1 //===-- Support/FoldingSet.cpp - Uniquing Hash Set --------------*- C++ -*-===//
3 // The LLVM Compiler Infrastructure
5 // This file is distributed under the University of Illinois Open Source
6 // License. See LICENSE.TXT for details.
8 //===----------------------------------------------------------------------===//
10 // This file implements a hash set that can be used to remove duplication of
11 // nodes in a graph. This code was originally created by Chris Lattner for use
12 // with SelectionDAGCSEMap, but was isolated to provide use across the llvm code
15 //===----------------------------------------------------------------------===//
17 #include "llvm/ADT/FoldingSet.h"
18 #include "llvm/Support/MathExtras.h"
23 //===----------------------------------------------------------------------===//
24 // FoldingSetNodeID Implementation
26 /// Add* - Add various data types to Bit data.
28 void FoldingSetNodeID::AddPointer(const void *Ptr) {
29 // Note: this adds pointers to the hash using sizes and endianness that
30 // depend on the host. It doesn't matter however, because hashing on
31 // pointer values in inherently unstable. Nothing should depend on the
32 // ordering of nodes in the folding set.
33 intptr_t PtrI = (intptr_t)Ptr;
34 Bits.push_back(unsigned(PtrI));
35 if (sizeof(intptr_t) > sizeof(unsigned))
36 Bits.push_back(unsigned(uint64_t(PtrI) >> 32));
38 void FoldingSetNodeID::AddInteger(signed I) {
41 void FoldingSetNodeID::AddInteger(unsigned I) {
44 void FoldingSetNodeID::AddInteger(long I) {
45 AddInteger((unsigned long)I);
47 void FoldingSetNodeID::AddInteger(unsigned long I) {
48 if (sizeof(long) == sizeof(int))
49 AddInteger(unsigned(I));
50 else if (sizeof(long) == sizeof(long long)) {
51 AddInteger((unsigned long long)I);
53 assert(0 && "unexpected sizeof(long)");
56 void FoldingSetNodeID::AddInteger(long long I) {
57 AddInteger((unsigned long long)I);
59 void FoldingSetNodeID::AddInteger(unsigned long long I) {
60 AddInteger(unsigned(I));
61 if ((uint64_t)(int)I != I)
62 Bits.push_back(unsigned(I >> 32));
65 void FoldingSetNodeID::AddString(const char *String) {
66 unsigned Size = static_cast<unsigned>(strlen(String));
70 unsigned Units = Size / 4;
72 const unsigned *Base = (const unsigned *)String;
74 // If the string is aligned do a bulk transfer.
75 if (!((intptr_t)Base & 3)) {
76 Bits.append(Base, Base + Units);
77 Pos = (Units + 1) * 4;
79 // Otherwise do it the hard way.
80 for ( Pos += 4; Pos <= Size; Pos += 4) {
81 unsigned V = ((unsigned char)String[Pos - 4] << 24) |
82 ((unsigned char)String[Pos - 3] << 16) |
83 ((unsigned char)String[Pos - 2] << 8) |
84 (unsigned char)String[Pos - 1];
89 // With the leftover bits.
91 // Pos will have overshot size by 4 - #bytes left over.
93 case 1: V = (V << 8) | (unsigned char)String[Size - 3]; // Fall thru.
94 case 2: V = (V << 8) | (unsigned char)String[Size - 2]; // Fall thru.
95 case 3: V = (V << 8) | (unsigned char)String[Size - 1]; break;
96 default: return; // Nothing left.
102 void FoldingSetNodeID::AddString(const std::string &String) {
103 unsigned Size = static_cast<unsigned>(String.size());
104 Bits.push_back(Size);
107 unsigned Units = Size / 4;
109 const unsigned *Base = (const unsigned *)String.data();
111 // If the string is aligned do a bulk transfer.
112 if (!((intptr_t)Base & 3)) {
113 Bits.append(Base, Base + Units);
114 Pos = (Units + 1) * 4;
116 // Otherwise do it the hard way.
117 for ( Pos += 4; Pos <= Size; Pos += 4) {
118 unsigned V = ((unsigned char)String[Pos - 4] << 24) |
119 ((unsigned char)String[Pos - 3] << 16) |
120 ((unsigned char)String[Pos - 2] << 8) |
121 (unsigned char)String[Pos - 1];
126 // With the leftover bits.
128 // Pos will have overshot size by 4 - #bytes left over.
129 switch (Pos - Size) {
130 case 1: V = (V << 8) | (unsigned char)String[Size - 3]; // Fall thru.
131 case 2: V = (V << 8) | (unsigned char)String[Size - 2]; // Fall thru.
