--- a/src/share/vm/runtime/thread.cpp	Tue Apr 07 11:20:51 2015 +0200
+++ b/src/share/vm/runtime/thread.cpp	Tue Apr 07 14:58:49 2015 +0200
@@ -62,6 +62,7 @@
 #include "runtime/memprofiler.hpp"
 #include "runtime/mutexLocker.hpp"
 #include "runtime/objectMonitor.hpp"
+#include "runtime/orderAccess.inline.hpp"
 #include "runtime/osThread.hpp"
 #include "runtime/safepoint.hpp"
 #include "runtime/sharedRuntime.hpp"
@@ -237,6 +238,8 @@
   // This initial value ==> never claimed.
   _oops_do_parity = 0;
 
+  _metadata_on_stack_buffer = NULL;
+
   // the handle mark links itself to last_handle_mark
   new HandleMark(this);
 
@@ -334,8 +337,7 @@
 #if INCLUDE_NMT
   // record thread's native stack, stack grows downward
   address stack_low_addr = stack_base() - stack_size();
-  MemTracker::record_thread_stack(stack_low_addr, stack_size(), this,
-      CURRENT_PC);
+  MemTracker::record_thread_stack(stack_low_addr, stack_size());
 #endif // INCLUDE_NMT
 }
 
@@ -353,7 +355,7 @@
 #if INCLUDE_NMT
   if (_stack_base != NULL) {
     address low_stack_addr = stack_base() - stack_size();
-    MemTracker::release_thread_stack(low_stack_addr, stack_size(), this);
+    MemTracker::release_thread_stack(low_stack_addr, stack_size());
 #ifdef ASSERT
     set_stack_base(NULL);
 #endif
@@ -837,7 +839,7 @@
   return false;
 }
 
-void Thread::oops_do(OopClosure* f, CLDToOopClosure* cld_f, CodeBlobClosure* cf) {
+void Thread::oops_do(OopClosure* f, CLDClosure* cld_f, CodeBlobClosure* cf) {
   active_handles()->oops_do(f);
   // Do oop for ThreadShadow
   f->do_oop((oop*)&_pending_exception);
@@ -867,6 +869,7 @@
       st->print("os_prio=%d ", os_prio);
     }
     st->print("tid=" INTPTR_FORMAT " ", this);
+    ext().print_on(st);
     osthread()->print_on(st);
   }
   debug_only(if (WizardMode) print_owned_locks_on(st);)
@@ -946,7 +949,7 @@
               cur != VMOperationRequest_lock &&
               cur != VMOperationQueue_lock) ||
               cur->rank() == Mutex::special) {
-          warning("Thread holding lock at safepoint that vm can block on: %s", cur->name());
+          fatal(err_msg("Thread holding lock at safepoint that vm can block on: %s", cur->name()));
         }
       }
     }
@@ -1476,9 +1479,6 @@
   set_monitor_chunks(NULL);
   set_next(NULL);
   set_thread_state(_thread_new);
-#if INCLUDE_NMT
-  set_recorder(NULL);
-#endif
   _terminated = _not_terminated;
   _privileged_stack_top = NULL;
   _array_for_gc = NULL;
@@ -1541,6 +1541,7 @@
   _popframe_condition = popframe_inactive;
   _popframe_preserved_args = NULL;
   _popframe_preserved_args_size = 0;
+  _frames_to_pop_failed_realloc = 0;
 
   pd_initialize();
 }
@@ -1564,7 +1565,6 @@
     _jni_attach_state = _not_attaching_via_jni;
   }
   assert(deferred_card_mark().is_empty(), "Default MemRegion ctor");
-  _safepoint_visible = false;
 }
 
 bool JavaThread::reguard_stack(address cur_sp) {
@@ -1627,7 +1627,6 @@
   thr_type = entry_point == &compiler_thread_entry ? os::compiler_thread :
                                                      os::java_thread;
   os::create_thread(this, thr_type, stack_sz);
-  _safepoint_visible = false;
   // The _osthread may be NULL here because we ran out of memory (too many threads active).
   // We need to throw and OutOfMemoryError - however we cannot do this here because the caller
   // may hold a lock and all locks must be unlocked before throwing the exception (throwing
@@ -1645,13 +1644,6 @@
       tty->print_cr("terminate thread %p", this);
   }
 
-  // By now, this thread should already be invisible to safepoint,
-  // and its per-thread recorder also collected.
-  assert(!is_safepoint_visible(), "wrong state");
-#if INCLUDE_NMT
-  assert(get_recorder() == NULL, "Already collected");
-#endif // INCLUDE_NMT
-
   // JSR166 -- return the parker to the free list
   Parker::Release(_parker);
   _parker = NULL ;
@@ -2780,7 +2772,7 @@
   }
 };
 
-void JavaThread::oops_do(OopClosure* f, CLDToOopClosure* cld_f, CodeBlobClosure* cf) {
+void JavaThread::oops_do(OopClosure* f, CLDClosure* cld_f, CodeBlobClosure* cf) {
   // Verify that the deferred card marks have been flushed.
   assert(deferred_card_mark().is_empty(), "Should be empty during GC");
 
