[libcxx-commits] [libcxx] [libcxx] p3008 atomic fp min/max (PR #186716)

[via libcxx-commits]

================
@@ -0,0 +1,65 @@
+//===----------------------------------------------------------------------===//
+//
+// 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, c++20, c++23
+// XFAIL: !has-64-bit-atomics
+
+// floating-point-type fetch_max(floating-point-type, memory_order = memory_order::seq_cst) const noexcept;
+
+#include <atomic>
+#include <cassert>
+#include <concepts>
+#include <type_traits>
+#include <utility>
+
+#include "test_macros.h"
+#include "atomic_helpers.h"
+#include "../atomics.types.operations/atomics.types.operations.req/atomic_fetch_fmax_helper.h"
+
+template <typename T>
+concept has_fetch_max = requires(T t) {
+  std::declval<T const>().fetch_max(std::declval<typename T::value_type>());
+  std::declval<T const>().fetch_max(std::declval<typename T::value_type>(), std::declval<std::memory_order>());
+};
+
+template <typename T>
+struct TestDoesNotHaveFetchMax {
+  void operator()() const { static_assert(!has_fetch_max<std::atomic_ref<T>>); }
+};
+
+template <typename T>
+struct TestFetchMax {
+  void operator()() const {
+    if constexpr (std::is_floating_point_v<T>) {
+      alignas(std::atomic_ref<T>::required_alignment) T x;
+      std::atomic_ref<T> const a(x);
+
+      auto load   = [&]() { return x; };
+      auto store  = [&](T val) { x = val; };
+      auto max_op = [&](T val, auto order) { return a.fetch_max(val, order); };
+
+      ASSERT_NOEXCEPT(a.fetch_max(T(0), std::memory_order_seq_cst));
+      test_fetch_fmax<T>(load, store, max_op);
+
+    } else {
+      static_assert(std::is_void_v<T>);
+    }
+  }
+};
+
+int main(int, char**) {
+  TestFetchMax<float>{}();
+  TestFetchMax<double>{}();
----------------
huixie90 wrote:

please test `long double`

https://github.com/llvm/llvm-project/pull/186716