132 case 3: V = (V << 8) | (unsigned char)String[Size - 1]; break;
133 default: return; // Nothing left.
139 /// ComputeHash - Compute a strong hash value for this FoldingSetNodeID, used to
140 /// lookup the node in the FoldingSetImpl.
141 unsigned FoldingSetNodeID::ComputeHash() const {
142 // This is adapted from SuperFastHash by Paul Hsieh.
143 unsigned Hash = static_cast<unsigned>(Bits.size());
144 for (const unsigned *BP = &Bits[0], *E = BP+Bits.size(); BP != E; ++BP) {
146 Hash += Data & 0xFFFF;
147 unsigned Tmp = ((Data >> 16) << 11) ^ Hash;
148 Hash = (Hash << 16) ^ Tmp;
152 // Force "avalanching" of final 127 bits.
162 /// operator== - Used to compare two nodes to each other.
164 bool FoldingSetNodeID::operator==(const FoldingSetNodeID &RHS)const{
165 if (Bits.size() != RHS.Bits.size()) return false;
166 return memcmp(&Bits[0], &RHS.Bits[0], Bits.size()*sizeof(Bits[0])) == 0;
170 //===----------------------------------------------------------------------===//
171 /// Helper functions for FoldingSetImpl.
173 /// GetNextPtr - In order to save space, each bucket is a
174 /// singly-linked-list. In order to make deletion more efficient, we make
175 /// the list circular, so we can delete a node without computing its hash.
176 /// The problem with this is that the start of the hash buckets are not
177 /// Nodes. If NextInBucketPtr is a bucket pointer, this method returns null:
178 /// use GetBucketPtr when this happens.
179 static FoldingSetImpl::Node *GetNextPtr(void *NextInBucketPtr) {
180 // The low bit is set if this is the pointer back to the bucket.
181 if (reinterpret_cast<intptr_t>(NextInBucketPtr) & 1)
184 return static_cast<FoldingSetImpl::Node*>(NextInBucketPtr);
189 static void **GetBucketPtr(void *NextInBucketPtr) {
190 intptr_t Ptr = reinterpret_cast<intptr_t>(NextInBucketPtr);
191 assert((Ptr & 1) && "Not a bucket pointer");
192 return reinterpret_cast<void**>(Ptr & ~intptr_t(1));
195 /// GetBucketFor - Hash the specified node ID and return the hash bucket for
196 /// the specified ID.
197 static void **GetBucketFor(const FoldingSetNodeID &ID,
198 void **Buckets, unsigned NumBuckets) {
199 // NumBuckets is always a power of 2.
200 unsigned BucketNum = ID.ComputeHash() & (NumBuckets-1);
201 return Buckets + BucketNum;
204 //===----------------------------------------------------------------------===//
205 // FoldingSetImpl Implementation
207 FoldingSetImpl::FoldingSetImpl(unsigned Log2InitSize) {
208 assert(5 < Log2InitSize && Log2InitSize < 32 &&
209 "Initial hash table size out of range");
210 NumBuckets = 1 << Log2InitSize;
211 Buckets = new void*[NumBuckets+1];
214 FoldingSetImpl::~FoldingSetImpl() {
217 void FoldingSetImpl::clear() {
218 // Set all but the last bucket to null pointers.
219 memset(Buckets, 0, NumBuckets*sizeof(void*));
221 // Set the very last bucket to be a non-null "pointer".
222 Buckets[NumBuckets] = reinterpret_cast<void*>(-1);
224 // Reset the node count to zero.
228 /// GrowHashTable - Double the size of the hash table and rehash everything.
230 void FoldingSetImpl::GrowHashTable() {
231 void **OldBuckets = Buckets;
232 unsigned OldNumBuckets = NumBuckets;
235 // Clear out new buckets.
236 Buckets = new void*[NumBuckets+1];
239 // Walk the old buckets, rehashing nodes into their new place.
241 for (unsigned i = 0; i != OldNumBuckets; ++i) {
242 void *Probe = OldBuckets[i];
243 if (!Probe) continue;
244 while (Node *NodeInBucket = GetNextPtr(Probe)) {
245 // Figure out the next link, remove NodeInBucket from the old link.
246 Probe = NodeInBucket->getNextInBucket();
247 NodeInBucket->SetNextInBucket(0);
249 // Insert the node into the new bucket, after recomputing the hash.
