SmallPtrSet.h

Go to the documentation of this file.

//===- llvm/ADT/SmallPtrSet.h - 'Normally small' pointer set ----*- C++ -*-===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
///
/// \file
/// This file defines the SmallPtrSet class.  See the doxygen comment for
/// SmallPtrSetImplBase for more details on the algorithm used.
//
//===----------------------------------------------------------------------===//
 
#ifndef LLVM_ADT_SMALLPTRSET_H
#define LLVM_ADT_SMALLPTRSET_H
 
#include "llvm/ADT/ADL.h"
#include "llvm/ADT/EpochTracker.h"
#include "llvm/ADT/STLForwardCompat.h"
#include "llvm/ADT/iterator_range.h"
#include "llvm/Support/Compiler.h"
#include "llvm/Support/MathExtras.h"
#include "llvm/Support/ReverseIteration.h"
#include "llvm/Support/type_traits.h"
#include <algorithm>
#include <cassert>
#include <cstddef>
#include <cstdlib>
#include <cstring>
#include <initializer_list>
#include <iterator>
#include <limits>
#include <utility>
 
namespace llvm {
 
template <typename PtrTy> class SmallPtrSetIterator;
 
/// SmallPtrSetImplBase - This is the common code shared among all the
/// SmallPtrSet<>'s, which is almost everything.  SmallPtrSet has two modes, one
/// for small and one for large sets.
///
/// Small sets use an array of pointers allocated in the SmallPtrSet object,
/// which is treated as a simple array of pointers.  When a pointer is added to
/// the set, the array is scanned to see if the element already exists, if not
/// the element is 'pushed back' onto the array.  If we run out of space in the
/// array, we grow into the 'large set' case.  SmallSet should be used when the
/// sets are often small.  In this case, no memory allocation is used, and only
/// light-weight and cache-efficient scanning is used.
///
/// Large sets use a linear-probed hash table with deletion implemented using
/// Knuth TAOCP 6.4 Algorithm R: `erase` opens a hole, walks forward sliding
/// each following entry whose probe path crosses the hole back into it (the
/// hole moves with each slide), and stops at the next empty slot.  Empty
/// buckets are represented with an illegal pointer value (-1) to allow null
/// pointers to be inserted; no tombstone state is needed.  The hash table is
/// resized when the table is 2/3 or more.  When this happens, the table is
/// doubled in size.
class SmallPtrSetImplBase : public DebugEpochBase {
  template <typename PtrTy> friend class SmallPtrSetIterator;
 
protected:
  /// The current set of buckets, in either small or big representation.
  const void **CurArray;
  /// CurArraySize - The allocated size of CurArray, always a power of two.
  unsigned CurArraySize;
 
  /// Number of elements in CurArray that contain a value.
  /// If small, all these elements are at the beginning of CurArray and the rest
  /// is uninitialized.
  unsigned NumEntries;
  /// Whether the set is in small representation.
  bool IsSmall;
 
  // Helpers to copy and move construct a SmallPtrSet.
  LLVM_ABI SmallPtrSetImplBase(const void **SmallStorage,
                               const SmallPtrSetImplBase &that);
  LLVM_ABI SmallPtrSetImplBase(const void **SmallStorage, unsigned SmallSize,
                               const void **RHSSmallStorage,
                               SmallPtrSetImplBase &&that);
 
  explicit SmallPtrSetImplBase(const void **SmallStorage, unsigned SmallSize)
      : CurArray(SmallStorage), CurArraySize(SmallSize), NumEntries(0),
        IsSmall(true) {
    assert(llvm::has_single_bit(SmallSize) &&
           "Initial size must be a power of two!");
  }
 
  ~SmallPtrSetImplBase() {
    if (!isSmall())
      free(CurArray);
  }
 
public:
  using size_type = unsigned;
 
  SmallPtrSetImplBase &operator=(const SmallPtrSetImplBase &) = delete;
 
  [[nodiscard]] bool empty() const { return size() == 0; }
  [[nodiscard]] size_type size() const { return NumEntries; }
  [[nodiscard]] size_type capacity() const { return CurArraySize; }
 
  void clear() {
    incrementEpoch();
    // If the capacity of the array is huge, and the # elements used is small,
    // shrink the array.
    if (!isSmall()) {
      if (size() * 4 < CurArraySize && CurArraySize > 32)
        return shrink_and_clear();
      // Fill the array with empty markers.
      memset(CurArray, -1, CurArraySize * sizeof(void *));
    }
 
