/* -*- 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/MemoryChecking.h" #include "mozilla/Types.h" #include #include #include #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. */ namespace detail { /* * These implementations assume float/double are 32/64-bit single/double * format number types compatible with the IEEE-754 standard. C++ doesn't * require this, but we required it in implementations of these algorithms that * preceded this header, so we shouldn't break anything to continue doing so. */ template struct FloatingPointTrait; template <> struct FloatingPointTrait { protected: using Bits = uint32_t; static constexpr unsigned kExponentWidth = 8; static constexpr unsigned kSignificandWidth = 23; }; template <> struct FloatingPointTrait { protected: using Bits = uint64_t; static constexpr unsigned kExponentWidth = 11; static constexpr unsigned kSignificandWidth = 52; }; } // namespace detail /* * This struct contains details regarding the encoding of floating-point * numbers that can be useful for direct bit manipulation. As of now, the * template parameter has to be float or double. * * The nested typedef |Bits| is the unsigned integral type with the same size * as T: uint32_t for float and uint64_t for double (static assertions * double-check these assumptions). * * kExponentBias is the offset that is subtracted from the exponent when * computing the value, i.e. one plus the opposite of the mininum possible * exponent. * kExponentShift is the shift that one needs to apply to retrieve the * exponent component of the value. * * kSignBit contains a bits mask. Bit-and-ing with this mask will result in * obtaining the sign bit. * kExponentBits contains the mask needed for obtaining the exponent bits and * kSignificandBits contains the mask needed for obtaining the significand * bits. * * Full details of how floating point number formats are encoded are beyond * the scope of this comment. For more information, see * http://en.wikipedia.org/wiki/IEEE_floating_point * http://en.wikipedia.org/wiki/Floating_point#IEEE_754:_floating_point_in_modern_computers */ template struct FloatingPoint final : private detail::FloatingPointTrait { private: using Base = detail::FloatingPointTrait; public: /** * An unsigned integral type suitable for accessing the bitwise representation * of T. */ using Bits = typename Base::Bits; static_assert(sizeof(T) == sizeof(Bits), "Bits must be same size as T"); /** The bit-width of the exponent component of T. */ using Base::kExponentWidth; /** The bit-width of the significand component of T. */ using Base::kSignificandWidth; static_assert(1 + kExponentWidth + kSignificandWidth == CHAR_BIT * sizeof(T), "sign bit plus bit widths should sum to overall bit width"); /** * The exponent field in an IEEE-754 floating point number consists of bits * encoding an unsigned number. The *actual* represented exponent (for all * values finite and not denormal) is that value, minus a bias |kExponentBias| * so that a useful range of numbers is represented. */ static constexpr unsigned kExponentBias = (1U << (kExponentWidth - 1)) - 1; /** * The amount by which the bits of the exponent-field in an IEEE-754 floating * point number are shifted from the LSB of the floating point type. */ static constexpr unsigned kExponentShift = kSignificandWidth; /** The sign bit in the floating point representation. */ static constexpr Bits kSignBit = static_cast(1) << (CHAR_BIT * sizeof(Bits) - 1); /** The exponent bits in the floating point representation. */ static constexpr Bits kExponentBits = ((static_cast(1) << kExponentWidth) - 1) << kSignificandWidth; /** The significand bits in the floating point representation. */ static constexpr Bits kSignificandBits = (static_cast(1) << kSignificandWidth) - 1; static_assert((kSignBit & kExponentBits) == 0, "sign bit shouldn't overlap exponent bits"); static_assert((kSignBit & kSignificandBits) == 0, "sign bit shouldn't overlap significand bits"); static_assert((kExponentBits & kSignificandBits) == 0, "exponent bits shouldn't overlap significand bits"); static_assert((kSignBit | kExponentBits | kSignificandBits) == ~Bits(0), "all bits accounted for"); }; /** Determines whether a float/double is NaN. */ template static MOZ_ALWAYS_INLINE bool IsNaN(T aValue) { /* * A float/double is NaN if all exponent bits are 1 and the significand * contains at least one non-zero bit. */ typedef FloatingPoint Traits; typedef typename Traits::Bits Bits; return (BitwiseCast(aValue) & Traits::kExponentBits) == Traits::kExponentBits && (BitwiseCast(aValue) & Traits::kSignificandBits) != 0; } /** Determines whether a float/double is +Infinity or -Infinity. */ template static MOZ_ALWAYS_INLINE bool IsInfinite(T aValue) { /* Infinities have all exponent bits set to 1 and an all-0 significand. */ typedef FloatingPoint Traits; typedef typename Traits::Bits Bits; Bits bits = BitwiseCast(aValue); return (bits & ~Traits::kSignBit) == Traits::kExponentBits; } /** Determines whether a float/double is not NaN or infinite. */ template static MOZ_ALWAYS_INLINE bool IsFinite(T aValue) { /* * NaN and Infinities are the only non-finite floats/doubles, and both have * all exponent bits set to 1. */ typedef FloatingPoint Traits; typedef typename Traits::Bits Bits; Bits bits = BitwiseCast(aValue); return (bits & Traits::kExponentBits) != Traits::kExponentBits; } /** * Determines whether a float/double is negative or -0. It is an error * to call this method on a float/double which is NaN. */ template static MOZ_ALWAYS_INLINE bool IsNegative(T aValue) { MOZ_ASSERT(!IsNaN(aValue), "NaN does not have a sign"); /* The sign bit is set if the double is negative. */ typedef FloatingPoint Traits; typedef typename Traits::Bits Bits; Bits bits = BitwiseCast(aValue); return (bits & Traits::kSignBit) != 0; } /** Determines whether a float/double represents -0. */ template static MOZ_ALWAYS_INLINE bool IsNegativeZero(T aValue) { /* Only the sign bit is set if the value is -0. */ typedef FloatingPoint Traits; typedef typename Traits::Bits Bits; Bits bits = BitwiseCast(aValue); return bits == Traits::kSignBit; } /** Determines wether a float/double represents +0. */ template static MOZ_ALWAYS_INLINE bool IsPositiveZero(T aValue) { /* All bits are zero if the value is +0. */ typedef FloatingPoint Traits; typedef typename Traits::Bits Bits; Bits bits = BitwiseCast(aValue); return bits == 0; } /** * Returns 0 if a float/double is NaN or infinite; * otherwise, the float/double is returned. */ template static MOZ_ALWAYS_INLINE T ToZeroIfNonfinite(T aValue) { return IsFinite(aValue) ? aValue : 0; } /** * Returns the exponent portion of the float/double. * * Zero is not special-cased, so ExponentComponent(0.0) is * -int_fast16_t(Traits::kExponentBias). */ template static MOZ_ALWAYS_INLINE int_fast16_t ExponentComponent(T aValue) { /* * The exponent component of a float/double is an unsigned number, biased * from its actual value. Subtract the bias to retrieve the actual exponent. */ typedef FloatingPoint Traits; typedef typename Traits::Bits Bits; Bits bits = BitwiseCast(aValue); return int_fast16_t((bits & Traits::kExponentBits) >> Traits::kExponentShift) - int_fast16_t(Traits::kExponentBias); } /** Returns +Infinity. */ template static MOZ_ALWAYS_INLINE T PositiveInfinity() { /* * Positive infinity has all exponent bits set, sign bit set to 0, and no * significand. */ typedef FloatingPoint Traits; return BitwiseCast(Traits::kExponentBits); } /** Returns -Infinity. */ template static MOZ_ALWAYS_INLINE T NegativeInfinity() { /* * Negative infinity has all exponent bits set, sign bit set to 1, and no * significand. */ typedef FloatingPoint Traits; return BitwiseCast(Traits::kSignBit | Traits::kExponentBits); } /** * Computes the bit pattern for an infinity with the specified sign bit. */ template struct InfinityBits { using Traits = FloatingPoint; static_assert(SignBit == 0 || SignBit == 1, "bad sign bit"); static constexpr typename Traits::Bits value = (SignBit * Traits::kSignBit) | Traits::kExponentBits; }; /** * Computes the bit pattern for a NaN with the specified sign bit and * significand bits. */ template ::Bits Significand> struct SpecificNaNBits { using Traits = FloatingPoint; static_assert(SignBit == 0 || SignBit == 1, "bad sign bit"); static_assert((Significand & ~Traits::kSignificandBits) == 0, "significand must only have significand bits set"); static_assert(Significand & Traits::kSignificandBits, "significand must be nonzero"); static constexpr typename Traits::Bits value = (SignBit * Traits::kSignBit) | Traits::kExponentBits | Significand; }; /** * Constructs a NaN value with the specified sign bit and significand bits. * * There is also a variant that returns the value directly. In most cases, the * two variants should be