view src/share/vm/gc_implementation/g1/heapRegionRemSet.hpp @ 2034:7e37af9d69ef

7011379: G1: overly long concurrent marking cycles Summary: This changeset introduces filtering of SATB buffers at the point when they are about to be enqueued. If this filtering clears enough entries on each buffer, the buffer can then be re-used and not enqueued. This cuts down the number of SATB buffers that need to be processed by the concurrent marking threads. Reviewed-by: johnc, ysr
author tonyp
date Wed, 19 Jan 2011 09:35:17 -0500
parents c18cbe5936b8
children 97ba643ea3ed
line wrap: on
line source
/*
 * Copyright (c) 2001, 2010, Oracle and/or its affiliates. All rights reserved.
 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
 *
 * This code is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License version 2 only, as
 * published by the Free Software Foundation.
 *
 * This code is distributed in the hope that it will be useful, but WITHOUT
 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
 * version 2 for more details (a copy is included in the LICENSE file that
 * accompanied this code).
 *
 * You should have received a copy of the GNU General Public License version
 * 2 along with this work; if not, write to the Free Software Foundation,
 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
 *
 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
 * or visit www.oracle.com if you need additional information or have any
 * questions.
 *
 */

#ifndef SHARE_VM_GC_IMPLEMENTATION_G1_HEAPREGIONREMSET_HPP
#define SHARE_VM_GC_IMPLEMENTATION_G1_HEAPREGIONREMSET_HPP

#include "gc_implementation/g1/sparsePRT.hpp"

// Remembered set for a heap region.  Represent a set of "cards" that
// contain pointers into the owner heap region.  Cards are defined somewhat
// abstractly, in terms of what the "BlockOffsetTable" in use can parse.

class G1CollectedHeap;
class G1BlockOffsetSharedArray;
class HeapRegion;
class HeapRegionRemSetIterator;
class PosParPRT;
class SparsePRT;


// The "_coarse_map" is a bitmap with one bit for each region, where set
// bits indicate that the corresponding region may contain some pointer
// into the owning region.

// The "_fine_grain_entries" array is an open hash table of PerRegionTables
// (PRTs), indicating regions for which we're keeping the RS as a set of
// cards.  The strategy is to cap the size of the fine-grain table,
// deleting an entry and setting the corresponding coarse-grained bit when
// we would overflow this cap.

// We use a mixture of locking and lock-free techniques here.  We allow
// threads to locate PRTs without locking, but threads attempting to alter
// a bucket list obtain a lock.  This means that any failing attempt to
// find a PRT must be retried with the lock.  It might seem dangerous that
// a read can find a PRT that is concurrently deleted.  This is all right,
// because:
//
//   1) We only actually free PRT's at safe points (though we reuse them at
//      other times).
//   2) We find PRT's in an attempt to add entries.  If a PRT is deleted,
//      it's _coarse_map bit is set, so the that we were attempting to add
//      is represented.  If a deleted PRT is re-used, a thread adding a bit,
//      thinking the PRT is for a different region, does no harm.

class OtherRegionsTable VALUE_OBJ_CLASS_SPEC {
  friend class HeapRegionRemSetIterator;

  G1CollectedHeap* _g1h;
  Mutex            _m;
  HeapRegion*      _hr;

  // These are protected by "_m".
  BitMap      _coarse_map;
  size_t      _n_coarse_entries;
  static jint _n_coarsenings;

  PosParPRT** _fine_grain_regions;
  size_t      _n_fine_entries;

#define SAMPLE_FOR_EVICTION 1
#if SAMPLE_FOR_EVICTION
  size_t        _fine_eviction_start;
  static size_t _fine_eviction_stride;
  static size_t _fine_eviction_sample_size;
#endif

  SparsePRT   _sparse_table;

  // These are static after init.
  static size_t _max_fine_entries;
  static size_t _mod_max_fine_entries_mask;

  // Requires "prt" to be the first element of the bucket list appropriate
  // for "hr".  If this list contains an entry for "hr", return it,
  // otherwise return "NULL".
  PosParPRT* find_region_table(size_t ind, HeapRegion* hr) const;

  // Find, delete, and return a candidate PosParPRT, if any exists,
  // adding the deleted region to the coarse bitmap.  Requires the caller
  // to hold _m, and the fine-grain table to be full.
  PosParPRT* delete_region_table();

  // If a PRT for "hr" is in the bucket list indicated by "ind" (which must
  // be the correct index for "hr"), delete it and return true; else return
  // false.
  bool del_single_region_table(size_t ind, HeapRegion* hr);

  static jint _cache_probes;
  static jint _cache_hits;

  // Indexed by thread X heap region, to minimize thread contention.
  static int** _from_card_cache;
  static size_t _from_card_cache_max_regions;
  static size_t _from_card_cache_mem_size;

public:
  OtherRegionsTable(HeapRegion* hr);