    NumEntries = 0;
  }
 
  void reserve(size_type NewNumEntries) {
    incrementEpoch();
    // Do nothing if we're given zero as a reservation size.
    if (NewNumEntries == 0)
      return;
    // No need to expand if we're small and NewNumEntries will fit in the space.
    if (isSmall() && NewNumEntries <= CurArraySize)
      return;
    // insert_imp_big will reallocate if stores is more than 2/3 full, on the
    // /final/ insertion.
    if (!isSmall() && ((NewNumEntries - 1) * 3) < (CurArraySize * 2))
      return;
    // We must Grow -- find the size where we'd be 2/3 full, then round up to
    // the next power of two.
    size_type NewSize = NewNumEntries + (NewNumEntries / 2);
    NewSize = llvm::bit_ceil(NewSize);
    // Like insert_imp_big, always allocate at least 128 elements.
    NewSize = std::max(128u, NewSize);
    Grow(NewSize);
  }
 
protected:
  static void *getEmptyMarker() {
    // Note that -1 is chosen to make clear() efficiently implementable with
    // memset and because it's not a valid pointer value.
    return reinterpret_cast<void *>(-1);
  }
 
  const void **EndPointer() const {
    return isSmall() ? CurArray + NumEntries : CurArray + CurArraySize;
  }
 
  iterator_range<const void **> small_buckets() {
    return make_range(CurArray, CurArray + NumEntries);
  }
 
  iterator_range<const void *const *> small_buckets() const {
    return {CurArray, CurArray + NumEntries};
  }
 
  iterator_range<const void **> buckets() {
    return make_range(CurArray, EndPointer());
  }
 
  iterator_range<const void *const *> buckets() const {
    return make_range(CurArray, EndPointer());
  }
 
  /// insert_imp - This returns true if the pointer was new to the set, false if
  /// it was already in the set.  This is hidden from the client so that the
  /// derived class can check that the right type of pointer is passed in.
  std::pair<const void *const *, bool> insert_imp(const void *Ptr) {
    if (isSmall()) {
      // Check to see if it is already in the set.
      for (const void *&Bucket : small_buckets()) {
        if (Bucket == Ptr)
          return {&Bucket, false};
      }
 
      // Nope, there isn't.  If we stay small, just 'pushback' now.
      if (NumEntries < CurArraySize) {
        CurArray[NumEntries++] = Ptr;
        incrementEpoch();
        return {CurArray + (NumEntries - 1), true};
      }
      // Otherwise, hit the big set case, which will call grow.
    }
    return insert_imp_big(Ptr);
  }
 
  /// erase_imp - If the set contains the specified pointer, remove it and
  /// return true, otherwise return false.  This is hidden from the client so
  /// that the derived class can check that the right type of pointer is passed
  /// in.
  bool erase_imp(const void *Ptr) {
    if (isSmall()) {
      for (const void *&Bucket : small_buckets()) {
        if (Bucket == Ptr) {
          Bucket = CurArray[--NumEntries];
          incrementEpoch();
          return true;
        }
      }
      return false;
    }
 
    auto *Bucket = doFind(Ptr);
    if (!Bucket)
      return false;
 
    eraseFromBucket(const_cast<const void **>(Bucket));
    --NumEntries;
    incrementEpoch();
    return true;
  }
 
  /// Returns the raw pointer needed to construct an iterator.  If element not
  /// found, this will be EndPointer.  Otherwise, it will be a pointer to the
  /// slot which stores Ptr;
  const void *const *find_imp(const void *Ptr) const {
    if (isSmall()) {
      // Linear search for the item.
      for (const void *const &Bucket : small_buckets())
        if (Bucket == Ptr)
          return &Bucket;
      return EndPointer();
    }
 
    // Big set case.
    if (auto *Bucket = doFind(Ptr))
      return Bucket;
    return EndPointer();
  }
 
  bool contains_imp(const void *Ptr) const {
    if (isSmall()) {
      // Linear search for the item.
      for (const void *const &Bucket : small_buckets())
        if (Bucket == Ptr)
          return true;
      return false;
    }
 
    return doFind(Ptr) != nullptr;
  }
 
  bool isSmall() const { return IsSmall; }
 
private:
  LLVM_ABI std::pair<const void *const *, bool> insert_imp_big(const void *Ptr);
 
