/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */ /* vim: set ts=8 sts=2 et sw=2 tw=80: */ /* This Source Code Form is subject to the terms of the Mozilla Public * License, v. 2.0. If a copy of the MPL was not distributed with this * file, You can obtain one at http://mozilla.org/MPL/2.0/. */ /* Various predicates and operations on IEEE-754 floating point types. */ #ifndef mozilla_FloatingPoint_h #define mozilla_FloatingPoint_h #include "mozilla/Assertions.h" #include "mozilla/Attributes.h" #include "mozilla/Casting.h" #include "mozilla/MathAlgorithms.h" #include "mozilla/Types.h" #include namespace mozilla { /* * It's reasonable to ask why we have this header at all. Don't isnan, * copysign, the built-in comparison operators, and the like solve these * problems? Unfortunately, they don't. We've found that various compilers * (MSVC, MSVC when compiling with PGO, and GCC on OS X, at least) miscompile * the standard methods in various situations, so we can't use them. Some of * these compilers even have problems compiling seemingly reasonable bitwise * algorithms! But with some care we've found algorithms that seem to not * trigger those compiler bugs. * * For the aforementioned reasons, be very wary of making changes to any of * these algorithms. If you must make changes, keep a careful eye out for * compiler bustage, particularly PGO-specific bustage. */ /* * These implementations all assume |double| is a 64-bit double format number * type, compatible with the IEEE-754 standard. C/C++ don't require this to be * the case. But we required this in implementations of these algorithms that * preceded this header, so we shouldn't break anything if we continue doing so. */ static_assert(sizeof(double) == sizeof(uint64_t), "double must be 64 bits"); const unsigned DoubleExponentBias = 1023; const unsigned DoubleExponentShift = 52; const uint64_t DoubleSignBit = 0x8000000000000000ULL; const uint64_t DoubleExponentBits = 0x7ff0000000000000ULL; const uint64_t DoubleSignificandBits = 0x000fffffffffffffULL; static_assert((DoubleSignBit & DoubleExponentBits) == 0, "sign bit doesn't overlap exponent bits"); static_assert((DoubleSignBit & DoubleSignificandBits) == 0, "sign bit doesn't overlap significand bits"); static_assert((DoubleExponentBits & DoubleSignificandBits) == 0, "exponent bits don't overlap significand bits"); static_assert((DoubleSignBit | DoubleExponentBits | DoubleSignificandBits) == ~uint64_t(0), "all bits accounted for"); /* * Ditto for |float| that must be a 32-bit double format number type, compatible * with the IEEE-754 standard. */ static_assert(sizeof(float) == sizeof(uint32_t), "float must be 32bits"); const unsigned FloatExponentBias = 127; const unsigned FloatExponentShift = 23; const uint32_t FloatSignBit = 0x80000000UL; const uint32_t FloatExponentBits = 0x7F800000UL; const uint32_t FloatSignificandBits = 0x007FFFFFUL; static_assert((FloatSignBit & FloatExponentBits) == 0, "sign bit doesn't overlap exponent bits"); static_assert((FloatSignBit & FloatSignificandBits) == 0, "sign bit doesn't overlap significand bits"); static_assert((FloatExponentBits & FloatSignificandBits) == 0, "exponent bits don't overlap significand bits"); static_assert((FloatSignBit | FloatExponentBits | FloatSignificandBits) == ~uint32_t(0), "all bits accounted for"); /** Determines whether a double is NaN. */ static MOZ_ALWAYS_INLINE bool IsNaN(double d) { /* * A double is NaN if all exponent bits are 1 and the significand contains at * least one non-zero bit. */ uint64_t bits = BitwiseCast(d); return (bits & DoubleExponentBits) == DoubleExponentBits && (bits & DoubleSignificandBits) != 0; } /** Determines whether a double is +Infinity or -Infinity. */ static MOZ_ALWAYS_INLINE bool IsInfinite(double d) { /* Infinities have all exponent bits set to 1 and an all-0 significand. */ uint64_t bits = BitwiseCast(d); return (bits & ~DoubleSignBit) == DoubleExponentBits; } /** Determines whether a double is not NaN or infinite. */ static MOZ_ALWAYS_INLINE bool IsFinite(double d) { /* * NaN and Infinities are the only non-finite doubles, and both have all * exponent bits set to 1. */ uint64_t bits = BitwiseCast(d); return (bits & DoubleExponentBits) != DoubleExponentBits; } /** * Determines whether a double is negative. It is an error to call this method * on a double which is NaN. */ static MOZ_ALWAYS_INLINE bool IsNegative(double d) { MOZ_ASSERT(!IsNaN(d), "NaN does not have a sign"); /* The sign bit is set if the double is negative. */ uint64_t bits = BitwiseCast(d); return (bits & DoubleSignBit) != 0; } /** Determines whether a double represents -0. */ static MOZ_ALWAYS_INLINE bool IsNegativeZero(double d) { /* Only the sign bit is set if the double is -0. */ uint64_t bits = BitwiseCast(d); return bits == DoubleSignBit; } /** * Returns the exponent portion of the double. * * Zero is not special-cased, so ExponentComponent(0.0) is * -int_fast16_t(DoubleExponentBias). */ static MOZ_ALWAYS_INLINE int_fast16_t ExponentComponent(double d) { /* * The exponent component of a double is an unsigned number, biased from its * actual value. Subtract the bias to retrieve the actual exponent. */ uint64_t bits = BitwiseCast(d); return int_fast16_t((bits & DoubleExponentBits) >> DoubleExponentShift) - int_fast16_t(DoubleExponentBias); } /** Returns +Infinity. */ static MOZ_ALWAYS_INLINE double PositiveInfinity() { /* * Positive infinity has all exponent bits set, sign bit set to 0, and no * significand. */ return BitwiseCast(DoubleExponentBits); } /** Returns -Infinity. */ static MOZ_ALWAYS_INLINE double NegativeInfinity() { /* * Negative infinity has all exponent bits set, sign bit set to 1, and no * significand. */ return BitwiseCast(DoubleSignBit | DoubleExponentBits); } /** Constructs a NaN value with the specified sign bit and significand bits. */ static MOZ_ALWAYS_INLINE double SpecificNaN(int signbit, uint64_t significand) { MOZ_ASSERT(signbit == 0 || signbit == 1); MOZ_ASSERT((significand & ~DoubleSignificandBits) == 0); MOZ_ASSERT(significand & DoubleSignificandBits); double d = BitwiseCast((signbit ? DoubleSignBit : 0) | DoubleExponentBits | significand); MOZ_ASSERT(IsNaN(d)); return d; } /** Computes the smallest non-zero positive double value. */ static MOZ_ALWAYS_INLINE double MinDoubleValue() { return BitwiseCast(uint64_t(1)); } /** * If d is equal to some int32_t value, set *i to that value and return true; * otherwise return false. * * Note that negative zero is "equal" to zero here. To test whether a value can * be losslessly converted to int32_t and back, use DoubleIsInt32 instead. */ static MOZ_ALWAYS_INLINE bool DoubleEqualsInt32(double d, int32_t* i) { /* * XXX Casting a double that doesn't truncate to int32_t, to int32_t, induces * undefined behavior. We should definitely fix this (bug 744965), but as * apparently it "works" in practice, it's not a pressing concern now. */ return d == (*i = int32_t(d)); } /** * If d can be converted to int32_t and back to an identical double value, * set *i to that value and return true; otherwise return false. * * The difference between this and DoubleEqualsInt32 is that this method returns * false for negative zero. */ static MOZ_ALWAYS_INLINE bool DoubleIsInt32(double d, int32_t* i) { return !IsNegativeZero(d) && DoubleEqualsInt32(d, i); } /** * Computes a NaN value. Do not use this method if you depend upon a particular * NaN value being returned. */ static MOZ_ALWAYS_INLINE double UnspecifiedNaN() { /* * If we can use any quiet NaN, we might as well use the all-ones NaN, * since it's cheap to materialize on common platforms (such as x64, where * this value can be represented in a 32-bit signed immediate field, allowing * it to be stored to memory in a