, you're expecting this to be called from your Blink CL,
right?

[kenrb, 2014-02-04 21:45:03]

https://codereview.chromium.org/132383005/diff/1/content/browser/frame_host/r...
File content/browser/frame_host/render_frame_host_manager.cc (right):

https://codereview.chromium.org/132383005/diff/1/content/browser/frame_host/r...
content/browser/frame_host/render_frame_host_manager.cc:532: // frame in its
parent's process.
On 2014/02/04 20:11:12, Charlie Reis wrote:
> I think there's something more to explain in this comment, since it's a bit
> subtle: if we do a second cross-process navigation in the subframe, we don't
> reset the connector.  Only the first cross-process subframe navigation needs
the
> connector, because that's when it leaves the parent's process.  (Hence the
> conditional on line 534.)

Done. Please check if the new comment makes sense to you.

https://codereview.chromium.org/132383005/diff/1/content/browser/frame_host/r...
content/browser/frame_host/render_frame_host_manager.cc:534: if
(!cross_process_frame_connector_)
On 2014/02/04 20:11:12, Charlie Reis wrote:
> Style nit: Needs braces.

Done.

https://codereview.chromium.org/132383005/diff/1/content/renderer/render_fram...
File content/renderer/render_frame_impl.cc (right):

https://codereview.chromium.org/132383005/diff/1/content/renderer/render_fram...
content/renderer/render_frame_impl.cc:441: frame_->setIsRemote(true);
On 2014/02/04 20:11:12, Charlie Reis wrote:
> Do we need to set this back to false when swapping it back in?  (Or is this
> still only called for subframes and we don't yet support swapping them back
in?)

The latter. We don't swap subframes back in yet.

https://codereview.chromium.org/132383005/diff/1/content/renderer/render_fram...
content/renderer/render_frame_impl.cc:1634: void
RenderFrameImpl::initializeChildFrame(const blink::WebRect& frame_rect,
On 2014/02/04 20:11:12, Charlie Reis wrote:
> Just to be sure I follow, you're expecting this to be called from your Blink
CL,
> right?

Yes it is implementing a virtual method from the WebFrameClient interface.

[Charlie Reis, 2014-02-04 21:52:41]

LGTM

https://codereview.chromium.org/132383005/diff/1/content/browser/frame_host/r...
File content/browser/frame_host/render_frame_host_manager.cc (right):

https://codereview.chromium.org/132383005/diff/1/content/browser/frame_host/r...
content/browser/frame_host/render_frame_host_manager.cc:532: // frame in its
parent's process.
On 2014/02/04 21:45:03, kenrb wrote:
> On 2014/02/04 20:11:12, Charlie Reis wrote:
> > I think there's something more to explain in this comment, since it's a bit
> > subtle: if we do a second cross-process navigation in the subframe, we don't
> > reset the connector.  Only the first cross-process subframe navigation needs
> the
> > connector, because that's when it leaves the parent's process.  (Hence the
> > conditional on line 534.)
> 
> Done. Please check if the new comment makes sense to you.

Thanks, that's much better.

[jam, 2014-02-04 23:31:18]

lgtm

[kenrb, 2014-02-14 20:07:16]

The CQ bit was checked by kenrb@chromium.org

[kenrb, 2014-02-14 20:07:20]

The CQ bit was unchecked by kenrb@chromium.org

[kenrb, 2014-02-14 20:07:25]

The CQ bit was checked by kenrb@chromium.org

[commit-bot: I haz the power, 2014-02-14 20:08:13]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kenrb@chromium.org/132383005/100001

[commit-bot: I haz the power, 2014-02-14 20:37:15]

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

[commit-bot: I haz the power, 2014-02-14 20:37:16]

Retried try job too often on mac_rel for step(s) base_unittests, browser_tests,
interactive_ui_tests, net_unittests, unit_tests
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=mac_rel&nu...

[kenrb, 2014-02-18 16:32:31]

The CQ bit was checked by kenrb@chromium.org

[commit-bot: I haz the power, 2014-02-18 16:33:55]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kenrb@chromium.org/132383005/100001

[commit-bot: I haz the power, 2014-02-18 16:34:10]

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

[commit-bot: I haz the power, 2014-02-18 16:34:10]

Failed to apply patch for content/renderer/render_view_impl.h:
While running patch -p1 --forward --force --no-backup-if-mismatch;
  patching file content/renderer/render_view_impl.h
  Hunk #1 FAILED at 807.
  1 out of 1 hunk FAILED -- saving rejects to file
content/renderer/render_view_impl.h.rej

Patch:       content/renderer/render_view_impl.h
Index: content/renderer/render_view_impl.h
diff --git a/content/renderer/render_view_impl.h
b/content/renderer/render_view_impl.h
index
479c526e04d0de723b2e358cf89efb4eeb88aa5e..d3477a24a868ac13641c38bda972b3fe013f8291
100644
--- a/content/renderer/render_view_impl.h
+++ b/content/renderer/render_view_impl.h
@@ -807,6 +807,9 @@ class CONTENT_EXPORT RenderViewImpl
     CONNECTION_ERROR,
   };
 
+  static void GetRedirectChain(blink::WebDataSource* ds,
+                               std::vector<GURL>* result);
+
   static blink::WebReferrerPolicy GetReferrerPolicyFromRequest(
       blink::WebFrame* frame,
       const blink::WebURLRequest& request);

[kenrb, 2014-02-18 17:29:20]

The CQ bit was checked by kenrb@chromium.org

[commit-bot: I haz the power, 2014-02-18 17:29:34]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kenrb@chromium.org/132383005/410001

[commit-bot: I haz the power, 2014-02-18 18:23:08]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kenrb@chromium.org/132383005/410001

[commit-bot: I haz the power, 2014-02-18 18:31:19]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kenrb@chromium.org/132383005/410001

[commit-bot: I haz the power, 2014-02-18 18:36:18]

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

[commit-bot: I haz the power, 2014-02-18 18:36:19]

Try jobs failed on following builders: linux_chromium_rel

[kenrb, 2014-02-18 19:04:02]

The CQ bit was checked by kenrb@chromium.org

[commit-bot: I haz the power, 2014-02-18 19:04:16]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/kenrb@chromium.org/132383005/680001

[commit-bot: I haz the power, 2014-02-19 08:02:18]

Change committed as 251986