< prev index next >

src/hotspot/share/runtime/objectMonitor.hpp

Print this page
rev 56634 : imported patch 8230876.patch
rev 56635 : v2.00 -> v2.05 (CR5/v2.05/8-for-jdk13) patches combined into one; merge with 8229212.patch; merge with jdk-14+11; merge with 8230184.patch; merge with 8230876.patch; merge with jdk-14+15; merge with jdk-14+18.
rev 56637 : Add OM_CACHE_LINE_SIZE so that ObjectMonitor cache line sizes can be experimented with independently of DEFAULT_CACHE_LINE_SIZE; for SPARC and X64 configs that use 128 for DEFAULT_CACHE_LINE_SIZE, we are experimenting with 64; move _previous_owner_tid and _allocation_state fields to share the cache line with ObjectMonitor::_header; put ObjectMonitor::_ref_count on its own cache line after _owner; add 'int* count_p' parameter to deflate_monitor_list() and deflate_monitor_list_using_JT() and push counter updates down to where the ObjectMonitors are actually removed from the in-use lists; monitors_iterate() async deflation check should use negative ref_count; add 'JavaThread* target' param to deflate_per_thread_idle_monitors_using_JT() add deflate_common_idle_monitors_using_JT() to make it clear which JavaThread* is the target of the work and which is the calling JavaThread* (self); g_free_list, g_om_in_use_list and g_om_in_use_count are now static to synchronizer.cpp (reduce scope); add more diagnostic info to some assert()'s; minor code cleanups and code motion; save_om_ptr() should detect a race with a deflating thread that is bailing out and cause a retry when the ref_count field is not positive; merge with jdk-14+11; add special GC support for TestHumongousClassLoader.java; merge with 8230184.patch; merge with jdk-14+14; merge with jdk-14+18.
rev 56638 : Merge the remainder of the lock-free monitor list changes from v2.06 with v2.06a and v2.06b after running the changes through the edit scripts; merge pieces from dcubed.monitor_deflate_conc.v2.06d in dcubed.monitor_deflate_conc.v2.06[ac]; merge pieces from dcubed.monitor_deflate_conc.v2.06e into dcubed.monitor_deflate_conc.v2.06c; merge with jdk-14+11; test work around for test/jdk/tools/jlink/multireleasejar/JLinkMultiReleaseJarTest.java should not been needed anymore; merge with jdk-14+18.
rev 56639 : loosen a couple more counter checks due to races observed in testing; simplify om_release() extraction of mid since list head or cur_mid_in_use is marked; simplify deflate_monitor_list() extraction of mid since there are no parallel deleters due to the safepoint; simplify deflate_monitor_list_using_JT() extraction of mid since list head or cur_mid_in_use is marked; prepend_block_to_lists() - simplify based on David H's comments; does not need load_acquire() or release_store() because of the cmpxchg(); prepend_to_common() - simplify to use mark_next_loop() for m and use mark_list_head() and release_store() for the non-empty list case; add more debugging for "Non-balanced monitor enter/exit" failure mode; fix race in inflate() in the "CASE: neutral" code path; install_displaced_markword_in_object() does not need to clear the header field since that is handled when the ObjectMonitor is moved from the global free list; LSuccess should clear boxReg to set ICC.ZF=1 to avoid depending on existing boxReg contents; update fast_unlock() to detect when object no longer refers to the same ObjectMonitor and take fast path exit instead; clarify fast_lock() code where we detect when object no longer refers to the same ObjectMonitor; add/update comments for movptr() calls where we move a literal into an Address; remove set_owner(); refactor setting of owner field into set_owner_from(2 versions), set_owner_from_BasicLock(), and try_set_owner_from(); the new functions include monitorinflation+owner logging; extract debug code from v2.06 and v2.07 and move to v2.07.debug; change 'jccb' -> 'jcc' and 'jmpb' -> 'jmp' as needed; checkpoint initial version of MacroAssembler::inc_om_ref_count(); update LP64 MacroAssembler::fast_lock() and fast_unlock() to use inc_om_ref_count(); fast_lock() return flag setting logic can use 'testptr(tmpReg, tmpReg)' instead of 'cmpptr(tmpReg, 0)' since that's more efficient; fast_unlock() LSuccess return flag setting logic can use 'testl (boxReg, 0)' instead of 'xorptr(boxReg, boxReg)' since that's more efficient; cleanup "fast-path" vs "fast path" and "slow-path" vs "slow path"; update MacroAssembler::rtm_inflated_locking() to use inc_om_ref_count(); update MacroAssembler::fast_lock() to preserve the flags before decrementing ref_count and restore the flags afterwards; this is more clean than depending on the contents of rax/tmpReg; coleenp CR - refactor async monitor deflation work from ServiceThread::service_thread_entry() to ObjectSynchronizer::deflate_idle_monitors_using_JT(); rehn,eosterlund CR - add support for HandshakeAfterDeflateIdleMonitors for platforms that don't have ObjectMonitor ref_count support implemented in C2 fast_lock() and fast_unlock().

