diff --git a/src/os/bsd/vm/os_bsd.cpp b/src/os/bsd/vm/os_bsd.cpp
index 47e5d16efdc820b6ccd2d54fe7acfee044132b85..456daba559c0933483eece391897195793d0cef4 100644
--- a/src/os/bsd/vm/os_bsd.cpp
+++ b/src/os/bsd/vm/os_bsd.cpp
@@ -2636,9 +2636,21 @@ int os::sleep(Thread* thread, jlong millis, bool interruptible) {
   }
 }
 
-int os::naked_sleep() {
-  // %% make the sleep time an integer flag. for now use 1 millisec.
-  return os::sleep(Thread::current(), 1, false);
+void os::naked_short_sleep(jlong ms) {
+  struct timespec req;
+
+  assert(ms < 1000, "Un-interruptable sleep, short time use only");
+  req.tv_sec = 0;
+  if (ms > 0) {
+    req.tv_nsec = (ms % 1000) * 1000000;
+  }
+  else {
+    req.tv_nsec = 1;
+  }
+
+  nanosleep(&req, NULL);
+
+  return;
 }
 
 // Sleep forever; naked call to OS-specific sleep; use with CAUTION
diff --git a/src/os/linux/vm/os_linux.cpp b/src/os/linux/vm/os_linux.cpp
index 54149e1825331655c701a4dd98445c881b25e0af..d0751f005adf006ebb0d6ab08b3eb41b0480453e 100644
--- a/src/os/linux/vm/os_linux.cpp
+++ b/src/os/linux/vm/os_linux.cpp
@@ -3871,9 +3871,33 @@ int os::sleep(Thread* thread, jlong millis, bool interruptible) {
   }
 }
 
-int os::naked_sleep() {
-  // %% make the sleep time an integer flag. for now use 1 millisec.
-  return os::sleep(Thread::current(), 1, false);
+//
+// Short sleep, direct OS call.
+//
+// Note: certain versions of Linux CFS scheduler (since 2.6.23) do not guarantee
+// sched_yield(2) will actually give up the CPU:
+//
+//   * Alone on this pariticular CPU, keeps running.
+//   * Before the introduction of "skip_buddy" with "compat_yield" disabled
+//     (pre 2.6.39).
+//
+// So calling this with 0 is an alternative.
+//
+void os::naked_short_sleep(jlong ms) {
+  struct timespec req;
+
+  assert(ms < 1000, "Un-interruptable sleep, short time use only");
+  req.tv_sec = 0;
+  if (ms > 0) {
+    req.tv_nsec = (ms % 1000) * 1000000;
+  }
+  else {
+    req.tv_nsec = 1;
+  }
+
+  nanosleep(&req, NULL);
+
+  return;
 }
 
 // Sleep forever; naked call to OS-specific sleep; use with CAUTION
diff --git a/src/os/solaris/vm/os_solaris.cpp b/src/os/solaris/vm/os_solaris.cpp
index a4d0178c8874c7e34ede94b1c63ad55292319ab3..533bbb766509096fbb6c6342c48c78aadbdfb921 100644
--- a/src/os/solaris/vm/os_solaris.cpp
+++ b/src/os/solaris/vm/os_solaris.cpp
@@ -3540,9 +3540,14 @@ int os::sleep(Thread* thread, jlong millis, bool interruptible) {
   return os_sleep(millis, interruptible);
 }
 
-int os::naked_sleep() {
-  // %% make the sleep time an integer flag. for now use 1 millisec.
-  return os_sleep(1, false);
+void os::naked_short_sleep(jlong ms) {
+  assert(ms < 1000, "Un-interruptable sleep, short time use only");
+
+  // usleep is deprecated and removed from POSIX, in favour of nanosleep, but
+  // Solaris requires -lrt for this.
+  usleep((ms * 1000));
+
+  return;
 }
 
 // Sleep forever; naked call to OS-specific sleep; use with CAUTION
diff --git a/src/os/windows/vm/os_windows.cpp b/src/os/windows/vm/os_windows.cpp
index 7daee35632e366e657bc5a48648ad99d158d8d0c..4a540b66e982be361b8222ac630f91eeaf214b5a 100644
--- a/src/os/windows/vm/os_windows.cpp
+++ b/src/os/windows/vm/os_windows.cpp
@@ -3496,6 +3496,16 @@ int os::sleep(Thread* thread, jlong ms, bool interruptable) {
   return result;
 }
 