@@ -3091,6 +3083,8 @@
   // Push the Java priority down to the native thread; needs Threads_lock
   Thread::set_priority(this, prio);
 
+  prepare_ext();
+
   // Add the new thread to the Threads list and set it in motion.
   // We must have threads lock in order to call Threads::add.
   // It is crucial that we do not block before the thread is
@@ -3316,7 +3310,7 @@
 }
 #endif
 
-void CompilerThread::oops_do(OopClosure* f, CLDToOopClosure* cld_f, CodeBlobClosure* cf) {
+void CompilerThread::oops_do(OopClosure* f, CLDClosure* cld_f, CodeBlobClosure* cf) {
   JavaThread::oops_do(f, cld_f, cf);
   if (_scanned_nmethod != NULL && cf != NULL) {
     // Safepoints can occur when the sweeper is scanning an nmethod so
@@ -3433,11 +3427,6 @@
   // intialize TLS
   ThreadLocalStorage::init();
 
-  // Bootstrap native memory tracking, so it can start recording memory
-  // activities before worker thread is started. This is the first phase
-  // of bootstrapping, VM is currently running in single-thread mode.
-  MemTracker::bootstrap_single_thread();
-
   // Initialize output stream logging
   ostream_init_log();
 
@@ -3497,9 +3486,6 @@
   // Initialize Java-Level synchronization subsystem
   ObjectMonitor::Initialize() ;
 
-  // Second phase of bootstrapping, VM is about entering multi-thread mode
-  MemTracker::bootstrap_multi_thread();
-
   // Initialize global modules
   jint status = init_globals();
   if (status != JNI_OK) {
@@ -3521,9 +3507,6 @@
   // real raw monitor. VM is setup enough here for raw monitor enter.
   JvmtiExport::transition_pending_onload_raw_monitors();
 
-  // Fully start NMT
-  MemTracker::start();
-
   // Create the VMThread
   { TraceTime timer("Start VMThread", TraceStartupTime);
     VMThread::create();
@@ -3979,6 +3962,24 @@
   }
 }
 
+JavaThread* Threads::find_java_thread_from_java_tid(jlong java_tid) {
+  assert(Threads_lock->owned_by_self(), "Must hold Threads_lock");
+
+  JavaThread* java_thread = NULL;
+  // Sequential search for now.  Need to do better optimization later.
+  for (JavaThread* thread = Threads::first(); thread != NULL; thread = thread->next()) {
+    oop tobj = thread->threadObj();
+    if (!thread->is_exiting() &&
+        tobj != NULL &&
+        java_tid == java_lang_Thread::thread_id(tobj)) {
+      java_thread = thread;
+      break;
+    }
+  }
+  return java_thread;
+}
+
+
 // Last thread running calls java.lang.Shutdown.shutdown()
 void JavaThread::invoke_shutdown_hooks() {
   HandleMark hm(this);
@@ -4175,8 +4176,6 @@
     daemon = false;
   }
 
-  p->set_safepoint_visible(true);
-
   ThreadService::add_thread(p, daemon);
 
   // Possible GC point.
@@ -4222,13 +4221,6 @@
     // to do callbacks into the safepoint code. However, the safepoint code is not aware
     // of this thread since it is removed from the queue.
     p->set_terminated_value();
-
-    // Now, this thread is not visible to safepoint
-    p->set_safepoint_visible(false);
-    // once the thread becomes safepoint invisible, we can not use its per-thread
-    // recorder. And Threads::do_threads() no longer walks this thread, so we have
-    // to release its per-thread recorder here.
-    MemTracker::thread_exiting(p);
   } // unlock Threads_lock
 
   // Since Events::log uses a lock, we grab it outside the Threads_lock
@@ -4253,22 +4245,22 @@
 // uses the Threads_lock to gurantee this property. It also makes sure that
 // all threads gets blocked when exiting or starting).
 
-void Threads::oops_do(OopClosure* f, CLDToOopClosure* cld_f, CodeBlobClosure* cf) {
+void Threads::oops_do(OopClosure* f, CLDClosure* cld_f, CodeBlobClosure* cf) {
   ALL_JAVA_THREADS(p) {
     p->oops_do(f, cld_f, cf);
   }
   VMThread::vm_thread()->oops_do(f, cld_f, cf);
 }
 
-void Threads::possibly_parallel_oops_do(OopClosure* f, CLDToOopClosure* cld_f, CodeBlobClosure* cf) {
+void Threads::possibly_parallel_oops_do(OopClosure* f, CLDClosure* cld_f, CodeBlobClosure* cf) {
   // Introduce a mechanism allowing parallel threads to claim threads as
   // root groups.  Overhead should be small enough to use all the time,
   // even in sequential code.
   SharedHeap* sh = SharedHeap::heap();
   // Cannot yet substitute active_workers for n_par_threads
   // because of G1CollectedHeap::verify() use of
-  // SharedHeap::process_strong_roots().  n_par_threads == 0 will
-  // turn off parallelism in process_strong_roots while active_workers
+  // SharedHeap::process_roots().  n_par_threads == 0 will
+  // turn off parallelism in process_roots while active_workers
   // is being used for parallelism elsewhere.
   bool is_par = sh->n_par_threads() > 0;
   assert(!is_par ||