PlzNavigate: Speculatively spawns a renderer process for navigations.

content/browser/frame_host/navigator_impl.cc

Index: content/browser/frame_host/navigator_impl.cc
diff --git a/content/browser/frame_host/navigator_impl.cc b/content/browser/frame_host/navigator_impl.cc
index e3b56215d4d26daf3218f1b52f0133bda53ffdc0..37aa8396dd98d0a8f09ace67bf5945ddbb539c1d 100644
--- a/content/browser/frame_host/navigator_impl.cc
+++ b/content/browser/frame_host/navigator_impl.cc
@@ -738,8 +738,7 @@ void NavigatorImpl::OnBeginNavigation(
   info->parent_is_main_frame = !frame_tree_node->parent() ?
       false : frame_tree_node->parent()->IsMainFrame();
 
-  // TODO(clamy): Inform the RenderFrameHostManager that a navigation is about
-  // to begin, so that it can speculatively spawn a new renderer if needed.
+  frame_tree_node->render_manager()->BeginNavigation(params, common_params);

[clamy, 2014/11/12 13:01:13]

Following the performance issues due to spawning a new RFH, shouldn't this take
place after the call to navigation_request->BeginNavigation?

[carlosk, 2014/11/13 14:40:25]

Yes! I had this already done locally once I realized work done here is message
driven so there will be no way for this thread to handle the response from the
IO thread before this call is done.

 
   navigation_request->BeginNavigation(info.Pass(), params.request_body);
 }
@@ -782,6 +781,7 @@ void NavigatorImpl::CancelNavigation(FrameTreeNode* frame_tree_node) {
   CHECK(CommandLine::ForCurrentProcess()->HasSwitch(
       switches::kEnableBrowserSideNavigation));
   navigation_request_map_.erase(frame_tree_node->frame_tree_node_id());
+  frame_tree_node->render_manager()->CancelNavigation();
 }
 
 void NavigatorImpl::LogResourceRequestTime(