File: OrderedHashTableHelper.h

package info (click to toggle)
webkit2gtk 2.48.5-1
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid
  • size: 429,764 kB
  • sloc: cpp: 3,697,587; javascript: 194,444; ansic: 169,997; python: 46,499; asm: 19,295; ruby: 18,528; perl: 16,602; xml: 4,650; yacc: 2,360; sh: 2,098; java: 1,993; lex: 1,327; pascal: 366; makefile: 298
file content (614 lines) | stat: -rw-r--r-- 30,931 bytes parent folder | download | duplicates (2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
/*
 * Copyright (C) 2024 Apple Inc. All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY APPLE INC. ``AS IS'' AND ANY
 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
 * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL APPLE INC. OR
 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
 * OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */

#pragma once

#include "HashMapHelper.h"
#include "JSImmutableButterfly.h"
#include "JSObject.h"

WTF_ALLOW_UNSAFE_BUFFER_USAGE_BEGIN

namespace JSC {

struct MapTraits {
    // DataTable:  [ Key_0, Val_0, NextKeyIndexInChain_0, Key_1, Val_1, NextKeyIndexInChain_1, ... ]
    //             | <------------ Entry_0 -----------> | <------------ Entry_1 -----------> |
    static constexpr uint8_t EntrySize = 3;
    static constexpr bool hasValueData = true;
};

struct SetTraits {
    // DataTable:  [ Key_0, NextKeyIndexInChain_0, Key_1, NextKeyIndexInChain_1, ... ]
    //             | <-------- Entry_0 --------> | <-------- Entry_1 --------> |
    static constexpr uint8_t EntrySize = 2;
    static constexpr bool hasValueData = false;
};

template<typename Traits>
class OrderedHashTable;

// ################ NonObsolete Table ################
//
//                      Count                                    Value(s)                           ValueType                  WriteBarrier
//               -------------------------------------------------------------------------------------------------------------------------------------
// TableStart ->           1          | AliveEntryCount                                            | TableSize                | NO
//                         1          | DeletedEntryCount                                          | TableSize                | NO
//                         1          | Capacity                                                   | TableSize                | NO
//                         1          | IterationEntry                                             | Entry                    | NO
//                    BucketCount     | HashTable:  { <BucketIndex, ChainStartKeyIndex>, ... }     | <TableIndex, TableIndex> | NO
//  TableEnd  -> Capacity * EntrySize | DataTable:  { <Entry_0, Data_0>, <Entry_1, Data_1>, ... }  | <TableIndex, JSValue>    | Yes (Key or Value)
//
// ################ Obsolete Table from Rehash ################
//
//                      Count                                    Value(s)                            ValueType                  WriteBarrier
//               -------------------------------------------------------------------------------------------------------------------------------------
// TableStart ->           1          | NextTable                                                   | JSCell                   | Yes
//                         1          | DeletedEntryCount                                           | TableSize                | NO
//                  DeletedEntryCount | DeletedEntries: [DeletedEntry_i, ...]                       | Entry                    | NO
//  TableEnd  ->          ...         | NotUsed                                                     | ...
//
// ################ Obsolete Table from Clear ################
//
//                      Count                                    Value(s)                            ValueType                  WriteBarrier
//               -------------------------------------------------------------------------------------------------------------------------------------
// TableStart ->           1          | NextTable                                                   | JSCell                   | Yes
//                         1          | ClearedTableSentinel                                        | TableSize                | NO
//  TableEnd  ->          ...         | NotUsed                                                     | ...
//
//
// This is based on the idea of CloseTable introduced in https://wiki.mozilla.org/User:Jorend/Deterministic_hash_tables.
//
// NonObsolete Table: A CloseTable flattened into a single array.
//
// - AliveEntryCount: The number of alive entries in the table.
// - DeletedEntryCount: The number of deleted entries in the table.
// - Capacity: The capacity of entries in the table.
// - IterationEntry: The temporary index only used for iteration.
// - HashTable: CloseTable is based on closed hashing. So, the each bucket in the HashTable points to the start EntryKeyIndex in the collision chain.
//   - FIXME: We can get rid of HashTable by applying OpenTable.
// - DataTable: The actual data table holds the tuples of key, value, and NextKeyIndexInChain for each entry.
//
//      Iterator                                   Map/Set
//          |                                         |
//          v                                         v
//    ObsoleteTable_0 --> ObsoleteTable_1 --> NonObsoleteTable_2
//
// Obsolete Table: Each rehash and clear would create a new table and update the obsolete one with additional info. Those info are used to
// transit the iterator (pointing to the obsolete table) to the latest alive table.
//
// - NextTable: A transition from the obsolete tables to the latest-non-obsolete one.
// - DeletedEntries: An array of the previously deleted entries used for updating iterator's index.
// - ClearedTableSentinel: The sentinel to indicate whether the obsolete table is retired due to a clearance.
//
// Note that all elements in the JSImmutableButterfly are in JSValue type. However, only the key and value in the DataTable are real JSValues.
// The others are used as unsigned integers wrapped by JSValue.
template<typename Traits>
class OrderedHashTableHelper {
public:
    using Storage = JSImmutableButterfly;
    using Helper = OrderedHashTableHelper<Traits>;
    using HashTable = OrderedHashTable<Traits>;
    using TableSize = uint32_t;
    using Entry = TableSize;
    using TableIndex = TableSize;