identical. However, in the specific case of x86 * chips, the behavior differs: returning floating-point values directly is done * through the x87 stack, and x87 loads and stores turn signaling NaNs into * quiet NaNs... silently. Returning floating-point values via outparam, * however, is done entirely within the SSE registers when SSE2 floating-point * is enabled in the compiler, which has semantics-preserving behavior you would * expect. * * If preserving the distinction between signaling NaNs and quiet NaNs is * important to you, you should use the outparam version. In all other cases, * you should use the direct return version. */ template static MOZ_ALWAYS_INLINE void SpecificNaN( int signbit, typename FloatingPoint::Bits significand, T* result) { typedef FloatingPoint Traits; MOZ_ASSERT(signbit == 0 || signbit == 1); MOZ_ASSERT((significand & ~Traits::kSignificandBits) == 0); MOZ_ASSERT(significand & Traits::kSignificandBits); BitwiseCast( (signbit ? Traits::kSignBit : 0) | Traits::kExponentBits | significand, result); MOZ_ASSERT(IsNaN(*result)); } template static MOZ_ALWAYS_INLINE T SpecificNaN(int signbit, typename FloatingPoint::Bits significand) { T t; SpecificNaN(signbit, significand, &t); return t; } /** Computes the smallest non-zero positive float/double value. */ template static MOZ_ALWAYS_INLINE T MinNumberValue() { typedef FloatingPoint Traits; typedef typename Traits::Bits Bits; return BitwiseCast(Bits(1)); } namespace detail { template inline bool NumberEqualsSignedInteger(Float aValue, SignedInteger* aInteger) { static_assert(std::is_same_v || std::is_same_v, "Float must be an IEEE-754 floating point type"); static_assert(std::is_signed_v, "this algorithm only works for signed types: a different one " "will be required for unsigned types"); static_assert(sizeof(SignedInteger) >= sizeof(int), "this function *might* require some finessing for signed types " "subject to integral promotion before it can be used on them"); MOZ_MAKE_MEM_UNDEFINED(aInteger, sizeof(*aInteger)); // NaNs and infinities are not integers. if (!IsFinite(aValue)) { return false; } // Otherwise do direct comparisons against the minimum/maximum |SignedInteger| // values that can be encoded in |Float|. constexpr SignedInteger MaxIntValue = std::numeric_limits::max(); // e.g. INT32_MAX constexpr SignedInteger MinValue = std::numeric_limits::min(); // e.g. INT32_MIN static_assert(IsPowerOfTwo(Abs(MinValue)), "MinValue should be is a small power of two, thus exactly " "representable in float/double both"); constexpr unsigned SignedIntegerWidth = CHAR_BIT * sizeof(SignedInteger); constexpr unsigned ExponentShift = FloatingPoint::kExponentShift; // Careful! |MaxIntValue| may not be the maximum |SignedInteger| value that // can be encoded in |Float|. Its |SignedIntegerWidth - 1| bits of precision // may exceed |Float|'s |ExponentShift + 1| bits of precision. If necessary, // compute the maximum |SignedInteger| that fits in |Float| from IEEE-754 // first principles. (|MinValue| doesn't have this problem because as a // [relatively] small power of two it's always representable in |Float|.) // Per C++11 [expr.const]p2, unevaluated subexpressions of logical AND/OR and // conditional expressions *may* contain non-constant expressions, without // making the enclosing expression not constexpr. MSVC implements this -- but // it sometimes warns about undefined behavior in unevaluated subexpressions. // This bites us if we initialize |MaxValue| the obvious way including an // |uint64_t(1) << (SignedIntegerWidth - 2 - ExponentShift)| subexpression. // Pull that shift-amount out and give it a not-too-huge value when it's in an // unevaluated subexpression. 🙄 constexpr unsigned PrecisionExceededShiftAmount = ExponentShift > SignedIntegerWidth - 1 ? 