  LLVM_ABI const void *const *doFind(const void *Ptr) const;
  LLVM_ABI void shrink_and_clear();
 
protected:
  /// Erase the entry at \p Bucket and close the resulting hole via Knuth
  /// TAOCP 6.4 Algorithm R. Caller must update \c NumEntries and the epoch.
  LLVM_ABI void eraseFromBucket(const void **Bucket);
 
  /// Allocate a larger backing store for the buckets and move it over.
  /// Passing the current size triggers a same-size rehash, used by batch
  /// erase to compact away empty slots left by mark-then-rebuild.
  LLVM_ABI void Grow(unsigned NewSize);
 
  /// swap - Swaps the elements of two sets.
  /// Note: This method assumes that both sets have the same small size.
  LLVM_ABI void swap(const void **SmallStorage, const void **RHSSmallStorage,
                     SmallPtrSetImplBase &RHS);
 
  LLVM_ABI void copyFrom(const void **SmallStorage,
                         const SmallPtrSetImplBase &RHS);
  LLVM_ABI void moveFrom(const void **SmallStorage, unsigned SmallSize,
                         const void **RHSSmallStorage,
                         SmallPtrSetImplBase &&RHS);
 
private:
  /// Code shared by moveFrom() and move constructor.
  void moveHelper(const void **SmallStorage, unsigned SmallSize,
                  const void **RHSSmallStorage, SmallPtrSetImplBase &&RHS);
  /// Code shared by copyFrom() and copy constructor.
  void copyHelper(const SmallPtrSetImplBase &RHS);
};
 
/// This implements a const_iterator for SmallPtrSet.
template <typename PtrTy>
class LLVM_DEBUGEPOCHBASE_HANDLEBASE_EMPTYBASE SmallPtrSetIterator
    : public DebugEpochBase::HandleBase {
  using PtrTraits = PointerLikeTypeTraits<PtrTy>;
  using BucketItTy =
      std::conditional_t<shouldReverseIterate(),
                         std::reverse_iterator<const void *const *>,
                         const void *const *>;
 
  BucketItTy Bucket = {};
  BucketItTy End = {};
 
  /// AdvanceIfNotValid - If the current bucket isn't valid, advance to a bucket
  /// that is.   This is guaranteed to stop because the end() bucket is marked
  /// valid.
  void AdvanceIfNotValid() {
    assert(Bucket <= End);
    while (Bucket != End && *Bucket == SmallPtrSetImplBase::getEmptyMarker())
      ++Bucket;
  }
 
public:
  using value_type = PtrTy;
  using reference = PtrTy;
  using pointer = PtrTy;
  using difference_type = std::ptrdiff_t;
  using iterator_category = std::forward_iterator_tag;
 
  SmallPtrSetIterator() = default;
 
  SmallPtrSetIterator(const void *const *BP, const void *const *E,
                      const DebugEpochBase &Epoch)
      : DebugEpochBase::HandleBase(&Epoch), Bucket(BucketItTy(BP)),
        End(BucketItTy(E)) {
    AdvanceIfNotValid();
  }
 
  [[nodiscard]] const PtrTy operator*() const {
    assert(isHandleInSync() && "invalid iterator access!");
    assert(Bucket < End);
    return PtrTraits::getFromVoidPointer(const_cast<void *>(*Bucket));
  }
 
  inline SmallPtrSetIterator &operator++() { // Preincrement
    assert(isHandleInSync() && "invalid iterator access!");
    ++Bucket;
    AdvanceIfNotValid();
    return *this;
  }
 
  SmallPtrSetIterator operator++(int) { // Postincrement
    SmallPtrSetIterator tmp = *this;
    ++*this;
    return tmp;
  }
 
  bool operator==(const SmallPtrSetIterator &RHS) const {
    return Bucket == RHS.Bucket;
  }
  bool operator!=(const SmallPtrSetIterator &RHS) const {
    return Bucket != RHS.Bucket;
  }
};
 
/// A templated base class for \c SmallPtrSet which provides the
/// typesafe interface that is common across all small sizes.
///
/// This is particularly useful for passing around between interface boundaries
/// to avoid encoding a particular small size in the interface boundary.
template <typename PtrType> class SmallPtrSetImpl : public SmallPtrSetImplBase {
  using ConstPtrType = typename add_const_past_pointer<PtrType>::type;
  using PtrTraits = PointerLikeTypeTraits<PtrType>;
  using ConstPtrTraits = PointerLikeTypeTraits<ConstPtrType>;
 
protected:
  // Forward constructors to the base.
  using SmallPtrSetImplBase::SmallPtrSetImplBase;
 
public:
  using iterator = SmallPtrSetIterator<PtrType>;
  using const_iterator = SmallPtrSetIterator<PtrType>;
  using key_type = ConstPtrType;
  using value_type = PtrType;
 