    static constexpr TableSize ClearedTableSentinel = -1;
    static constexpr TableIndex InvalidTableIndex = -1;
    static constexpr uint8_t EntrySize = Traits::EntrySize;
    static constexpr uint8_t ChainOffset = Traits::EntrySize - 1;

    static constexpr uint8_t LoadFactor = 1;
    static constexpr uint8_t InitialCapacity = 8;
    static constexpr TableSize LargeCapacity = 2 << 15;

    static_assert(EntrySize == MapTraits::EntrySize || EntrySize == SetTraits::EntrySize);

    OrderedHashTableHelper() = delete;

    ALWAYS_INLINE static TableSize toNumber(JSValue number) { return static_cast<TableSize>(number.asInt32()); }
    ALWAYS_INLINE static constexpr TableSize asNumber(Storage& storage, TableIndex index) { return toNumber(get(storage, index)); }
    ALWAYS_INLINE static JSValue toJSValue(Entry entry)
    {
        JSValue value = JSValue();
        OrderedHashTableTraits::set(&value, entry);
        return value;
    }

    ALWAYS_INLINE static bool isDeleted(VM& vm, JSValue value) { return value.isCell() && value.asCell() == vm.orderedHashTableDeletedValue(); }
    ALWAYS_INLINE static constexpr bool isValidTableIndex(TableIndex index) { return index != InvalidTableIndex; }
    ALWAYS_INLINE static JSValue invalidTableIndex() { return toJSValue(InvalidTableIndex); }

    ALWAYS_INLINE static JSValue get(Storage& storage, TableIndex index) { return storage.toButterfly()->contiguous().atUnsafe(index).get(); }
    ALWAYS_INLINE static JSValue* slot(Storage& storage, TableIndex index) { return storage.toButterfly()->contiguous().atUnsafe(index).slot(); }

    ALWAYS_INLINE static void set(Storage& storage, TableIndex index, JSValue value) { return storage.toButterfly()->contiguous().atUnsafe(index).setWithoutWriteBarrier(value); }
    ALWAYS_INLINE static void set(Storage& storage, TableIndex index, TableSize number) { OrderedHashTableTraits::set(slot(storage, index), number); }
    ALWAYS_INLINE static void setWithWriteBarrier(VM& vm, Storage& storage, TableIndex index, JSValue value) { storage.toButterfly()->contiguous().atUnsafe(index).set(vm, &storage, value); }