single instruction). */ return SpecificNaN(1, 0xfffffffffffffULL); } /** * Compare two doubles for equality, *without* equating -0 to +0, and equating * any NaN value to any other NaN value. (The normal equality operators equate * -0 with +0, and they equate NaN to no other value.) */ static inline bool DoublesAreIdentical(double d1, double d2) { if (IsNaN(d1)) return IsNaN(d2); return BitwiseCast(d1) == BitwiseCast(d2); } /** Determines whether a float is NaN. */ static MOZ_ALWAYS_INLINE bool IsFloatNaN(float f) { /* * A float is NaN if all exponent bits are 1 and the significand contains at * least one non-zero bit. */ uint32_t bits = BitwiseCast(f); return (bits & FloatExponentBits) == FloatExponentBits && (bits & FloatSignificandBits) != 0; } /** Constructs a NaN value with the specified sign bit and significand bits. */ static MOZ_ALWAYS_INLINE float SpecificFloatNaN(int signbit, uint32_t significand) { MOZ_ASSERT(signbit == 0 || signbit == 1); MOZ_ASSERT((significand & ~FloatSignificandBits) == 0); MOZ_ASSERT(significand & FloatSignificandBits); float f = BitwiseCast((signbit ? FloatSignBit : 0) | FloatExponentBits | significand); MOZ_ASSERT(IsFloatNaN(f)); return f; } namespace detail { template struct FuzzyEqualsEpsilon; template<> struct FuzzyEqualsEpsilon { // A number near 1e-5 that is exactly representable in // floating point static const float value() { return 1.0f / (1 << 17); } }; template<> struct FuzzyEqualsEpsilon { // A number near 1e-12 that is exactly representable in // a double static const double value() { return 1.0 / (1LL << 40); } }; } // namespace detail /** * Compare two floating point values for equality, modulo rounding error. That * is, the two values are considered equal if they are both not NaN and if they * are less than or equal to epsilon apart. The default value of epsilon is near * 1e-5. * * For most scenarios you will want to use FuzzyEqualsMultiplicative instead, * as it is more reasonable over the entire range of floating point numbers. * This additive version should only be used if you know the range of the numbers * you are dealing with is bounded and stays around the same order of magnitude. */ template static MOZ_ALWAYS_INLINE bool FuzzyEqualsAdditive(T val1, T val2, T epsilon = detail::FuzzyEqualsEpsilon::value()) { static_assert(IsFloatingPoint::value, "floating point type required"); return Abs(val1 - val2) <= epsilon; } /** * Compare two floating point values for equality, allowing for rounding error * relative to the magnitude of the values. That is, the two values are * considered equal if they are both not NaN and they are less than or equal to * some epsilon apart, where the epsilon is scaled by the smaller of the two * argument values. * * In most cases you will want to use this rather than FuzzyEqualsAdditive, as * this function effectively masks out differences in the bottom few bits of * the floating point numbers being compared, regardless of what order of magnitude * those numbers are at. */ template static MOZ_ALWAYS_INLINE bool FuzzyEqualsMultiplicative(T val1, T val2, T epsilon = detail::FuzzyEqualsEpsilon::value()) { static_assert(IsFloatingPoint::value, "floating point type required"); // can't use std::min because of bug 965340 T smaller = Abs(val1) < Abs(val2) ? Abs(val1) : Abs(val2); return Abs(val1 - val2) <= epsilon * smaller; } /** * Returns true if the given value can be losslessly represented as an IEEE-754 * single format number, false otherwise. All NaN values are considered * representable (notwithstanding that the exact bit pattern of a double format * NaN value can't be exactly represented in single format). * * This function isn't inlined to avoid buggy optimizations by MSVC. */ MOZ_WARN_UNUSED_RESULT extern MFBT_API bool IsFloat32Representable(double x); } /* namespace mozilla */ #endif /* mozilla_FloatingPoint_h */