@@ -118,34 +118,58 @@
 //
 //   - The _recursions field should be of type int, or int32_t but not
 //     intptr_t. There's no reason to use a 64-bit type for this field
 //     in a 64-bit JVM.
 
+#ifndef OM_CACHE_LINE_SIZE
+// Use DEFAULT_CACHE_LINE_SIZE if not already specified for
+// the current build platform.
+#define OM_CACHE_LINE_SIZE DEFAULT_CACHE_LINE_SIZE
+#endif
+
 class ObjectMonitor {
+  friend class ObjectMonitorHandle;
   friend class ObjectSynchronizer;
   friend class ObjectWaiter;
   friend class VMStructs;
   JVMCI_ONLY(friend class JVMCIVMStructs;)
 
   // The sync code expects the header field to be at offset zero (0).
   // Enforced by the assert() in header_addr().
   volatile markWord _header;        // displaced object header word - mark
   void* volatile _object;           // backward object pointer - strong root
- public:
-  ObjectMonitor* _next_om;          // Next ObjectMonitor* linkage
- private:
+  typedef enum {
+    Free = 0,  // Free must be 0 for monitor to be free after memset(..,0,..).
+    New,
+    Old
+  } AllocationState;
+  AllocationState _allocation_state;
   // Separate _header and _owner on different cache lines since both can
-  // have busy multi-threaded access. _header and _object are set at
-  // initial inflation and _object doesn't change until deflation so
-  // _object is a good choice to share the cache line with _header.
-  // _next_om shares _header's cache line for pre-monitor list historical
-  // reasons. _next_om only changes if the next ObjectMonitor is deflated.
-  DEFINE_PAD_MINUS_SIZE(0, DEFAULT_CACHE_LINE_SIZE,
-                        sizeof(volatile markWord) + sizeof(void* volatile) +
-                        sizeof(ObjectMonitor *));
+  // have busy multi-threaded access. _header, _object and _allocation_state
+  // are set at initial inflation. _object and _allocation_state don't
+  // change until deflation so _object and _allocation_state are good
+  // choices to share the cache line with _header.
+  DEFINE_PAD_MINUS_SIZE(0, OM_CACHE_LINE_SIZE, sizeof(volatile markWord) +
+                        sizeof(void* volatile) + sizeof(AllocationState));
+  // Used by async deflation as a marker in the _owner field:
+  #define DEFLATER_MARKER reinterpret_cast<void*>(-1)
   void* volatile _owner;            // pointer to owning thread OR BasicLock
   volatile jlong _previous_owner_tid;  // thread id of the previous owner of the monitor
+  // Separate _owner and _ref_count on different cache lines since both
+  // can have busy multi-threaded access. _previous_owner_tid is only
+  // changed by ObjectMonitor::exit() so it is a good choice to share the
+  // cache line with _owner.
+  DEFINE_PAD_MINUS_SIZE(1, OM_CACHE_LINE_SIZE, sizeof(void* volatile) +
+                        sizeof(volatile jlong));
+  volatile jint _ref_count;         // ref count for ObjectMonitor* and used by the async deflation
+                                    // protocol. See ObjectSynchronizer::deflate_monitor_using_JT().
+  // Separate _ref_count and _next_om on different cache lines since
+  // both can have busy multi-threaded access.
+  DEFINE_PAD_MINUS_SIZE(2, OM_CACHE_LINE_SIZE, sizeof(volatile jint));
+ public:                            // for static synchronizer.cpp access:
+  ObjectMonitor* volatile _next_om;  // Next ObjectMonitor* linkage
+ private:
   volatile intptr_t _recursions;    // recursion count, 0 for first entry
   ObjectWaiter* volatile _EntryList;  // Threads blocked on entry or reentry.
                                       // The list is actually composed of WaitNodes,
                                       // acting as proxies for Threads.
 