    /* -------------------------------- AliveEntryCount, DeletedEntryCount, Capacity, and IterationEntry -------------------------------- */
    ALWAYS_INLINE static constexpr TableIndex aliveEntryCountIndex() { return 0; }
    ALWAYS_INLINE static constexpr TableIndex deletedEntryCountIndex() { return aliveEntryCountIndex() + 1; }
    ALWAYS_INLINE static constexpr TableIndex capacityIndex() { return deletedEntryCountIndex() + 1; }
    ALWAYS_INLINE static constexpr TableIndex iterationEntryIndex() { return capacityIndex() + 1; }

    ALWAYS_INLINE static constexpr TableSize aliveEntryCount(Storage& storage) { return asNumber(storage, aliveEntryCountIndex()); }
    ALWAYS_INLINE static constexpr TableSize deletedEntryCount(Storage& storage) { return asNumber(storage, deletedEntryCountIndex()); }
    ALWAYS_INLINE static constexpr TableSize usedCapacity(Storage& storage) { return aliveEntryCount(storage) + deletedEntryCount(storage); }
    ALWAYS_INLINE static constexpr TableSize capacity(Storage& storage) { return asNumber(storage, capacityIndex()); }
    ALWAYS_INLINE static constexpr TableSize iterationEntry(Storage& storage) { return asNumber(storage, iterationEntryIndex()); }

    ALWAYS_INLINE static constexpr void incrementAliveEntryCount(Storage& storage) { OrderedHashTableTraits::increment(slot(storage, aliveEntryCountIndex())); }
    ALWAYS_INLINE static constexpr void decrementAliveEntryCount(Storage& storage) { OrderedHashTableTraits::decrement(slot(storage, aliveEntryCountIndex())); }
    ALWAYS_INLINE static constexpr void incrementDeletedEntryCount(Storage& storage) { OrderedHashTableTraits::increment(slot(storage, deletedEntryCountIndex())); }

    /* -------------------------------- Hash table -------------------------------- */
    ALWAYS_INLINE static constexpr TableSize bucketCount(TableSize capacity) { return capacity / LoadFactor; }

    ALWAYS_INLINE static constexpr TableIndex hashTableStartIndex() { return iterationEntryIndex() + 1; }
    ALWAYS_INLINE static constexpr TableIndex hashTableEndIndex(TableSize capacity) { return hashTableStartIndex() + bucketCount(capacity) - 1; }

    ALWAYS_INLINE static constexpr TableIndex bucketIndex(TableIndex hashTableStartIndex, TableSize bucketCount, TableSize hash) { return hashTableStartIndex + (hash & (bucketCount - 1)); }
    ALWAYS_INLINE static constexpr TableIndex bucketIndex(TableSize capacity, uint32_t hash) { return bucketIndex(hashTableStartIndex(), bucketCount(capacity), hash); }

    /* -------------------------------- Data table -------------------------------- */
    ALWAYS_INLINE static constexpr TableSize dataTableSize(TableSize capacity) { return capacity * EntrySize; }

    ALWAYS_INLINE static constexpr TableIndex dataTableStartIndex(TableSize capacity) { return hashTableEndIndex(capacity) + 1; }
    ALWAYS_INLINE static constexpr TableIndex dataTableEndIndex(TableSize capacity) { return dataTableStartIndex(capacity) + dataTableSize(capacity) - 1; }
    ALWAYS_INLINE static constexpr TableIndex entryDataStartIndex(TableIndex dataTableStartIndex, Entry entry) { return dataTableStartIndex + entry * EntrySize; }

    ALWAYS_INLINE static void setKeyOrValueData(VM& vm, Storage& storage, TableIndex index, JSValue value) { setWithWriteBarrier(vm, storage, index, value); }
    ALWAYS_INLINE static constexpr void deleteData(VM& vm, Storage& storage, TableIndex index) { return set(storage, index, vm.orderedHashTableDeletedValue()); }

