[mojo-edk] Add support for transferring mach ports.

content/browser/renderer_host/render_process_host_impl.cc

 // Copyright 2012 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 // Represents the browser side of the browser <--> renderer communication
 // channel. There will be one RenderProcessHost per renderer process.
 
 #include "content/browser/renderer_host/render_process_host_impl.h"
 
 #include <algorithm>
@@ skipping 605 matching lines @@
   // render process. This ensures that when a test is being run in one of the
   // single process modes, the global attachment broker is the privileged
   // attachment broker, rather than an unprivileged attachment broker.
 #if defined(OS_MACOSX) && !defined(OS_IOS)
   IPC::AttachmentBrokerPrivileged::CreateBrokerIfNeeded(
       MachBroker::GetInstance());
 #else
   IPC::AttachmentBrokerPrivileged::CreateBrokerIfNeeded();
 #endif  // defined(OS_MACOSX) && !defined(OS_IOS)
 #endif  // USE_ATTACHMENT_BROKER
+#if defined(OS_MACOSX) && !defined(OS_IOS)
+  mojo::edk::SetMachPortProviderIfNeeded(MachBroker::GetInstance());

[Ken Rockot(use gerrit already), 2016/03/09 08:34:31]

I'm confused about why this (and really the AttachmentBrokerPrivileged above) is
created here instead of somewhere a bit more generic. If this is a
process-global thing, why is RPHI responsible for ensuring it exists, and why
does every RPHI instance do it? Why not BrowserMainLoop or some other very-early
initialization?

[erikchen, 2016/03/09 19:13:16]

That would be cleaner.

Right now, we actually call this same logic in 3 different places (you'll have
to do the same for mojo::edk as well)
https://code.google.com/p/chromium/codesearch#search/&q=brokerprivileged::Cre...

The idea was that we should attempt to lazy initialize, and only perform this
setup when we need it (which is when the browser tries to make a child process).

[Anand Mistry (off Chromium), 2016/03/11 07:44:17]

I agree, but think it's outside the scope of this CL.
I've added it to the one other place. The other (ChildProcessHostImpl) doesn't
actually create the broker on mac. Maybe that's a bug?

+#endif  // defined(OS_MACOSX) && !defined(OS_IOS)
 }
 
 // static
 void RenderProcessHostImpl::ShutDownInProcessRenderer() {
   DCHECK(g_run_renderer_in_process_);
 
   switch (g_all_hosts.Pointer()->size()) {
     case 0:
       return;
     case 1: {
@@ skipping 2173 matching lines @@
 
     // Skip widgets in other processes.
     if (rvh->GetProcess()->GetID() != GetID())
       continue;
 
     rvh->OnWebkitPreferencesChanged();
   }
 }
 
 }  // namespace content