  SmallPtrSetImpl(const SmallPtrSetImpl &) = delete;
 
  /// Inserts Ptr if and only if there is no element in the container equal to
  /// Ptr. The bool component of the returned pair is true if and only if the
  /// insertion takes place, and the iterator component of the pair points to
  /// the element equal to Ptr.
  std::pair<iterator, bool> insert(PtrType Ptr) {
    auto p = insert_imp(PtrTraits::getAsVoidPointer(Ptr));
    return {makeIterator(p.first), p.second};
  }
 
  /// Insert the given pointer with an iterator hint that is ignored. This is
  /// identical to calling insert(Ptr), but allows SmallPtrSet to be used by
  /// std::insert_iterator and std::inserter().
  iterator insert(iterator, PtrType Ptr) { return insert(Ptr).first; }
 
  /// Remove pointer from the set.
  ///
  /// Returns whether the pointer was in the set. Invalidates iterators if
  /// true is returned. To remove elements while iterating over the set, use
  /// remove_if() instead.
  bool erase(PtrType Ptr) {
    return erase_imp(PtrTraits::getAsVoidPointer(Ptr));
  }
 
  /// Remove elements that match the given predicate.
  ///
  /// This method is a safe replacement for the following pattern, which is not
  /// valid, because the erase() calls would invalidate the iterator:
  ///
  ///     for (PtrType *Ptr : Set)
  ///       if (Pred(P))
  ///         Set.erase(P);
  ///
  /// Returns whether anything was removed. The predicate must not access the
  /// set being modified: it may inspect the element passed to it and return
  /// true to request removal, but must not read (e.g. count()/find()) or
  /// otherwise mutate the set. If anything is removed, all iterators and
  /// references into the set are invalidated.
  template <typename UnaryPredicate> bool remove_if(UnaryPredicate P) {
    bool Removed = false;
    if (isSmall()) {
      auto Buckets = small_buckets();
      const void **APtr = Buckets.begin(), **E = Buckets.end();
      while (APtr != E) {
        PtrType Ptr = PtrTraits::getFromVoidPointer(const_cast<void *>(*APtr));
        if (P(Ptr)) {
          *APtr = *--E;
          --NumEntries;
          incrementEpoch();
          Removed = true;
        } else {
          ++APtr;
        }
      }
      return Removed;
    }
 
    // Mark-then-rebuild: one pass to clear matches without sliding (which
    // would re-walk the cluster on every erase), then a single rehash to
    // restore the linear-probe invariant.  O(N) total, vs O(N * cluster)
    // for repeated per-match Algorithm R erases.
    for (const void *&Bucket : buckets()) {
      if (Bucket == getEmptyMarker())
        continue;
      PtrType Ptr = PtrTraits::getFromVoidPointer(const_cast<void *>(Bucket));
      if (P(Ptr)) {
        Bucket = getEmptyMarker();
        --NumEntries;
        Removed = true;
      }
    }
    if (Removed) {
      incrementEpoch();
      Grow(CurArraySize);
    }
    return Removed;
  }
 
  /// count - Return 1 if the specified pointer is in the set, 0 otherwise.
  [[nodiscard]] size_type count(ConstPtrType Ptr) const {
    return contains_imp(ConstPtrTraits::getAsVoidPointer(Ptr));
  }
  [[nodiscard]] iterator find(ConstPtrType Ptr) const {
    return makeIterator(find_imp(ConstPtrTraits::getAsVoidPointer(Ptr)));
  }
  [[nodiscard]] bool contains(ConstPtrType Ptr) const {
    return contains_imp(ConstPtrTraits::getAsVoidPointer(Ptr));
  }
 
  template <typename IterT> void insert(IterT I, IterT E) {
    for (; I != E; ++I)
      insert(*I);
  }
 
  void insert(std::initializer_list<PtrType> IL) {
    insert(IL.begin(), IL.end());
  }
 
  template <typename Range> void insert_range(Range &&R) {
    insert(adl_begin(R), adl_end(R));
  }
 
  [[nodiscard]] iterator begin() const {
    if constexpr (shouldReverseIterate())
      return makeIterator(EndPointer() - 1);
    else
      return makeIterator(CurArray);
  }
  [[nodiscard]] iterator end() const { return makeIterator(EndPointer()); }
 
private:
  /// Create an iterator that dereferences to same place as the given pointer.
  iterator makeIterator(const void *const *P) const {
    if constexpr (shouldReverseIterate())
      return iterator(P == EndPointer() ? CurArray : P + 1, CurArray, *this);
    else
      return iterator(P, EndPointer(), *this);
  }
};
 