    ALWAYS_INLINE static void addToChain(Storage& storage, TableIndex bucketIndex, TableIndex newChainStartKeyIndex)
    {
        JSValue prevChainStartKeyIndex = get(storage, bucketIndex);
        set(storage, bucketIndex, newChainStartKeyIndex);
        set(storage, newChainStartKeyIndex + ChainOffset, prevChainStartKeyIndex);
    }

    /* -------------------------------- OrderedHashTable -------------------------------- */
    ALWAYS_INLINE static constexpr TableSize tableSize(TableSize capacity)
    {
        TableSize result = 4 /* AliveEntryCount, DeletedEntryCount, Capacity, and IterationEntry */
            + capacity / LoadFactor /* BucketCount */
            + capacity * EntrySize; /* DataTableSize */
        ASSERT(result == tableSizeSlow(capacity));
        return result;
    }
    ALWAYS_INLINE static constexpr TableSize tableSizeSlow(TableSize capacity) { return dataTableEndIndex(capacity) + 1; }

    /* -------------------------------- Obsolete table -------------------------------- */
    ALWAYS_INLINE static constexpr bool isObsolete(Storage& storage) { return !!nextTable(storage); }
    ALWAYS_INLINE static constexpr Storage* nextTable(Storage& storage)
    {
        JSValue* value = slot(storage, aliveEntryCountIndex());
        if (value->isCell()) {
            ASSERT(jsDynamicCast<Storage*>(*value));
            return jsCast<Storage*>(*value);
        }
        return nullptr;
    }
    ALWAYS_INLINE static constexpr void setNextTable(VM& vm, Storage& storage, Storage* next) { setWithWriteBarrier(vm, storage, aliveEntryCountIndex(), next); }

    ALWAYS_INLINE static constexpr TableIndex deletedEntriesStartIndex() { return capacityIndex(); }

    ALWAYS_INLINE static constexpr bool isCleared(Storage& storage) { return asNumber(storage, deletedEntryCountIndex()) == ClearedTableSentinel; }
    ALWAYS_INLINE static constexpr void setClearedTableSentinel(Storage& storage) { set(storage, deletedEntryCountIndex(), ClearedTableSentinel); }

    /* ------------------------------------------------------------------------------------ */

    ALWAYS_INLINE static Storage* tryCreate(VM& vm, int length)
    {
        return Storage::tryCreate(vm, vm.immutableButterflyStructure(CopyOnWriteArrayWithContiguous), length);
    }
    ALWAYS_INLINE static Storage* tryCreate(JSGlobalObject* globalObject, TableSize aliveEntryCount = 0, TableSize deletedEntryCount = 0, TableSize capacity = InitialCapacity)
    {
        VM& vm = getVM(globalObject);
        auto scope = DECLARE_THROW_SCOPE(vm);
        ASSERT(!(capacity & (capacity - 1)));

        TableSize length = tableSize(capacity);
        if (UNLIKELY(length > IndexingHeader::maximumLength)) {
            throwOutOfMemoryError(globalObject, scope);
            return nullptr;
        }

        Storage* storage = tryCreate(vm, length);
        if (UNLIKELY(!storage)) {
            throwOutOfMemoryError(globalObject, scope);
            return nullptr;
        }

        Storage& storageRef = *storage;
        set(storageRef, aliveEntryCountIndex(), aliveEntryCount);
        set(storageRef, deletedEntryCountIndex(), deletedEntryCount);
        set(storageRef, capacityIndex(), capacity);
        return storage;
    }

    enum class UpdateDeletedEntries : uint8_t {
        Yes,
        No
    };
    template<UpdateDeletedEntries update = UpdateDeletedEntries::No>
    ALWAYS_INLINE static Storage* copyImpl(JSGlobalObject* globalObject, Storage& base, TableSize newCapacity)
    {
        VM& vm = getVM(globalObject);
        auto scope = DECLARE_THROW_SCOPE(vm);