  HeapRegion* hr() const { return _hr; }

  // For now.  Could "expand" some tables in the future, so that this made
  // sense.
  void add_reference(OopOrNarrowOopStar from, int tid);

  void add_reference(OopOrNarrowOopStar from) {
    return add_reference(from, 0);
  }

  // Removes any entries shown by the given bitmaps to contain only dead
  // objects.
  void scrub(CardTableModRefBS* ctbs, BitMap* region_bm, BitMap* card_bm);

  // Not const because it takes a lock.
  size_t occupied() const;
  size_t occ_fine() const;
  size_t occ_coarse() const;
  size_t occ_sparse() const;

  static jint n_coarsenings() { return _n_coarsenings; }

  // Returns size in bytes.
  // Not const because it takes a lock.
  size_t mem_size() const;
  static size_t static_mem_size();
  static size_t fl_mem_size();

  bool contains_reference(OopOrNarrowOopStar from) const;
  bool contains_reference_locked(OopOrNarrowOopStar from) const;

  void clear();

  // Specifically clear the from_card_cache.
  void clear_fcc();

  // "from_hr" is being cleared; remove any entries from it.
  void clear_incoming_entry(HeapRegion* from_hr);

  // Declare the heap size (in # of regions) to the OtherRegionsTable.
  // (Uses it to initialize from_card_cache).
  static void init_from_card_cache(size_t max_regions);

  // Declares that only regions i s.t. 0 <= i < new_n_regs are in use.
  // Make sure any entries for higher regions are invalid.
  static void shrink_from_card_cache(size_t new_n_regs);

  static void print_from_card_cache();

};


class HeapRegionRemSet : public CHeapObj {
  friend class VMStructs;
  friend class HeapRegionRemSetIterator;

public:
  enum Event {
    Event_EvacStart, Event_EvacEnd, Event_RSUpdateEnd
  };

private:
  G1BlockOffsetSharedArray* _bosa;
  G1BlockOffsetSharedArray* bosa() const { return _bosa; }

  OtherRegionsTable _other_regions;

  enum ParIterState { Unclaimed, Claimed, Complete };
  volatile ParIterState _iter_state;
  volatile jlong _iter_claimed;

  // Unused unless G1RecordHRRSOops is true.

  static const int MaxRecorded = 1000000;
  static OopOrNarrowOopStar* _recorded_oops;
  static HeapWord**          _recorded_cards;
  static HeapRegion**        _recorded_regions;
  static int                 _n_recorded;

  static const int MaxRecordedEvents = 1000;
  static Event*       _recorded_events;
  static int*         _recorded_event_index;
  static int          _n_recorded_events;

  static void print_event(outputStream* str, Event evnt);

public:
  HeapRegionRemSet(G1BlockOffsetSharedArray* bosa,
                   HeapRegion* hr);

  static int num_par_rem_sets();
  static void setup_remset_size();

  HeapRegion* hr() const {
    return _other_regions.hr();
  }

  size_t occupied() const {
    return _other_regions.occupied();
  }
  size_t occ_fine() const {
    return _other_regions.occ_fine();
  }
  size_t occ_coarse() const {
    return _other_regions.occ_coarse();
  }
  size_t occ_sparse() const {
    return _other_regions.occ_sparse();
  }

  static jint n_coarsenings() { return OtherRegionsTable::n_coarsenings(); }

  /* Used in the sequential case.  Returns "true" iff this addition causes
     the size limit to be reached. */
  void add_reference(OopOrNarrowOopStar from) {
    _other_regions.add_reference(from);
  }

  /* Used in the parallel case.  Returns "true" iff this addition causes
     the size limit to be reached. */
  void add_reference(OopOrNarrowOopStar from, int tid) {
    _other_regions.add_reference(from, tid);
  }

  // Removes any entries shown by the given bitmaps to contain only dead
  // objects.
  void scrub(CardTableModRefBS* ctbs, BitMap* region_bm, BitMap* card_bm);

  // The region is being reclaimed; clear its remset, and any mention of
  // entries for this region in other remsets.
  void clear();

  // Forget any entries due to pointers from "from_hr".
  void clear_incoming_entry(HeapRegion* from_hr) {
    _other_regions.clear_incoming_entry(from_hr);
  }

#if 0
  virtual void cleanup() = 0;
#endif

  // Should be called from single-threaded code.
  void init_for_par_iteration();
  // Attempt to claim the region.  Returns true iff this call caused an
  // atomic transition from Unclaimed to Claimed.
  bool claim_iter();
  // Sets the iteration state to "complete".
  void set_iter_complete();
  // Returns "true" iff the region's iteration is complete.
  bool iter_is_complete();