@@ -156,11 +180,12 @@
   volatile int _Spinner;            // for exit->spinner handoff optimization
   volatile int _SpinDuration;
 
   volatile jint  _contentions;      // Number of active contentions in enter(). It is used by is_busy()
                                     // along with other fields to determine if an ObjectMonitor can be
-                                    // deflated. See ObjectSynchronizer::deflate_monitor().
+                                    // deflated. See ObjectSynchronizer::deflate_monitor() and
+                                    // ObjectSynchronizer::deflate_monitor_using_JT().
  protected:
   ObjectWaiter* volatile _WaitSet;  // LL of threads wait()ing on the monitor
   volatile jint  _waiters;          // number of waiting threads
  private:
   volatile int _WaitSetLock;        // protects Wait Queue - simple spinlock

@@ -198,10 +223,11 @@
   // TODO-FIXME: the "offset" routines should return a type of off_t instead of int ...
   // ByteSize would also be an appropriate type.
   static int header_offset_in_bytes()      { return offset_of(ObjectMonitor, _header); }
   static int object_offset_in_bytes()      { return offset_of(ObjectMonitor, _object); }
   static int owner_offset_in_bytes()       { return offset_of(ObjectMonitor, _owner); }
+  static int ref_count_offset_in_bytes()   { return offset_of(ObjectMonitor, _ref_count); }
   static int recursions_offset_in_bytes()  { return offset_of(ObjectMonitor, _recursions); }
   static int cxq_offset_in_bytes()         { return offset_of(ObjectMonitor, _cxq); }
   static int succ_offset_in_bytes()        { return offset_of(ObjectMonitor, _succ); }
   static int EntryList_offset_in_bytes()   { return offset_of(ObjectMonitor, _EntryList); }
 

@@ -223,18 +249,40 @@
   volatile markWord* header_addr();
   void               set_header(markWord hdr);
 
   intptr_t is_busy() const {
     // TODO-FIXME: assert _owner == null implies _recursions = 0
-    return _contentions|_waiters|intptr_t(_owner)|intptr_t(_cxq)|intptr_t(_EntryList);
+    // We do not include _ref_count in the is_busy() check because
+    // _ref_count is for indicating that the ObjectMonitor* is in
+    // use which is orthogonal to whether the ObjectMonitor itself
+    // is in use for a locking operation.
+    intptr_t ret_code = _contentions | _waiters | intptr_t(_cxq) | intptr_t(_EntryList);
+    if (!AsyncDeflateIdleMonitors) {
+      ret_code |= intptr_t(_owner);
+    } else {
+      if (_owner != DEFLATER_MARKER) {
+        ret_code |= intptr_t(_owner);
+      }
+    }
+    return ret_code;
   }
   const char* is_busy_to_string(stringStream* ss);
 
   intptr_t  is_entered(Thread* current) const;
 
-  void*     owner() const;
-  void      set_owner(void* owner);
+  void*     owner() const;  // Returns NULL if DEFLATER_MARKER is observed.
+  // Returns true if owner field == DEFLATER_MARKER and false otherwise.
+  bool      owner_is_DEFLATER_MARKER();
+  // Set _owner field to new_value; current value must match old_value.
+  void      set_owner_from(void* new_value, void* old_value);
+  // Set _owner field to new_value; current value must match old_value1 or old_value2.
+  void      set_owner_from(void* new_value, void* old_value1, void* old_value2);
+  // Set _owner field to self; current value must match basic_lock_p.
+  void      set_owner_from_BasicLock(Thread* self, void* basic_lock_p);
+  // Try to set _owner field to new_value if the current value matches
+  // old_value. Otherwise, does not change the _owner field.
+  void*     try_set_owner_from(void* new_value, void* old_value);
 
   jint      waiters() const;
 
   jint      contentions() const;
   intptr_t  recursions() const                                         { return _recursions; }