        TableSize baseCapacity = capacity(base);
        TableSize baseAliveEntryCount = aliveEntryCount(base);
        TableSize baseUsedCapacity = usedCapacity(base);
        ASSERT(!isObsolete(base));
        ASSERT_UNUSED(baseAliveEntryCount, newCapacity >= std::max(static_cast<TableSize>(InitialCapacity), baseAliveEntryCount));
        ASSERT_UNUSED(baseCapacity, baseUsedCapacity <= baseCapacity);

        Storage* copy = tryCreate(globalObject, baseAliveEntryCount, 0, newCapacity);
        RETURN_IF_EXCEPTION(scope, nullptr);

        TableIndex baseEntryKeyIndex = dataTableStartIndex(baseCapacity) - EntrySize;
        TableIndex baseDeletedEntriesIndex = deletedEntriesStartIndex();

        Storage& copyRef = *copy;
        TableIndex newEntryKeyIndex = dataTableStartIndex(newCapacity) - EntrySize;
        TableIndex newHashTableStartIndex = hashTableStartIndex();
        TableIndex newBucketCount = bucketCount(newCapacity);

        for (Entry baseEntry = 0; baseEntry < baseUsedCapacity; ++baseEntry) {
            baseEntryKeyIndex += EntrySize;
            JSValue baseKey = get(base, baseEntryKeyIndex);
            ASSERT(!baseKey.isEmpty());

            // Step 1: Copy DataTable only for the alive entries.
            if (isDeleted(vm, baseKey)) {
                if constexpr (update == UpdateDeletedEntries::Yes)
                    set(base, baseDeletedEntriesIndex++, baseEntry);
                continue;
            }

            // Step 2: Copy the key and value from the base table to the new table.
            newEntryKeyIndex += EntrySize;
            setKeyOrValueData(vm, copyRef, newEntryKeyIndex, baseKey);
            if constexpr (Traits::hasValueData) {
                JSValue baseValue = get(base, baseEntryKeyIndex + 1);
                setKeyOrValueData(vm, copyRef, newEntryKeyIndex + 1, baseValue);
            }

            // Step 3: Compute for the hash value and add to the chain in the new table. Note that the
            // key stored in the base table is already normalized.
            TableSize hash = jsMapHash(globalObject, vm, baseKey);
            RETURN_IF_EXCEPTION(scope, nullptr);
            TableIndex newBucketIndex = bucketIndex(newHashTableStartIndex, newBucketCount, hash);
            addToChain(copyRef, newBucketIndex, newEntryKeyIndex);
        }

        return copy;
    }
    ALWAYS_INLINE static Storage* copy(JSGlobalObject* globalObject, Storage& base)
    {
        return copyImpl<>(globalObject, base, capacity(base));
    }

    enum class UpdateOwnerStorage : uint8_t {
        Yes,
        No
    };
    template<UpdateOwnerStorage update>
    ALWAYS_INLINE static Storage* rehash(JSGlobalObject* globalObject, HashTable* owner, Storage& base, TableSize newCapacity)
    {
        VM& vm = getVM(globalObject);
        auto scope = DECLARE_THROW_SCOPE(vm);

        Storage* copy = copyImpl<UpdateDeletedEntries::Yes>(globalObject, base, newCapacity);
        RETURN_IF_EXCEPTION(scope, nullptr);

        setNextTable(vm, base, copy);
        if constexpr (update == UpdateOwnerStorage::Yes)
            owner->m_storage.set(vm, owner, copy);
        return copy;
    }

    ALWAYS_INLINE static void clear(JSGlobalObject* globalObject, HashTable* owner, Storage& base)
    {
        VM& vm = getVM(globalObject);
        auto scope = DECLARE_THROW_SCOPE(vm);

