PlzNavigate: Speculatively spawns a renderer process for navigations.

content/browser/frame_host/render_frame_host_impl.cc

Index: content/browser/frame_host/render_frame_host_impl.cc
diff --git a/content/browser/frame_host/render_frame_host_impl.cc b/content/browser/frame_host/render_frame_host_impl.cc
index af0c6fcc9a43af5997fc99815ca11e7f44696bee..f875406b2127fb19be04e2356ceab52041728de0 100644
--- a/content/browser/frame_host/render_frame_host_impl.cc
+++ b/content/browser/frame_host/render_frame_host_impl.cc
@@ -1401,6 +1401,10 @@ void RenderFrameHostImpl::CommitNavigation(
   // TODO(clamy): Check if we have to add security checks for the browser plugin
   // guests.
 
+  // Get back to a clean state, in case we start a new navigation without
+  // completing a RFH swap or unload handler.

[nasko, 2015/01/06 00:03:17]

Can you give an example of what this case is? I can see only one legal state
transition - STATE_SWAPPED_OUT to STATE_DEFAULT. If this is the case, we should
CHECK the expectation here.

[carlosk, 2015/01/08 16:05:55]

clamy@ just got me through one: if a still-swapping-out RFH (with a slow unload
handling) is swapped back in before it finishes unloading it will not be in the
SWAPPED_OUT state but in the PENDING_SWAP_OUT instead. In fact this is the case
being tested in
NavigatorTestWithBrowserSideNavigation::SpeculativeRendererReuseSwappedOutRFH.

[nasko, 2015/01/08 23:36:37]

Acknowledged.

+  SetState(RenderFrameHostImpl::STATE_DEFAULT);
+
   Send(new FrameMsg_CommitNavigation(
       routing_id_, response->head, body->GetURL(),
       common_params, commit_params));