+//
+// Short sleep, direct OS call.
+//
+// ms = 0, means allow others (if any) to run.
+//
+void os::naked_short_sleep(jlong ms) {
+  assert(ms < 1000, "Un-interruptable sleep, short time use only");
+  Sleep(ms);
+}
+
 // Sleep forever; naked call to OS-specific sleep; use with CAUTION
 void os::infinite_sleep() {
   while (true) {    // sleep forever ...
diff --git a/src/share/vm/runtime/os.hpp b/src/share/vm/runtime/os.hpp
index 17fcd3bdd79540229f8ad20f438db14655dd064d..47ebeb6bbb5b9fbf8b3e17d24f0744cc012feb11 100644
--- a/src/share/vm/runtime/os.hpp
+++ b/src/share/vm/runtime/os.hpp
@@ -430,7 +430,10 @@ class os: AllStatic {
   static intx current_thread_id();
   static int current_process_id();
   static int sleep(Thread* thread, jlong ms, bool interruptable);
-  static int naked_sleep();
+  // Short standalone OS sleep suitable for slow path spin loop.
+  // Ignores Thread.interrupt() (so keep it short).
+  // ms = 0, will sleep for the least amount of time allowed by the OS.
+  static void naked_short_sleep(jlong ms);
   static void infinite_sleep(); // never returns, use with CAUTION
   static void yield();        // Yields to all threads with same priority
   enum YieldResult {
diff --git a/src/share/vm/runtime/park.cpp b/src/share/vm/runtime/park.cpp
index 6380570ef508b0e7057c0bab9d7bd3c3b68c7de2..0ab5b5b6786d76e999d2dec90ed62fe3b133d9af 100644
--- a/src/share/vm/runtime/park.cpp
+++ b/src/share/vm/runtime/park.cpp
@@ -59,58 +59,22 @@ ParkEvent * ParkEvent::Allocate (Thread * t) {
 
   // Start by trying to recycle an existing but unassociated
   // ParkEvent from the global free list.
-  for (;;) {
-    ev = FreeList ;
-    if (ev == NULL) break ;
-    // 1: Detach - sequester or privatize the list
-    // Tantamount to ev = Swap (&FreeList, NULL)
-    if (Atomic::cmpxchg_ptr (NULL, &FreeList, ev) != ev) {
-       continue ;
+  // Using a spin lock since we are part of the mutex impl.
+  // 8028280: using concurrent free list without memory management can leak
+  // pretty badly it turns out.
+  Thread::SpinAcquire(&ListLock, "ParkEventFreeListAllocate");
+  {
+    ev = FreeList;
+    if (ev != NULL) {
+      FreeList = ev->FreeNext;
     }
-
-    // We've detached the list.  The list in-hand is now
-    // local to this thread.   This thread can operate on the
-    // list without risk of interference from other threads.
-    // 2: Extract -- pop the 1st element from the list.
-    ParkEvent * List = ev->FreeNext ;
-    if (List == NULL) break ;
-    for (;;) {
-        // 3: Try to reattach the residual list
-        guarantee (List != NULL, "invariant") ;
-        ParkEvent * Arv =  (ParkEvent *) Atomic::cmpxchg_ptr (List, &FreeList, NULL) ;
-        if (Arv == NULL) break ;
-
-        // New nodes arrived.  Try to detach the recent arrivals.
-        if (Atomic::cmpxchg_ptr (NULL, &FreeList, Arv) != Arv) {
-            continue ;
-        }
-        guarantee (Arv != NULL, "invariant") ;
-        // 4: Merge Arv into List
-        ParkEvent * Tail = List ;
-        while (Tail->FreeNext != NULL) Tail = Tail->FreeNext ;
-        Tail->FreeNext = Arv ;
-    }
-    break ;
   }
+  Thread::SpinRelease(&ListLock);
 
   if (ev != NULL) {
     guarantee (ev->AssociatedWith == NULL, "invariant") ;
   } else {
     // Do this the hard way -- materialize a new ParkEvent.
-    // In rare cases an allocating thread might detach a long list --
-    // installing null into FreeList -- and then stall or be obstructed.
-    // A 2nd thread calling Allocate() would see FreeList == null.
-    // The list held privately by the 1st thread is unavailable to the 2nd thread.
-    // In that case the 2nd thread would have to materialize a new ParkEvent,
-    // even though free ParkEvents existed in the system.  In this case we end up
-    // with more ParkEvents in circulation than we need, but the race is
-    // rare and the outcome is benign.  Ideally, the # of extant ParkEvents
-    // is equal to the maximum # of threads that existed at any one time.
-    // Because of the race mentioned above, segments of the freelist
-    // can be transiently inaccessible.  At worst we may end up with the
-    // # of ParkEvents in circulation slightly above the ideal.
-    // Note that if we didn't have the TSM/immortal constraint, then
-    // when reattaching, above, we could trim the list.
     ev = new ParkEvent () ;
     guarantee ((intptr_t(ev) & 0xFF) == 0, "invariant") ;
   }
@@ -124,13 +88,14 @@ void ParkEvent::Release (ParkEvent * ev) {
   if (ev == NULL) return ;
   guarantee (ev->FreeNext == NULL      , "invariant") ;
   ev->AssociatedWith = NULL ;
-  for (;;) {
-    // Push ev onto FreeList
-    // The mechanism is "half" lock-free.
-    ParkEvent * List = FreeList ;
-    ev->FreeNext = List ;
-    if (Atomic::cmpxchg_ptr (ev, &FreeList, List) == List) break ;
+  // Note that if we didn't have the TSM/immortal constraint, then
+  // when reattaching we could trim the list.
+  Thread::SpinAcquire(&ListLock, "ParkEventFreeListRelease");
+  {
+    ev->FreeNext = FreeList;
+    FreeList = ev;
   }
+  Thread::SpinRelease(&ListLock);
 }
 