  // Support for claiming blocks of cards during iteration
  void set_iter_claimed(size_t x) { _iter_claimed = (jlong)x; }
  size_t iter_claimed() const { return (size_t)_iter_claimed; }
  // Claim the next block of cards
  size_t iter_claimed_next(size_t step) {
    size_t current, next;
    do {
      current = iter_claimed();
      next = current + step;
    } while (Atomic::cmpxchg((jlong)next, &_iter_claimed, (jlong)current) != (jlong)current);
    return current;
  }

  // Initialize the given iterator to iterate over this rem set.
  void init_iterator(HeapRegionRemSetIterator* iter) const;

#if 0
  // Apply the "do_card" method to the start address of every card in the
  // rem set.  Returns false if some application of the closure aborted.
  virtual bool card_iterate(CardClosure* iter) = 0;
#endif

  // The actual # of bytes this hr_remset takes up.
  size_t mem_size() {
    return _other_regions.mem_size()
      // This correction is necessary because the above includes the second
      // part.
      + sizeof(this) - sizeof(OtherRegionsTable);
  }

  // Returns the memory occupancy of all static data structures associated
  // with remembered sets.
  static size_t static_mem_size() {
    return OtherRegionsTable::static_mem_size();
  }

  // Returns the memory occupancy of all free_list data structures associated
  // with remembered sets.
  static size_t fl_mem_size() {
    return OtherRegionsTable::fl_mem_size();
  }

  bool contains_reference(OopOrNarrowOopStar from) const {
    return _other_regions.contains_reference(from);
  }
  void print() const;

  // Called during a stop-world phase to perform any deferred cleanups.
  // The second version may be called by parallel threads after then finish
  // collection work.
  static void cleanup();
  static void par_cleanup();

  // Declare the heap size (in # of regions) to the HeapRegionRemSet(s).
  // (Uses it to initialize from_card_cache).
  static void init_heap(size_t max_regions) {
    OtherRegionsTable::init_from_card_cache(max_regions);
  }

  // Declares that only regions i s.t. 0 <= i < new_n_regs are in use.
  static void shrink_heap(size_t new_n_regs) {
    OtherRegionsTable::shrink_from_card_cache(new_n_regs);
  }

#ifndef PRODUCT
  static void print_from_card_cache() {
    OtherRegionsTable::print_from_card_cache();
  }
#endif

  static void record(HeapRegion* hr, OopOrNarrowOopStar f);
  static void print_recorded();
  static void record_event(Event evnt);

  // Run unit tests.
#ifndef PRODUCT
  static void test();
#endif

};

class HeapRegionRemSetIterator : public CHeapObj {

  // The region over which we're iterating.
  const HeapRegionRemSet* _hrrs;

  // Local caching of HRRS fields.
  const BitMap*             _coarse_map;
  PosParPRT**               _fine_grain_regions;

  G1BlockOffsetSharedArray* _bosa;
  G1CollectedHeap*          _g1h;

  // The number yielded since initialization.
  size_t _n_yielded_fine;
  size_t _n_yielded_coarse;
  size_t _n_yielded_sparse;

  // If true we're iterating over the coarse table; if false the fine
  // table.
  enum IterState {
    Sparse,
    Fine,
    Coarse
  };
  IterState _is;

  // In both kinds of iteration, heap offset of first card of current
  // region.
  size_t _cur_region_card_offset;
  // Card offset within cur region.
  size_t _cur_region_cur_card;

  // Coarse table iteration fields:

  // Current region index;
  int _coarse_cur_region_index;
  int _coarse_cur_region_cur_card;

  bool coarse_has_next(size_t& card_index);

  // Fine table iteration fields:

  // Index of bucket-list we're working on.
  int _fine_array_index;
  // Per Region Table we're doing within current bucket list.
  PosParPRT* _fine_cur_prt;

  /* SparsePRT::*/ SparsePRTIter _sparse_iter;

  void fine_find_next_non_null_prt();

  bool fine_has_next();
  bool fine_has_next(size_t& card_index);

public:
  // We require an iterator to be initialized before use, so the
  // constructor does little.
  HeapRegionRemSetIterator();

  void initialize(const HeapRegionRemSet* hrrs);

  // If there remains one or more cards to be yielded, returns true and
  // sets "card_index" to one of those cards (which is then considered
  // yielded.)   Otherwise, returns false (and leaves "card_index"
  // undefined.)
  bool has_next(size_t& card_index);

  size_t n_yielded_fine() { return _n_yielded_fine; }
  size_t n_yielded_coarse() { return _n_yielded_coarse; }
  size_t n_yielded_sparse() { return _n_yielded_sparse; }
  size_t n_yielded() {
    return n_yielded_fine() + n_yielded_coarse() + n_yielded_sparse();
  }
};

#if 0
class CardClosure: public Closure {
public:
  virtual void do_card(HeapWord* card_start) = 0;
};

#endif

#endif // SHARE_VM_GC_IMPLEMENTATION_G1_HEAPREGIONREMSET_HPP