        Storage* next = tryCreate(globalObject);
        RETURN_IF_EXCEPTION(scope, void());

        setClearedTableSentinel(base);
        setNextTable(vm, base, next);
        owner->m_storage.set(vm, owner, next);
    }

    struct FindResult {
        JSValue normalizedKey;
        uint32_t hash;
        TableIndex bucketIndex;
        TableIndex entryKeyIndex;
        // The keyIndex and keySlot may be redundant here. Let's leave it for now
        // since the following OpenTable patch can get rid of most of index related info for us.
        JSValue* entryKeySlot;
    };
    ALWAYS_INLINE static FindResult find(JSGlobalObject* globalObject, Storage& storage, JSValue key)
    {
        VM& vm = getVM(globalObject);
        auto scope = DECLARE_THROW_SCOPE(vm);
        ASSERT(!isObsolete(storage));

        if (!aliveEntryCount(storage))
            return { JSValue(), 0, InvalidTableIndex, InvalidTableIndex, nullptr };

        key = normalizeMapKey(key);
        TableSize hash = jsMapHash(globalObject, vm, key);
        RETURN_IF_EXCEPTION(scope, { });
        return find(globalObject, storage, key, hash);
    }
    ALWAYS_INLINE static FindResult find(JSGlobalObject* globalObject, Storage& storage, JSValue normalizedKey, TableSize hash)
    {
        VM& vm = getVM(globalObject);
        TableIndex bucketIndex = Helper::bucketIndex(capacity(storage), hash);
        ASSERT(!isObsolete(storage) && normalizeMapKey(normalizedKey) == normalizedKey);

        JSValue keyIndexValue = get(storage, bucketIndex);
        while (!keyIndexValue.isEmpty()) {
            TableIndex entryKeyIndex = toNumber(keyIndexValue);
            JSValue* entryKeySlot = slot(storage, entryKeyIndex);
            // Fixme: Maybe we can compress the searching path by updating the chain with non-deleted entry.
            if (!isDeleted(vm, *entryKeySlot) && areKeysEqual(globalObject, normalizedKey, *entryKeySlot))
                return { normalizedKey, hash, bucketIndex, entryKeyIndex, entryKeySlot };
            keyIndexValue = get(storage, entryKeyIndex + ChainOffset);
        }
        return { normalizedKey, hash, bucketIndex, InvalidTableIndex, nullptr };
    }

    ALWAYS_INLINE static Storage* expandIfNeeded(JSGlobalObject* globalObject, HashTable* owner, Storage& base)
    {
        ASSERT(!isObsolete(base));
        TableSize capacity = Helper::capacity(base);
        TableSize deletedEntryCount = Helper::deletedEntryCount(base);
        TableSize usedCapacity = Helper::aliveEntryCount(base) + deletedEntryCount;

        bool isSmallCapacity = capacity < LargeCapacity;
        TableSize expandFactor = isSmallCapacity ? 2 : 1;

        if (isSmallCapacity) {
            if (usedCapacity < (capacity >> 1))
                return &base;
        } else {
            if (usedCapacity < ((capacity >> 2) * 3))
                return &base;
        }

        TableSize newCapacity = Checked<TableSize>(capacity) << expandFactor;
        if (deletedEntryCount >= (capacity >> 1))
            newCapacity = capacity; // No need to expanded. Just clear the deleted entries.
        return rehash<UpdateOwnerStorage::No>(globalObject, owner, base, newCapacity);
    }
    template<typename FindKeyFunctor>
    ALWAYS_INLINE static void addImpl(JSGlobalObject* globalObject, HashTable* owner, Storage& base, JSValue key, JSValue value, const FindKeyFunctor& findKeyFunctor)
    {
        VM& vm = getVM(globalObject);
        auto scope = DECLARE_THROW_SCOPE(vm);
        ASSERT(!isObsolete(base));

        auto result = findKeyFunctor();
        RETURN_IF_EXCEPTION(scope, void());

        if (isValidTableIndex(result.entryKeyIndex)) {
            if constexpr (Traits::hasValueData)
                setKeyOrValueData(vm, base, result.entryKeyIndex + 1, value);
            return;
        }

        scope.release();
        addImpl(globalObject, owner, base, key, value, result);
    }
    ALWAYS_INLINE static void addImpl(JSGlobalObject* globalObject, HashTable* owner, Storage& base, JSValue key, JSValue value, FindResult& result)
    {
        VM& vm = getVM(globalObject);
        auto scope = DECLARE_THROW_SCOPE(vm);
        ASSERT(!isObsolete(base));

