| //===----------------------------------------------------------------------===// |
| // |
| // 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 |
| // |
| //===----------------------------------------------------------------------===// |
| // UNSUPPORTED: c++03, c++11, c++14, c++17 |
| // XFAIL: !has-64-bit-atomics |
| // UNSUPPORTED: LIBCXX-AIX-FIXME |
| // UNSUPPORTED: !non-lockfree-atomics |
| // Hangs with msan. |
| // UNSUPPORTED: msan |
| |
| // floating-point-type operator+=(floating-point-type) volatile noexcept; |
| // floating-point-type operator+=(floating-point-type) noexcept; |
| |
| #include <atomic> |
| #include <cassert> |
| #include <concepts> |
| #include <type_traits> |
| #include <vector> |
| |
| #include "test_helper.h" |
| #include "test_macros.h" |
| |
| #ifndef TEST_HAS_NO_THREADS |
| # include "make_test_thread.h" |
| # include <thread> |
| #endif |
| |
| template <class T> |
| concept HasVolatilePlusEquals = requires(volatile std::atomic<T>& a, T t) { a += t; }; |
| |
| template <class T, template <class> class MaybeVolatile = std::type_identity_t> |
| void test_impl() { |
| static_assert(HasVolatilePlusEquals<T> == std::atomic<T>::is_always_lock_free); |
| static_assert(noexcept(std::declval<MaybeVolatile<std::atomic<T>>&>() += T(0))); |
| |
| // += |
| { |
| MaybeVolatile<std::atomic<T>> a(T(3.1)); |
| std::same_as<T> decltype(auto) r = a += T(1.2); |
| assert(r == T(3.1) + T(1.2)); |
| assert(a.load() == T(3.1) + T(1.2)); |
| } |
| |
| #ifndef TEST_HAS_NO_THREADS |
| // += concurrent |
| { |
| constexpr auto number_of_threads = 4; |
| constexpr auto loop = 1000; |
| |
| MaybeVolatile<std::atomic<T>> at; |
| |
| std::vector<std::thread> threads; |
| threads.reserve(number_of_threads); |
| for (auto i = 0; i < number_of_threads; ++i) { |
| threads.push_back(support::make_test_thread([&at]() { |
| for (auto j = 0; j < loop; ++j) { |
| at += T(1.234); |
| } |
| })); |
| } |
| |
| for (auto& thread : threads) { |
| thread.join(); |
| } |
| |
| const auto times = [](T t, int n) { |
| T res(0); |
| for (auto i = 0; i < n; ++i) { |
| res += t; |
| } |
| return res; |
| }; |
| |
| assert(at.load() == times(T(1.234), number_of_threads * loop)); |
| } |
| |
| // memory_order::seq_cst |
| { |
| auto plus_equals = [](MaybeVolatile<std::atomic<T>>& x, T old_value, T new_val) { x += (new_val - old_value); }; |
| auto load = [](MaybeVolatile<std::atomic<T>>& x) { return x.load(); }; |
| test_seq_cst<T, MaybeVolatile>(plus_equals, load); |
| } |
| #endif |
| } |
| |
| template <class T> |
| void test() { |
| test_impl<T>(); |
| if constexpr (std::atomic<T>::is_always_lock_free) { |
| test_impl<T, std::add_volatile_t>(); |
| } |
| } |
| |
| int main(int, char**) { |
| test<float>(); |
| test<double>(); |
| test<long double>(); |
| |
| return 0; |
| } |