0 : SignedIntegerWidth - 2 - ExponentShift; constexpr SignedInteger MaxValue = ExponentShift > SignedIntegerWidth - 1 ? MaxIntValue : SignedInteger((uint64_t(1) << (SignedIntegerWidth - 1)) - (uint64_t(1) << PrecisionExceededShiftAmount)); if (static_cast(MinValue) <= aValue && aValue <= static_cast(MaxValue)) { auto possible = static_cast(aValue); if (static_cast(possible) == aValue) { *aInteger = possible; return true; } } return false; } template inline bool NumberIsSignedInteger(Float aValue, SignedInteger* aInteger) { static_assert(std::is_same_v || std::is_same_v, "Float must be an IEEE-754 floating point type"); static_assert(std::is_signed_v, "this algorithm only works for signed types: a different one " "will be required for unsigned types"); static_assert(sizeof(SignedInteger) >= sizeof(int), "this function *might* require some finessing for signed types " "subject to integral promotion before it can be used on them"); MOZ_MAKE_MEM_UNDEFINED(aInteger, sizeof(*aInteger)); if (IsNegativeZero(aValue)) { return false; } return NumberEqualsSignedInteger(aValue, aInteger); } } // namespace detail /** * If |aValue| is identical to some |int32_t| value, set |*aInt32| to that value * and return true. Otherwise return false, leaving |*aInt32| in an * indeterminate state. * * This method returns false for negative zero. If you want to consider -0 to * be 0, use NumberEqualsInt32 below. */ template static MOZ_ALWAYS_INLINE bool NumberIsInt32(T aValue, int32_t* aInt32) { return detail::NumberIsSignedInteger(aValue, aInt32); } /** * If |aValue| is equal to some int32_t value (where -0 and +0 are considered * equal), set |*aInt32| to that value and return true. Otherwise return false, * leaving |*aInt32| in an indeterminate state. * * |NumberEqualsInt32(-0.0, ...)| will return true. To test whether a value can * be losslessly converted to |int32_t| and back, use NumberIsInt32 above. */ template static MOZ_ALWAYS_INLINE bool NumberEqualsInt32(T aValue, int32_t* aInt32) { return detail::NumberEqualsSignedInteger(aValue, aInt32); } /** * Computes a NaN value. Do not use this method if you depend upon a particular * NaN value being returned. */ template static MOZ_ALWAYS_INLINE T 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). */ typedef FloatingPoint Traits; return SpecificNaN(1, Traits::kSignificandBits); } /** * 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.) */ template static inline bool NumbersAreIdentical(T aValue1, T aValue2) { typedef FloatingPoint Traits; typedef typename Traits::Bits Bits; if (IsNaN(aValue1)) { return IsNaN(aValue2); } return BitwiseCast(aValue1) == BitwiseCast(aValue2); } /** * Return true iff |aValue| and |aValue2| are equal (ignoring sign if both are * zero) or both NaN. */ template static inline bool EqualOrBothNaN(T aValue1, T aValue2) { if (IsNaN(aValue1)) { return IsNaN(aValue2); } return aValue1 == aValue2; } /** * Return NaN if either |aValue1| or |aValue2| is NaN, or the minimum of * |aValue1| and |aValue2| otherwise. */ template static inline T NaNSafeMin(T aValue1, T aValue2) { if (IsNaN(aValue1) || IsNaN(aValue2)) { return UnspecifiedNaN(); } return std::min(aValue1, aValue2); } /** * Return NaN if either |aValue1| or |aValue2| is NaN, or the maximum of * |aValue1| and |aValue2| otherwise. */ template static inline T NaNSafeMax(T aValue1, T aValue2) { if (IsNaN(aValue1) || IsNaN(aValue2)) { return UnspecifiedNaN(); } return std::max(aValue1, aValue2); } namespace detail { template struct FuzzyEqualsEpsilon; template <> struct FuzzyEqualsEpsilon { // A number near 1e-5 that is exactly representable in a float. static float value() { return 1.0f / (1 << 17); } }; template <> struct FuzzyEqualsEpsilon { // A number near 1e-12 that is exactly representable in a double. static 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 aEpsilon apart. The default value of aEpsilon 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 aValue1, T aValue2, T aEpsilon = detail::FuzzyEqualsEpsilon::value()) { static_assert(std::is_floating_point_v, "floating point type required"); return Abs(aValue1 - aValue2) <= aEpsilon; } /** * 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 aEpsilon apart, where the aEpsilon 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 aValue1, T aValue2, T aEpsilon = detail::FuzzyEqualsEpsilon::value()) { static_assert(std::is_floating_point_v, "floating point type required"); // can't use std::min because of bug 965340 T smaller = Abs(aValue1) < Abs(aValue2) ? Abs(aValue1) : Abs(aValue2); return Abs(aValue1 - aValue2) <= aEpsilon * smaller; } /** * Returns true if |aValue| can be losslessly represented as an IEEE-754 single * precision number, false otherwise. All NaN values are considered * representable (even though the bit patterns of double precision NaNs can't * all be exactly represented in single precision). */ MOZ_MUST_USE extern MFBT_API bool IsFloat32Representable(double aValue); } /* namespace mozilla */ #endif /* mozilla_FloatingPoint_h */