1 /*
  2  * Copyright (c) 2024, 2025, Oracle and/or its affiliates. All rights reserved.
  3  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
  4  *
  5  * This code is free software; you can redistribute it and/or modify it
  6  * under the terms of the GNU General Public License version 2 only, as
  7  * published by the Free Software Foundation.
  8  *
  9  * This code is distributed in the hope that it will be useful, but WITHOUT
 10  * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 11  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
 12  * version 2 for more details (a copy is included in the LICENSE file that
 13  * accompanied this code).
 14  *
 15  * You should have received a copy of the GNU General Public License version
 16  * 2 along with this work; if not, write to the Free Software Foundation,
 17  * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
 18  *
 19  * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
 20  * or visit www.oracle.com if you need additional information or have any
 21  * questions.
 22  *
 23  */
 24 
 25 #ifndef SHARE_CDS_ARCHIVEHEAPWRITER_HPP
 26 #define SHARE_CDS_ARCHIVEHEAPWRITER_HPP
 27 
 28 #include "cds/heapShared.hpp"
 29 #include "memory/allocation.hpp"
 30 #include "memory/allStatic.hpp"
 31 #include "oops/oopHandle.hpp"
 32 #include "utilities/bitMap.hpp"
 33 #include "utilities/exceptions.hpp"
 34 #include "utilities/growableArray.hpp"
 35 #include "utilities/macros.hpp"
 36 #include "utilities/resourceHash.hpp"
 37 
 38 class MemRegion;
 39 
 40 class ArchiveHeapInfo {
 41   MemRegion _buffer_region;             // Contains the archived objects to be written into the CDS archive.
 42   CHeapBitMap _oopmap;
 43   CHeapBitMap _ptrmap;
 44   HeapRootSegments _heap_root_segments;
 45 
 46 public:
 47   ArchiveHeapInfo() : _buffer_region(), _oopmap(128, mtClassShared), _ptrmap(128, mtClassShared) {}
 48   bool is_used() { return !_buffer_region.is_empty(); }
 49 
 50   MemRegion buffer_region() { return _buffer_region; }
 51   void set_buffer_region(MemRegion r) { _buffer_region = r; }
 52 
 53   char* buffer_start() { return (char*)_buffer_region.start(); }
 54   size_t buffer_byte_size() { return _buffer_region.byte_size();    }
 55 
 56   CHeapBitMap* oopmap() { return &_oopmap; }
 57   CHeapBitMap* ptrmap() { return &_ptrmap; }
 58 
 59   void set_heap_root_segments(HeapRootSegments segments) { _heap_root_segments = segments; };
 60   HeapRootSegments heap_root_segments() { return _heap_root_segments; }
 61 };
 62 
 63 #if INCLUDE_CDS_JAVA_HEAP
 64 class ArchiveHeapWriter : AllStatic {
 65   friend class HeapShared;
 66   // ArchiveHeapWriter manipulates three types of addresses:
 67   //
 68   //     "source" vs "buffered" vs "requested"
 69   //
 70   // (Note: the design and convention is the same as for the archiving of Metaspace objects.
 71   //  See archiveBuilder.hpp.)
 72   //
 73   // - "source objects" are regular Java objects allocated during the execution
 74   //   of "java -Xshare:dump". They can be used as regular oops.
 75   //
 76   //   Between HeapShared::start_scanning_for_oops() and HeapShared::end_scanning_for_oops(),
 77   //   we recursively search for the oops that need to be stored into the CDS archive.
 78   //   These are entered into HeapShared::archived_object_cache().
 79   //
 80   // - "buffered objects" are copies of the "source objects", and are stored in into
 81   //   ArchiveHeapWriter::_buffer, which is a GrowableArray that sits outside of
 82   //   the valid heap range. Therefore we avoid using the addresses of these copies
 83   //   as oops. They are usually called "buffered_addr" in the code (of the type "address").
 84   //
 85   //   The buffered objects are stored contiguously, possibly with interleaving fillers
 86   //   to make sure no objects span across boundaries of MIN_GC_REGION_ALIGNMENT.