250 GetNodeProfile(ID, NodeInBucket);
251 InsertNode(NodeInBucket, GetBucketFor(ID, Buckets, NumBuckets));
259 /// FindNodeOrInsertPos - Look up the node specified by ID. If it exists,
260 /// return it. If not, return the insertion token that will make insertion
263 *FoldingSetImpl::FindNodeOrInsertPos(const FoldingSetNodeID &ID,
266 void **Bucket = GetBucketFor(ID, Buckets, NumBuckets);
267 void *Probe = *Bucket;
271 FoldingSetNodeID OtherID;
272 while (Node *NodeInBucket = GetNextPtr(Probe)) {
273 GetNodeProfile(OtherID, NodeInBucket);
277 Probe = NodeInBucket->getNextInBucket();
281 // Didn't find the node, return null with the bucket as the InsertPos.
286 /// InsertNode - Insert the specified node into the folding set, knowing that it
287 /// is not already in the map. InsertPos must be obtained from
288 /// FindNodeOrInsertPos.
289 void FoldingSetImpl::InsertNode(Node *N, void *InsertPos) {
290 assert(N->getNextInBucket() == 0);
291 // Do we need to grow the hashtable?
292 if (NumNodes+1 > NumBuckets*2) {
295 GetNodeProfile(ID, N);
296 InsertPos = GetBucketFor(ID, Buckets, NumBuckets);
301 /// The insert position is actually a bucket pointer.
302 void **Bucket = static_cast<void**>(InsertPos);
304 void *Next = *Bucket;
306 // If this is the first insertion into this bucket, its next pointer will be
307 // null. Pretend as if it pointed to itself, setting the low bit to indicate
308 // that it is a pointer to the bucket.
310 Next = reinterpret_cast<void*>(reinterpret_cast<intptr_t>(Bucket)|1);
312 // Set the node's next pointer, and make the bucket point to the node.
313 N->SetNextInBucket(Next);
317 /// RemoveNode - Remove a node from the folding set, returning true if one was
318 /// removed or false if the node was not in the folding set.
319 bool FoldingSetImpl::RemoveNode(Node *N) {
320 // Because each bucket is a circular list, we don't need to compute N's hash
322 void *Ptr = N->getNextInBucket();
323 if (Ptr == 0) return false; // Not in folding set.
326 N->SetNextInBucket(0);
328 // Remember what N originally pointed to, either a bucket or another node.
329 void *NodeNextPtr = Ptr;
331 // Chase around the list until we find the node (or bucket) which points to N.
333 if (Node *NodeInBucket = GetNextPtr(Ptr)) {
335 Ptr = NodeInBucket->getNextInBucket();
337 // We found a node that points to N, change it to point to N's next node,
338 // removing N from the list.
340 NodeInBucket->SetNextInBucket(NodeNextPtr);
344 void **Bucket = GetBucketPtr(Ptr);
347 // If we found that the bucket points to N, update the bucket to point to
350 *Bucket = NodeNextPtr;
357 /// GetOrInsertNode - If there is an existing simple Node exactly
358 /// equal to the specified node, return it. Otherwise, insert 'N' and it
360 FoldingSetImpl::Node *FoldingSetImpl::GetOrInsertNode(FoldingSetImpl::Node *N) {
362 GetNodeProfile(ID, N);
364 if (Node *E = FindNodeOrInsertPos(ID, IP))
370 //===----------------------------------------------------------------------===//
371 // FoldingSetIteratorImpl Implementation
373 FoldingSetIteratorImpl::FoldingSetIteratorImpl(void **Bucket) {
374 // Skip to the first non-null non-self-cycle bucket.
375 while (*Bucket != reinterpret_cast<void*>(-1) &&
376 (*Bucket == 0 || GetNextPtr(*Bucket) == 0))
379 NodePtr = static_cast<FoldingSetNode*>(*Bucket);
382 void FoldingSetIteratorImpl::advance() {
383 // If there is another link within this bucket, go to it.
384 void *Probe = NodePtr->getNextInBucket();
386 if (FoldingSetNode *NextNodeInBucket = GetNextPtr(Probe))
387 NodePtr = NextNodeInBucket;
389 // Otherwise, this is the last link in this bucket.
390 void **Bucket = GetBucketPtr(Probe);
392 // Skip to the next non-null non-self-cycle bucket.
395 } while (*Bucket != reinterpret_cast<void*>(-1) &&
396 (*Bucket == 0 || GetNextPtr(*Bucket) == 0));
398 NodePtr = static_cast<FoldingSetNode*>(*Bucket);
402 //===----------------------------------------------------------------------===//
403 // FoldingSetBucketIteratorImpl Implementation
405 FoldingSetBucketIteratorImpl::FoldingSetBucketIteratorImpl(void **Bucket) {
406 Ptr = (*Bucket == 0 || GetNextPtr(*Bucket) == 0) ? (void*) Bucket : *Bucket;