        ASSERT(!isValidTableIndex(result.entryKeyIndex));

        Storage* candidate = expandIfNeeded(globalObject, owner, base);
        RETURN_IF_EXCEPTION(scope, void());

        bool firstAliveEntry = result.normalizedKey.isEmpty();
        if (UNLIKELY(firstAliveEntry)) {
            result.normalizedKey = normalizeMapKey(key);
            result.hash = jsMapHash(globalObject, vm, result.normalizedKey);
            RETURN_IF_EXCEPTION(scope, void());
        }

        Storage& candidateRef = *candidate;
        TableSize capacity = Helper::capacity(candidateRef);
        Entry newEntry = usedCapacity(candidateRef);
        TableIndex newEntryKeyIndex = entryDataStartIndex(dataTableStartIndex(capacity), newEntry);
        incrementAliveEntryCount(candidateRef);

        bool rehashed = &base != candidate;
        if (UNLIKELY(rehashed || firstAliveEntry))
            result.bucketIndex = bucketIndex(capacity, result.hash);

        addToChain(candidateRef, result.bucketIndex, newEntryKeyIndex);
        setKeyOrValueData(vm, candidateRef, newEntryKeyIndex, result.normalizedKey);
        if constexpr (Traits::hasValueData)
            setKeyOrValueData(vm, candidateRef, newEntryKeyIndex + 1, value);

        if (UNLIKELY(rehashed))
            owner->m_storage.set(vm, owner, candidate);
    }
    ALWAYS_INLINE static void add(JSGlobalObject* globalObject, HashTable* owner, Storage& storage, JSValue key, JSValue value)
    {
        addImpl(globalObject, owner, storage, key, value, [&]() ALWAYS_INLINE_LAMBDA {
            return find(globalObject, storage, key);
        });
    }
    ALWAYS_INLINE static void addNormalized(JSGlobalObject* globalObject, HashTable* owner, Storage& storage, JSValue normalizedKey, JSValue value, TableSize hash)
    {
        ASSERT(normalizeMapKey(normalizedKey) == normalizedKey);
        addImpl(globalObject, owner, storage, normalizedKey, value, [&]() ALWAYS_INLINE_LAMBDA {
            return find(globalObject, storage, normalizedKey, hash);
        });
    }

    ALWAYS_INLINE static void shrinkIfNeeded(JSGlobalObject* globalObject, HashTable* owner, Storage& base)
    {
        ASSERT(!isObsolete(base));
        TableSize capacity = Helper::capacity(base);
        TableSize aliveEntryCount = Helper::aliveEntryCount(base);
        if (aliveEntryCount >= (capacity >> 3))
            return;
        if (capacity == InitialCapacity)
            return;
        rehash<UpdateOwnerStorage::Yes>(globalObject, owner, base, capacity >> 1);
    }
    template<typename FindKeyFunctor>
    ALWAYS_INLINE static bool removeImpl(JSGlobalObject* globalObject, HashTable* owner, Storage& storage, const FindKeyFunctor& findKeyFunctor)
    {
        VM& vm = getVM(globalObject);
        auto scope = DECLARE_THROW_SCOPE(vm);
        ASSERT(!isObsolete(storage));

        auto result = findKeyFunctor();
        RETURN_IF_EXCEPTION(scope, false);

        if (!isValidTableIndex(result.entryKeyIndex))
            return false;

        deleteData(vm, storage, result.entryKeyIndex);
        if constexpr (Traits::hasValueData)
            deleteData(vm, storage, result.entryKeyIndex + 1);
        incrementDeletedEntryCount(storage);
        decrementAliveEntryCount(storage);