 87   //
 88   // - Each archived object has a "requested address" -- at run time, if the object
 89   //   can be mapped at this address, we can avoid relocation.
 90   //
 91   // The requested address is implemented differently depending on UseCompressedOops:
 92   //
 93   // UseCompressedOops == true:
 94   //   The archived objects are stored assuming that the runtime COOPS compression
 95   //   scheme is exactly the same as in dump time (or else a more expensive runtime relocation
 96   //   would be needed.)
 97   //
 98   //   At dump time, we assume that the runtime heap range is exactly the same as
 99   //   in dump time. The requested addresses of the archived objects are chosen such that
100   //   they would occupy the top end of a G1 heap (TBD when dumping is supported by other
101   //   collectors. See JDK-8298614).
102   //
103   // UseCompressedOops == false:
104   //   At runtime, the heap range is usually picked (randomly) by the OS, so we will almost always
105   //   need to perform relocation. Hence, the goal of the "requested address" is to ensure that
106   //   the contents of the archived objects are deterministic. I.e., the oop fields of archived
107   //   objects will always point to deterministic addresses.
108   //
109   //   For G1, the archived heap is written such that the lowest archived object is placed
110   //   at NOCOOPS_REQUESTED_BASE. (TBD after JDK-8298614).
111   // ----------------------------------------------------------------------
112 
113 public:
114   static const intptr_t NOCOOPS_REQUESTED_BASE = 0x10000000;
115 
116   // The minimum region size of all collectors that are supported by CDS.
117   // G1 heap region size can never be smaller than 1M.
118   // Shenandoah heap region size can never be smaller than 256K.
119   static constexpr int MIN_GC_REGION_ALIGNMENT = 256 * K;
120 
121 private:
122   class EmbeddedOopRelocator;
123   struct NativePointerInfo {
124     oop _src_obj;
125     int _field_offset;
126   };
127 
128   static GrowableArrayCHeap<u1, mtClassShared>* _buffer;
129 
130   // The number of bytes that have written into _buffer (may be smaller than _buffer->length()).
131   static size_t _buffer_used;
132 
133   // The heap root segments information.
134   static HeapRootSegments _heap_root_segments;
135 
136   // The address range of the requested location of the archived heap objects.
137   static address _requested_bottom;
138   static address _requested_top;
139 
140   static GrowableArrayCHeap<NativePointerInfo, mtClassShared>* _native_pointers;
141   static GrowableArrayCHeap<oop, mtClassShared>* _source_objs;
142 
143   // We sort _source_objs_order to minimize the number of bits in ptrmap and oopmap.
144   // See comments near the body of ArchiveHeapWriter::compare_objs_by_oop_fields().
145   // The objects will be written in the order of:
146   //_source_objs->at(_source_objs_order->at(0)._index)
147   // source_objs->at(_source_objs_order->at(1)._index)
148   // source_objs->at(_source_objs_order->at(2)._index)
149   // ...
150   struct HeapObjOrder {
151     int _index;    // The location of this object in _source_objs
152     int _rank;     // A lower rank means the object will be written at a lower location.
153   };
154   static GrowableArrayCHeap<HeapObjOrder, mtClassShared>* _source_objs_order;
155 
156   typedef ResizeableResourceHashtable<size_t, oop,
157       AnyObj::C_HEAP,
158       mtClassShared> BufferOffsetToSourceObjectTable;
159   static BufferOffsetToSourceObjectTable* _buffer_offset_to_source_obj_table;
160 
161   static void allocate_buffer();
162   static void ensure_buffer_space(size_t min_bytes);
163 
164   // Both Java bytearray and GrowableArraty use int indices and lengths. Do a safe typecast with range check
165   static int to_array_index(size_t i) {
166     assert(i <= (size_t)max_jint, "must be");
167     return (int)i;
168   }
169   static int to_array_length(size_t n) {
170     return to_array_index(n);
171   }
172 
173   template <typename T> static T offset_to_buffered_address(size_t offset) {
174     return (T)(_buffer->adr_at(to_array_index(offset)));
175   }
176 
177   static address buffer_bottom() {
178     return offset_to_buffered_address<address>(0);
179   }
180 
181   // The exclusive end of the last object that was copied into the buffer.
