/
githubmirror
/
node
Обзор
Документация
Войти
/
githubmirror
/
node
Код
Запросы
0
Пакеты
0
Релизы
0
Аналитика
Безопасность
main
deps/v8/src/base/numerics/wrapping_math.h
45 строк
2 KB
Michaël Zasso
deps: update V8 to 13.7.152.9
18 май 2025, 10:42
18 май 2025, 10:42
fff0d15
Код
Авторство
О чём код?
// Copyright 2023 The Chromium Authors // Use of this source code is governed by a BSD-style license that can be // found in the LICENSE file. // Slightly adapted for inclusion in V8. // Copyright 2025 the V8 project authors. All rights reserved. #ifndef V8_BASE_NUMERICS_WRAPPING_MATH_H_ #define V8_BASE_NUMERICS_WRAPPING_MATH_H_ #include <type_traits> namespace v8::base { // Returns `a + b` with overflow defined to wrap around, i.e. modulo 2^N where N // is the bit width of `T`. template <typename T> inline constexpr T WrappingAdd(T a, T b) { static_assert(std::is_integral_v<T>); // Unsigned arithmetic wraps, so convert to the corresponding unsigned type. // Note that, if `T` is smaller than `int`, e.g. `int16_t`, the values are // promoted to `int`, which brings us back to undefined overflow. This is fine // here because the sum of any two `int16_t`s fits in `int`, but `WrappingMul` // will need a more complex implementation. using Unsigned = std::make_unsigned_t<T>; return static_cast<T>(static_cast<Unsigned>(a) + static_cast<Unsigned>(b)); } // Returns `a - b` with overflow defined to wrap around, i.e. modulo 2^N where N // is the bit width of `T`. template <typename T> inline constexpr T WrappingSub(T a, T b) { static_assert(std::is_integral_v<T>); // Unsigned arithmetic wraps, so convert to the corresponding unsigned type. // Note that, if `T` is smaller than `int`, e.g. `int16_t`, the values are // promoted to `int`, which brings us back to undefined overflow. This is fine // here because the difference of any two `int16_t`s fits in `int`, but // `WrappingMul` will need a more complex implementation. using Unsigned = std::make_unsigned_t<T>; return static_cast<T>(static_cast<Unsigned>(a) - static_cast<Unsigned>(b)); } } // namespace v8::base #endif // V8_BASE_NUMERICS_WRAPPING_MATH_H_