@@ -261,11 +309,11 @@
     // _cxq == 0 _succ == NULL _owner == NULL _waiters == 0
     // _contentions == 0 EntryList  == NULL
     // _recursions == 0 _WaitSet == NULL
     DEBUG_ONLY(stringStream ss;)
     assert((is_busy() | _recursions) == 0, "freeing in-use monitor: %s, "
-           "recursions=" INTPTR_FORMAT, is_busy_to_string(&ss), _recursions);
+           "recursions=" INTX_FORMAT, is_busy_to_string(&ss), _recursions);
     _succ          = NULL;
     _EntryList     = NULL;
     _cxq           = NULL;
     _WaitSet       = NULL;
     _recursions    = 0;

@@ -274,23 +322,34 @@
  public:
 
   void*     object() const;
   void*     object_addr();
   void      set_object(void* obj);
+  void      set_allocation_state(AllocationState s);
+  AllocationState allocation_state() const;
+  bool      is_free() const;
+  bool      is_active() const;
+  bool      is_old() const;
+  bool      is_new() const;
+  void      dec_ref_count();
+  void      inc_ref_count();
+  jint      ref_count() const;
 
   // Returns true if the specified thread owns the ObjectMonitor. Otherwise
   // returns false and throws IllegalMonitorStateException (IMSE).
   bool      check_owner(Thread* THREAD);
   void      clear();
+  void      clear_using_JT();
 
   void      enter(TRAPS);
   void      exit(bool not_suspended, TRAPS);
   void      wait(jlong millis, bool interruptable, TRAPS);
   void      notify(TRAPS);
   void      notifyAll(TRAPS);
 
   void      print() const;
+  void      print_debug_style_on(outputStream* st) const;
   void      print_on(outputStream* st) const;
 
 // Use the following at your own risk
   intptr_t  complete_exit(TRAPS);
   void      reenter(intptr_t recursions, TRAPS);

@@ -302,12 +361,44 @@
   void      DequeueSpecificWaiter(ObjectWaiter* waiter);
   void      EnterI(TRAPS);
   void      ReenterI(Thread* self, ObjectWaiter* self_node);
   void      UnlinkAfterAcquire(Thread* self, ObjectWaiter* self_node);
   int       TryLock(Thread* self);
-  int       NotRunnable(Thread* self, Thread * Owner);
+  int       NotRunnable(Thread* self, Thread* Owner);
   int       TrySpin(Thread* self);
   void      ExitEpilog(Thread* self, ObjectWaiter* Wakee);
   bool      ExitSuspendEquivalent(JavaThread* self);
+  void      install_displaced_markword_in_object(const oop obj);
 };
 
+// A helper object for managing an ObjectMonitor*'s ref_count. There
+// are special safety considerations when async deflation is used.
+class ObjectMonitorHandle : public StackObj {
+ private:
+  ObjectMonitor* _om_ptr;
+ public:
+  ObjectMonitorHandle() { _om_ptr = NULL; }
+  ~ObjectMonitorHandle();
+
+  ObjectMonitor* om_ptr() const { return _om_ptr; }
+  // Save the ObjectMonitor* associated with the specified markWord and
+  // increment the ref_count.
+  bool save_om_ptr(oop object, markWord mark);
+
+  // For internal used by ObjectSynchronizer::monitors_iterate().
+  ObjectMonitorHandle(ObjectMonitor* _om_ptr);
+  // For internal use by ObjectSynchronizer::inflate().
+  void set_om_ptr(ObjectMonitor* om_ptr);
+};
+
+// Macro to use guarantee() for more strict AsyncDeflateIdleMonitors
+// checks and assert() otherwise.
+#define ADIM_guarantee(p, ...)       \
+  do {                               \
+    if (AsyncDeflateIdleMonitors) {  \
+      guarantee(p, __VA_ARGS__);     \
+    } else {                         \
+      assert(p, __VA_ARGS__);        \
+    }                                \
+  } while (0)
+
 #endif // SHARE_RUNTIME_OBJECTMONITOR_HPP
< prev index next >