182   static address buffer_top() {
183     return buffer_bottom() + _buffer_used;
184   }
185 
186   static bool in_buffer(address buffered_addr) {
187     return (buffer_bottom() <= buffered_addr) && (buffered_addr < buffer_top());
188   }
189 
190   static size_t buffered_address_to_offset(address buffered_addr) {
191     assert(in_buffer(buffered_addr), "sanity");
192     return buffered_addr - buffer_bottom();
193   }
194 
195   static void root_segment_at_put(objArrayOop segment, int index, oop root);
196   static objArrayOop allocate_root_segment(size_t offset, int element_count);
197   static void copy_roots_to_buffer(GrowableArrayCHeap<oop, mtClassShared>* roots);
198   static void copy_source_objs_to_buffer(GrowableArrayCHeap<oop, mtClassShared>* roots);
199   static size_t copy_one_source_obj_to_buffer(oop src_obj);
200   static void update_stats(oop src_obj);
201 
202   static void maybe_fill_gc_region_gap(size_t required_byte_size);
203   static size_t filler_array_byte_size(int length);
204   static int filler_array_length(size_t fill_bytes);
205   static HeapWord* init_filler_array_at_buffer_top(int array_length, size_t fill_bytes);
206 
207   static void set_requested_address(ArchiveHeapInfo* info);
208   static void relocate_embedded_oops(GrowableArrayCHeap<oop, mtClassShared>* roots, ArchiveHeapInfo* info);
209   static void compute_ptrmap(ArchiveHeapInfo *info);
210   static bool is_in_requested_range(oop o);
211   static oop requested_obj_from_buffer_offset(size_t offset);
212 
213   static oop load_oop_from_buffer(oop* buffered_addr);
214   static oop load_oop_from_buffer(narrowOop* buffered_addr);
215   inline static void store_oop_in_buffer(oop* buffered_addr, oop requested_obj);
216   inline static void store_oop_in_buffer(narrowOop* buffered_addr, oop requested_obj);
217 
218   template <typename T> static oop load_source_oop_from_buffer(T* buffered_addr);
219   template <typename T> static void store_requested_oop_in_buffer(T* buffered_addr, oop request_oop);
220 
221   template <typename T> static T* requested_addr_to_buffered_addr(T* p);
222   template <typename T> static void relocate_field_in_buffer(T* field_addr_in_buffer, CHeapBitMap* oopmap);
223   template <typename T> static void mark_oop_pointer(T* buffered_addr, CHeapBitMap* oopmap);
224 
225   static void update_header_for_requested_obj(oop requested_obj, oop src_obj, Klass* src_klass);
226 
227   static int compare_objs_by_oop_fields(HeapObjOrder* a, HeapObjOrder* b);
228   static void sort_source_objs();
229 
230 public:
231   static void init() NOT_CDS_JAVA_HEAP_RETURN;
232   static void add_source_obj(oop src_obj);
233   static bool is_too_large_to_archive(size_t size);
234   static bool is_too_large_to_archive(oop obj);
235   static bool is_string_too_large_to_archive(oop string);
236   static void write(GrowableArrayCHeap<oop, mtClassShared>*, ArchiveHeapInfo* heap_info);
237   static address requested_address();  // requested address of the lowest achived heap object
238   static size_t get_filler_size_at(address buffered_addr);
239 
240   static void mark_native_pointer(oop src_obj, int offset);
241   static bool is_marked_as_native_pointer(ArchiveHeapInfo* heap_info, oop src_obj, int field_offset);
242   static oop source_obj_to_requested_obj(oop src_obj);
243   static oop buffered_addr_to_source_obj(address buffered_addr);
244   static address buffered_addr_to_requested_addr(address buffered_addr);
245 
246 };
247 #endif // INCLUDE_CDS_JAVA_HEAP
248 #endif // SHARE_CDS_ARCHIVEHEAPWRITER_HPP