Fix occasional deadlock caused by checkpoint root marking.

There was a race where a new worker thread would attach during the
checkpoint. This caused the thread to wait since suspend count != 0.
But when we decremented the suspend count, we did not broadcast to
the resume condition.

Added a create peer parameter to Thread::Attach and
AttachCurrentThread. This is used by the threadpool since we don't
need a java peer for worker threads.

Change-Id: I632926b5a6b52eeb0684b6e1dcbf3db42ba3d35c
diff --git a/src/runtime.h b/src/runtime.h
index 02f5206..94ab0c4 100644
--- a/src/runtime.h
+++ b/src/runtime.h
@@ -163,7 +163,8 @@
   jobject GetSystemThreadGroup() const;
 
   // Attaches the calling native thread to the runtime.
-  bool AttachCurrentThread(const char* thread_name, bool as_daemon, jobject thread_group);
+  bool AttachCurrentThread(const char* thread_name, bool as_daemon, jobject thread_group,
+                           bool create_peer);
 
   void CallExitHook(jint status);