/// Equality comparison for SmallPtrSet.
///
/// Iterates over elements of LHS confirming that each value from LHS is also in
/// RHS, and that no additional values are in RHS.
template <typename PtrType>
[[nodiscard]] bool operator==(const SmallPtrSetImpl<PtrType> &LHS,
                              const SmallPtrSetImpl<PtrType> &RHS) {
  if (LHS.size() != RHS.size())
    return false;
 
  for (const auto *KV : LHS)
    if (!RHS.count(KV))
      return false;
 
  return true;
}
 
/// Inequality comparison for SmallPtrSet.
///
/// Equivalent to !(LHS == RHS).
template <typename PtrType>
[[nodiscard]] bool operator!=(const SmallPtrSetImpl<PtrType> &LHS,
                              const SmallPtrSetImpl<PtrType> &RHS) {
  return !(LHS == RHS);
}
 
/// SmallPtrSet - This class implements a set which is optimized for holding
/// SmallSize or less elements.  This internally rounds up SmallSize to the next
/// power of two if it is not already a power of two.  See the comments above
/// SmallPtrSetImplBase for details of the algorithm.
template <class PtrType, unsigned SmallSize>
class SmallPtrSet : public SmallPtrSetImpl<PtrType> {
  // In small mode SmallPtrSet uses linear search for the elements, so it is
  // not a good idea to choose this value too high. You may consider using a
  // DenseSet<> instead if you expect many elements in the set.
  static_assert(SmallSize <= 32, "SmallSize should be small");
 
  using BaseT = SmallPtrSetImpl<PtrType>;
 
  // Make sure that SmallSize is a power of two, round up if not.
  static constexpr size_t SmallSizePowTwo = llvm::bit_ceil_constexpr(SmallSize);
  /// SmallStorage - Fixed size storage used in 'small mode'.
  const void *SmallStorage[SmallSizePowTwo];
 
public:
  SmallPtrSet() : BaseT(SmallStorage, SmallSizePowTwo) {}
  SmallPtrSet(const SmallPtrSet &that) : BaseT(SmallStorage, that) {}
  SmallPtrSet(SmallPtrSet &&that)
      : BaseT(SmallStorage, SmallSizePowTwo, that.SmallStorage,
              std::move(that)) {}
 
  template <typename It>
  SmallPtrSet(It I, It E) : BaseT(SmallStorage, SmallSizePowTwo) {
    this->insert(I, E);
  }
 
  template <typename Range>
  SmallPtrSet(llvm::from_range_t, Range &&R)
      : SmallPtrSet(adl_begin(R), adl_end(R)) {}
 
  SmallPtrSet(std::initializer_list<PtrType> IL)
      : BaseT(SmallStorage, SmallSizePowTwo) {
    this->insert(IL.begin(), IL.end());
  }
 
  SmallPtrSet<PtrType, SmallSize> &
  operator=(const SmallPtrSet<PtrType, SmallSize> &RHS) {
    if (&RHS != this)
      this->copyFrom(SmallStorage, RHS);
    return *this;
  }
 
  SmallPtrSet<PtrType, SmallSize> &
  operator=(SmallPtrSet<PtrType, SmallSize> &&RHS) {
    if (&RHS != this)
      this->moveFrom(SmallStorage, SmallSizePowTwo, RHS.SmallStorage,
                     std::move(RHS));
    return *this;
  }
 
  SmallPtrSet<PtrType, SmallSize> &
  operator=(std::initializer_list<PtrType> IL) {
    this->clear();
    this->insert(IL.begin(), IL.end());
    return *this;
  }
 
  /// swap - Swaps the elements of two sets.
  void swap(SmallPtrSet<PtrType, SmallSize> &RHS) {
    SmallPtrSetImplBase::swap(SmallStorage, RHS.SmallStorage, RHS);
  }
};
 
} // namespace llvm
 
namespace std {
 
/// Implement std::swap in terms of SmallPtrSet swap.
template <class T, unsigned N>
inline void swap(llvm::SmallPtrSet<T, N> &LHS, llvm::SmallPtrSet<T, N> &RHS) {
  LHS.swap(RHS);
}
 
} // namespace std
 
#endif // LLVM_ADT_SMALLPTRSET_H