        scope.release();
        shrinkIfNeeded(globalObject, owner, storage);
        return true;
    }
    ALWAYS_INLINE static bool remove(JSGlobalObject* globalObject, HashTable* owner, Storage& storage, JSValue key)
    {
        return removeImpl(globalObject, owner, storage, [&]() ALWAYS_INLINE_LAMBDA {
            return find(globalObject, storage, key);
        });
    }
    ALWAYS_INLINE static bool removeNormalized(JSGlobalObject* globalObject, HashTable* owner, Storage& storage, JSValue normalizedKey, TableSize hash)
    {
        ASSERT(normalizeMapKey(normalizedKey) == normalizedKey);
        return removeImpl(globalObject, owner, storage, [&]() ALWAYS_INLINE_LAMBDA {
            return find(globalObject, storage, normalizedKey, hash);
        });
    }

    template<typename Functor>
    ALWAYS_INLINE static Storage& transit(Storage& storage, const Functor& functor)
    {
        Storage* ptr = &storage;
        while (isObsolete(*ptr)) {
            functor(*ptr);
            ptr = nextTable(*ptr);
        }
        return *ptr;
    }
    struct TransitionResult {
        Storage* storage;
        Entry entry;
        JSValue key;
        JSValue value;
    };
    ALWAYS_INLINE static TransitionResult transitAndNext(VM& vm, Storage& storage, Entry from)
    {
        Storage& candidate = transit(storage, [&](Storage& obsolete) ALWAYS_INLINE_LAMBDA {
            if (!from)
                return;

            if (isCleared(obsolete)) {
                from = 0;
                return;
            }

            TableIndex deletedEntryCount = Helper::deletedEntryCount(obsolete);
            if (!deletedEntryCount)
                return;

            TableIndex start = deletedEntriesStartIndex();
            TableIndex end = start + deletedEntryCount;
            Entry fromCopy = from;
            for (TableIndex i = start; i < end; ++i) {
                Entry deletedEntry = toNumber(get(obsolete, i));
                if (deletedEntry >= fromCopy)
                    break;
                --from;
            }
        });

        ASSERT(!isObsolete(candidate));
        TableSize capacity = Helper::capacity(candidate);
        TableSize usedCapacity = Helper::usedCapacity(candidate);
        TableIndex entryKeyIndex = entryDataStartIndex(dataTableStartIndex(capacity), from) - EntrySize;
        for (Entry entry = from; entry < usedCapacity; ++entry) {
            entryKeyIndex += EntrySize;
            JSValue key = get(candidate, entryKeyIndex);
            if (isDeleted(vm, key))
                continue;

            JSValue value;
            if constexpr (Traits::hasValueData)
                value = get(candidate, entryKeyIndex + 1);
            return { &candidate, entry, key, value };
        }
        return { };
    }

    ALWAYS_INLINE static JSValue getKey(Storage& storage, Entry entry)
    {
        ASSERT(!isObsolete(storage));
        return get(storage, entryDataStartIndex(dataTableStartIndex(capacity(storage)), entry));
    }
    ALWAYS_INLINE static JSValue getValue(Storage& storage, Entry entry)
    {
        ASSERT(!isObsolete(storage) && EntrySize == 3);
        return get(storage, entryDataStartIndex(dataTableStartIndex(capacity(storage)), entry) + 1);
    }
    ALWAYS_INLINE static JSCell* nextAndUpdateIterationEntry(VM& vm, Storage& storage, Entry entry)
    {
        auto result = transitAndNext(vm, storage, entry);
        if (!result.storage)
            return vm.orderedHashTableSentinel();
        set(*result.storage, iterationEntryIndex(), result.entry);
        return result.storage;
    }
    ALWAYS_INLINE static JSValue getIterationEntry(Storage& storage) { return toJSValue(iterationEntry(storage)); }
    ALWAYS_INLINE static JSValue getIterationEntryKey(Storage& storage) { return getKey(storage, iterationEntry(storage)); }
    ALWAYS_INLINE static JSValue getIterationEntryValue(Storage& storage) { return getValue(storage, iterationEntry(storage)); }
};

} // namespace JSC

WTF_ALLOW_UNSAFE_BUFFER_USAGE_END