 // Override operator new and delete so we can ensure that the
@@ -164,56 +129,21 @@ Parker * Parker::Allocate (JavaThread * t) {
 
   // Start by trying to recycle an existing but unassociated
   // Parker from the global free list.
-  for (;;) {
-    p = FreeList ;
-    if (p  == NULL) break ;
-    // 1: Detach
-    // Tantamount to p = Swap (&FreeList, NULL)
-    if (Atomic::cmpxchg_ptr (NULL, &FreeList, p) != p) {
-       continue ;
+  // 8028280: using concurrent free list without memory management can leak
+  // pretty badly it turns out.
+  Thread::SpinAcquire(&ListLock, "ParkerFreeListAllocate");
+  {
+    p = FreeList;
+    if (p != NULL) {
+      FreeList = p->FreeNext;
     }
-
-    // We've detached the list.  The list in-hand is now
-    // local to this thread.   This thread can operate on the
-    // list without risk of interference from other threads.
-    // 2: Extract -- pop the 1st element from the list.
-    Parker * List = p->FreeNext ;
-    if (List == NULL) break ;
-    for (;;) {
-        // 3: Try to reattach the residual list
-        guarantee (List != NULL, "invariant") ;
-        Parker * Arv =  (Parker *) Atomic::cmpxchg_ptr (List, &FreeList, NULL) ;
-        if (Arv == NULL) break ;
-
-        // New nodes arrived.  Try to detach the recent arrivals.
-        if (Atomic::cmpxchg_ptr (NULL, &FreeList, Arv) != Arv) {
-            continue ;
-        }
-        guarantee (Arv != NULL, "invariant") ;
-        // 4: Merge Arv into List
-        Parker * Tail = List ;
-        while (Tail->FreeNext != NULL) Tail = Tail->FreeNext ;
-        Tail->FreeNext = Arv ;
-    }
-    break ;
   }
+  Thread::SpinRelease(&ListLock);
 
   if (p != NULL) {
     guarantee (p->AssociatedWith == NULL, "invariant") ;
   } else {
     // Do this the hard way -- materialize a new Parker..
-    // In rare cases an allocating thread might detach
-    // a long list -- installing null into FreeList --and
-    // then stall.  Another thread calling Allocate() would see
-    // FreeList == null and then invoke the ctor.  In this case we
-    // end up with more Parkers in circulation than we need, but
-    // the race is rare and the outcome is benign.
-    // Ideally, the # of extant Parkers is equal to the
-    // maximum # of threads that existed at any one time.
-    // Because of the race mentioned above, segments of the
-    // freelist can be transiently inaccessible.  At worst
-    // we may end up with the # of Parkers in circulation
-    // slightly above the ideal.
     p = new Parker() ;
   }
   p->AssociatedWith = t ;          // Associate p with t
@@ -227,11 +157,12 @@ void Parker::Release (Parker * p) {
   guarantee (p->AssociatedWith != NULL, "invariant") ;
   guarantee (p->FreeNext == NULL      , "invariant") ;
   p->AssociatedWith = NULL ;
-  for (;;) {
-    // Push p onto FreeList
-    Parker * List = FreeList ;
-    p->FreeNext = List ;
-    if (Atomic::cmpxchg_ptr (p, &FreeList, List) == List) break ;
+
+  Thread::SpinAcquire(&ListLock, "ParkerFreeListRelease");
+  {
+    p->FreeNext = FreeList;
+    FreeList = p;
   }
+  Thread::SpinRelease(&ListLock);
 }
 
diff --git a/src/share/vm/runtime/thread.cpp b/src/share/vm/runtime/thread.cpp
index 9f31c25b7c6479bb90c5bc71712d1242ad8d847b..af24fac2c6a6207459fd7002363695e63815d0a4 100644
--- a/src/share/vm/runtime/thread.cpp
+++ b/src/share/vm/runtime/thread.cpp
@@ -4446,9 +4446,7 @@ void Thread::SpinAcquire (volatile int * adr, const char * LockName) {
         ++ctr ;
         if ((ctr & 0xFFF) == 0 || !os::is_MP()) {
            if (Yields > 5) {
-             // Consider using a simple NakedSleep() instead.
-             // Then SpinAcquire could be called by non-JVM threads
-             Thread::current()->_ParkEvent->park(1) ;
+             os::naked_short_sleep(1);
            } else {
              os::NakedYield() ;
              ++Yields ;