Improve distribution of V3Hash/V3Hasher

- Better combining, without multiplication, which means a 0 hash value
is now allowed.
- Do not OR in bottom bits (this was used to avoid a 0 hash but had the
side effect of hashing 0 and 1 to the same value, which are actually
common inputs.
- Hash whole content of V3Number. This does not seem to be noticeable in
runtime, but quite often the bottom word can be a special value like
zero while the rest of the content varies.
This commit is contained in:
Geza Lore 2021-06-09 21:03:32 +01:00
parent 5555f20bd2
commit 24b3816f5e
2 changed files with 30 additions and 21 deletions

View File

@ -24,40 +24,44 @@
// V3Hash -- Generic hashing
class V3Hash final {
// A 32-bit hash value. A value of 0 is illegal.
uint32_t m_value;
uint32_t m_value; // The 32-bit hash value.
public:
// CONSTRUCTORS
V3Hash()
: m_value{0} {}
explicit V3Hash(uint32_t val)
: m_value{val + 0x9e3779b9} {} // This is the same as 'V3Hash() + val'
explicit V3Hash(int32_t val)
: m_value{static_cast<uint32_t>(val)} {}
explicit V3Hash(size_t val)
: m_value{static_cast<uint32_t>(val)} {}
explicit V3Hash(const std::string& val);
// METHODS
uint32_t value() const { return m_value; }
// OPERATORS
// Comparisons
bool operator==(const V3Hash& rh) const { return m_value == rh.m_value; }
bool operator!=(const V3Hash& rh) const { return m_value != rh.m_value; }
bool operator<(const V3Hash& rh) const { return m_value < rh.m_value; }
V3Hash operator+(uint32_t value) const {
const uint64_t prod = (static_cast<uint64_t>(m_value) * 31) + value;
return V3Hash(static_cast<uint32_t>(prod ^ (prod >> 32)));
}
V3Hash operator+(int32_t value) const { return *this + static_cast<uint32_t>(value); }
V3Hash operator+(const V3Hash& that) const { return *this + that.m_value; }
V3Hash& operator+=(const V3Hash& that) {
*this = *this + that.m_value;
return *this;
// '+' combines hashes
V3Hash operator+(const V3Hash& that) const {
return V3Hash(m_value ^ (that.m_value + 0x9e3779b9 + (m_value << 6) + (m_value >> 2)));
}
V3Hash operator+(uint32_t value) const { return *this + V3Hash(value); }
V3Hash operator+(int32_t value) const { return *this + V3Hash(value); }
V3Hash operator+(size_t value) const { return *this + V3Hash(value); }
V3Hash operator+(const std::string& value) const { return *this + V3Hash(value); }
// '+=' combines in place
V3Hash& operator+=(const V3Hash& that) { return *this = *this + that; }
V3Hash& operator+=(uint32_t value) { return *this += V3Hash(value); }
V3Hash& operator+=(int32_t value) { return *this += V3Hash(value); }
V3Hash& operator+=(size_t value) { return *this += V3Hash(value); }
V3Hash& operator+=(const std::string& that) { return *this += V3Hash(that); }
// CONSTRUCTORS
V3Hash()
: m_value{1} {}
explicit V3Hash(uint32_t val)
: m_value{val | 1} {}
explicit V3Hash(int32_t val)
: m_value{static_cast<uint32_t>(val)} {}
explicit V3Hash(const std::string& val);
};
std::ostream& operator<<(std::ostream& os, const V3Hash& rhs);

View File

@ -24,6 +24,7 @@
#include <algorithm>
#include <cerrno>
#include <cmath>
#include <functional>
constexpr int MAX_SPRINTF_DOUBLE_SIZE
= 1100; // Maximum characters with a sprintf %e/%f/%g (really 1079)
@ -883,7 +884,11 @@ string V3Number::toString() const {
return str;
}
V3Hash V3Number::toHash() const { return V3Hash(m_width * (m_value[0] | 1)); }
V3Hash V3Number::toHash() const {
V3Hash hash(m_width);
for (int i = 0; i < words(); ++i) { hash += m_value[i]; }
return hash;
}
uint32_t V3Number::edataWord(int eword) const {
UASSERT(!isFourState(), "edataWord with 4-state " << *this);