atomic_wait.h

Go to the documentation of this file.

// -*- C++ -*- header.
 
// Copyright (C) 2020-2026 Free Software Foundation, Inc.
//
// This file is part of the GNU ISO C++ Library.  This library is free
// software; you can redistribute it and/or modify it under the
// terms of the GNU General Public License as published by the
// Free Software Foundation; either version 3, or (at your option)
// any later version.
 
// This library is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
// GNU General Public License for more details.
 
// Under Section 7 of GPL version 3, you are granted additional
// permissions described in the GCC Runtime Library Exception, version
// 3.1, as published by the Free Software Foundation.
 
// You should have received a copy of the GNU General Public License and
// a copy of the GCC Runtime Library Exception along with this program;
// see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
// <http://www.gnu.org/licenses/>.
 
/** @file bits/atomic_wait.h
 *  This is an internal header file, included by other library headers.
 *  Do not attempt to use it directly. @headername{atomic}
 */
 
#ifndef _GLIBCXX_ATOMIC_WAIT_H
#define _GLIBCXX_ATOMIC_WAIT_H 1
 
#ifdef _GLIBCXX_SYSHDR
#pragma GCC system_header
#endif
 
#include <bits/version.h>
 
#if __glibcxx_atomic_wait
#include <bits/gthr.h>
#include <ext/numeric_traits.h>
 
#include <bits/stl_pair.h>
 
namespace std _GLIBCXX_VISIBILITY(default)
{
_GLIBCXX_BEGIN_NAMESPACE_VERSION
 
  namespace __detail
  {
    // TODO: this needs to be false for types with padding, e.g. __int20.
    // TODO: should this be true only for integral, enum, and pointer types?
    template<typename _Tp>
      concept __waitable
        = is_scalar_v<_Tp> && __builtin_popcountg(sizeof(_Tp)) == 1
            && (sizeof(_Tp) <= sizeof(__UINT64_TYPE__));
  }
 
#if defined _GLIBCXX_HAVE_LINUX_FUTEX
  namespace __detail
  {
    // Use futex syscall on int objects.
    using __platform_wait_t = int;
    inline constexpr size_t __platform_wait_alignment = 4;
  }
  // Defined to true for a subset of __waitable types which are statically
  // known to definitely be able to use futex, not a proxy wait.
  template<typename _Tp>
    inline constexpr bool __platform_wait_uses_type
      = __detail::__waitable<_Tp>
          && sizeof(_Tp) == sizeof(int) && alignof(_Tp) >= 4;
#elif defined __APPLE__ \
      && __ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ >= 101200
  namespace __detail
  {
    using __platform_wait_t = __INT32_TYPE__;
    inline constexpr size_t __platform_wait_alignment = 4;
  }
  // Defined to true for a subset of __waitable types which are statically
  // known to definitely be able to use __ulock_wait, not a proxy wait.
  // We know that OS Versions later than 10.15 support 64b wait types even
  // though we must make the __platform_wait_t 32b for compatibility with
  // earlier versions of __ulock_xxxx.
  template<typename _Tp>
    inline constexpr bool __platform_wait_uses_type
      = __detail::__waitable<_Tp>
#  if __ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ < 101500
          && sizeof(_Tp) == 4 && alignof(_Tp) >= 4;
#  else
          && ((sizeof(_Tp) == 4 && alignof(_Tp) >= 4)
                || (sizeof(_Tp) == 8 && alignof(_Tp) >= 8));
#  endif
#elif defined __FreeBSD__ && __SIZEOF_LONG__ == 8
  namespace __detail
  {
    using __platform_wait_t = __UINT64_TYPE__;
    inline constexpr size_t __platform_wait_alignment = 8;
  }
  template<typename _Tp>
    inline constexpr bool __platform_wait_uses_type
      = __detail::__waitable<_Tp>
          && ((sizeof(_Tp) == 4 && alignof(_Tp) >= 4)
                || (sizeof(_Tp) == 8 && alignof(_Tp) >= 8));
#else
// define _GLIBCX_HAVE_PLATFORM_WAIT and implement __platform_wait()
// and __platform_notify() if there is a more efficient primitive supported
// by the platform (e.g. __ulock_wait()/__ulock_wake()) which is better than
// a mutex/condvar based wait.
  namespace __detail
  {
# if defined __OpenBSD__ || defined __DragonFly__
    // These targets provide 32-bit futex-like syscalls.
    // We don't currently make use of them, but we want to in future.
    using __platform_wait_t = unsigned int;
# elif ATOMIC_LONG_LOCK_FREE == 2
    using __platform_wait_t = unsigned long;
# else
    using __platform_wait_t = unsigned int;
# endif
    inline constexpr size_t __platform_wait_alignment
      = sizeof(__platform_wait_t) < __alignof__(__platform_wait_t)
          ? __alignof__(__platform_wait_t) : sizeof(__platform_wait_t);
  } // namespace __detail
 
  // This must be false for the general case where we don't know of any
  // futex-like syscall.
  template<typename>
    inline constexpr bool __platform_wait_uses_type = false;
#endif
 
  namespace __detail
  {
    inline void
    __thread_yield() noexcept
    {
#if defined _GLIBCXX_HAS_GTHREADS && defined _GLIBCXX_USE_SCHED_YIELD
      __gthread_yield();
#endif
    }
 
    inline void
    __thread_relax() noexcept
    {
#if defined __i386__ || defined __x86_64__
      __builtin_ia32_pause();
#else
      __thread_yield();
#endif
    }
 
    // return true if equal
    template<typename _Tp>
      inline bool
      __atomic_eq(const _Tp& __a, const _Tp& __b)
      {
        // TODO make this do the correct padding bit ignoring comparison
        return __builtin_memcmp(std::addressof(__a), std::addressof(__b),
                                sizeof(_Tp)) == 0;
      }
 
    // Storage for up to 64 bits of value, should be considered opaque bits.
    using __wait_value_type = __UINT64_TYPE__;
 
    // lightweight std::optional<__wait_value_type>
    struct __wait_result_type
    {
      __wait_value_type _M_val;
      unsigned char _M_has_val : 1; // _M_val value was loaded before return.
      unsigned char _M_timeout : 1; // Waiting function ended with timeout.
      unsigned char _M_unused : 6;  // padding
    };
 
    enum class __wait_flags : __UINT_LEAST32_TYPE__
    {
       __abi_version = 0x00000000,
       // currently unused = 1,
       __track_contention = 2,
       __do_spin = 4,
       __spin_only = 8, // Ignored unless __do_spin is also set.
       // __abi_version_mask = 0xff000000,
    };
 
    [[__gnu__::__always_inline__]]
    constexpr __wait_flags
    operator|(__wait_flags __l, __wait_flags __r) noexcept
    {
      using _Ut = underlying_type_t<__wait_flags>;
      return static_cast<__wait_flags>(static_cast<_Ut>(__l)
                                         | static_cast<_Ut>(__r));
    }
 
    [[__gnu__::__always_inline__]]
    constexpr __wait_flags&
    operator|=(__wait_flags& __l, __wait_flags __r) noexcept
    { return __l = __l | __r; }
 
    // Simple aggregate containing arguments used by implementation details.
    struct __wait_args_base
    {
      __wait_flags _M_flags;
      int _M_order = __ATOMIC_ACQUIRE; // Memory order for loads from _M_obj.
      __wait_value_type _M_old = 0;  // Previous value of *_M_obj.
      void* _M_wait_state = nullptr; // For proxy wait and tracking contention.
      const void* _M_obj = nullptr;  // The address of the object to wait on.
      unsigned char _M_obj_size = 0; // The size of that object.
 
      // Test whether _M_flags & __flags is non-zero.
      bool
      operator&(__wait_flags __flags) const noexcept
      {
         using _Ut = underlying_type_t<__wait_flags>;
         return static_cast<_Ut>(_M_flags) & static_cast<_Ut>(__flags);
      }
    };
 
    // Utility for populating a __wait_args_base structure.
    struct __wait_args : __wait_args_base
    {
      template<typename _Tp> requires (!is_same_v<_Tp, __wait_args>)
        explicit
        __wait_args(const _Tp* __addr, bool __bare_wait = false) noexcept
        : __wait_args_base{ _S_flags_for(__addr, __bare_wait) }
        {
          _M_obj = __addr; // Might be replaced by _M_setup_wait
          if constexpr (__waitable<_Tp>)
            // __wait_impl might be able to wait directly on __addr
            // instead of using a proxy, depending on its size.
            _M_obj_size = sizeof(_Tp);
        }
 
      __wait_args(const __platform_wait_t* __addr, __platform_wait_t __old,
                  int __order, bool __bare_wait = false) noexcept
      : __wait_args(__addr, __bare_wait)
      {
        _M_order = __order;
        _M_old = __old;
      }
 
      __wait_args(const __wait_args&) noexcept = default;
      __wait_args& operator=(const __wait_args&) noexcept = default;
 
      template<typename _Tp, typename _ValFn>
        _Tp
        _M_setup_wait(const _Tp* __addr, _ValFn __vfn)
        {
          static_assert(is_same_v<_Tp, decay_t<decltype(__vfn())>>);
 
          if constexpr (!__platform_wait_uses_type<_Tp>)
            if (_M_setup_proxy_wait(__addr))
              {
                // We will use a proxy wait for this object.
                // The library has set _M_wait_state, _M_obj, _M_obj_size,
                // and _M_old.
                // Call __vfn to load the current value from *__addr
                // (which must happen after the call to _M_setup_proxy_wait).
                return __vfn();
              }
 
          // We will use a futex-like operation to wait on this object,
          // so just load the value, store it into _M_old, and return it.
          return _M_store(__vfn());
        }
 
      // Called after a wait returns, to prepare to wait again.
      template<typename _Tp, typename _ValFn>
        _Tp
        _M_on_wake(const _Tp* __addr, _ValFn __vfn, __wait_result_type __res)
        {
          if constexpr (!__platform_wait_uses_type<_Tp>) // maybe a proxy wait
            if (_M_obj != __addr) // definitely a proxy wait
              {
                if (__res._M_has_val)
                  // Previous wait loaded a recent value from the proxy.
                  _M_old = __res._M_val;
                else // Load a new value from the proxy and store in _M_old.
                  (void) _M_setup_proxy_wait(nullptr);
                // Read the current value of *__addr
                return __vfn();
              }
 
          if (__res._M_has_val) // The previous wait loaded a recent value.
            {
              _M_old = __res._M_val;
 
              // Not a proxy wait, so the value in __res._M_val was loaded
              // from *__addr and we don't need to call __vfn().
              if constexpr (sizeof(_Tp) == sizeof(__UINT64_TYPE__))
                return __builtin_bit_cast(_Tp, (__UINT64_TYPE__)_M_old);
              else if constexpr (sizeof(_Tp) == sizeof(__UINT32_TYPE__))
                return __builtin_bit_cast(_Tp, (__UINT32_TYPE__)_M_old);
              else if constexpr (sizeof(_Tp) == sizeof(__UINT16_TYPE__))
                return __builtin_bit_cast(_Tp, (__UINT16_TYPE__)_M_old);
              else if constexpr (sizeof(_Tp) == sizeof(__UINT8_TYPE__))
                return __builtin_bit_cast(_Tp, (__UINT8_TYPE__)_M_old);
              else // Should be a proxy wait for this size!
                __glibcxx_assert(false);
            }
          else
            return _M_store(__vfn());
        }
 
    private:
      // Store __val in _M_old.
      // pre: This must be a non-proxy wait.
      template<typename _Tp>
        [[__gnu__::__always_inline__]]
        _Tp
        _M_store(_Tp __val)
        {
          // We have to consider various sizes, because a future libstdc++.so
          // might enable non-proxy waits for additional sizes.
          if constexpr (sizeof(_Tp) == sizeof(__UINT64_TYPE__))
            _M_old = __builtin_bit_cast(__UINT64_TYPE__, __val);
          else if constexpr (sizeof(_Tp) == sizeof(__UINT32_TYPE__))
            _M_old = __builtin_bit_cast(__UINT32_TYPE__, __val);
          else if constexpr (sizeof(_Tp) == sizeof(__UINT16_TYPE__))
            _M_old = __builtin_bit_cast(__UINT16_TYPE__, __val);
          else if constexpr (sizeof(_Tp) == sizeof(__UINT8_TYPE__))
            _M_old = __builtin_bit_cast(__UINT8_TYPE__, __val);
          else // Should be a proxy wait for this size!
            __glibcxx_assert(false);
          return __val;
        }
 
      // Prepare `*this` for a call to `__wait_impl` or `__wait_until_impl`.
      // See comments in src/c++20/atomic.cc for more details.
      bool
      _M_setup_proxy_wait(const void* __addr);
 
      template<typename _Tp>
        static constexpr __wait_flags
        _S_flags_for(const _Tp*, bool __bare_wait) noexcept
        {
          using enum __wait_flags;
          __wait_flags __res = __abi_version | __do_spin;
          if (!__bare_wait)
            __res |= __track_contention;
          return __res;
        }
    };
 
    __wait_result_type
    __wait_impl(const void* __addr, __wait_args_base&);
 
    void
    __notify_impl(const void* __addr, bool __all, const __wait_args_base&);
  } // namespace __detail
 
  // Wait on __addr while __pred(__vfn()) is false.
  // If __bare_wait is false, increment a counter while waiting.
  // For callers that keep their own count of waiters, use __bare_wait=true.
  // The effect of __vfn() must be an atomic load from __addr and nothing else.
  template<typename _Tp, typename _Pred, typename _ValFn>
    void
    __atomic_wait_address(const _Tp* __addr, _Pred&& __pred, _ValFn&& __vfn,
                          bool __bare_wait = false) noexcept
    {
      __detail::__wait_args __args{ __addr, __bare_wait };
      _Tp __val = __args._M_setup_wait(__addr, __vfn);
      while (!__pred(__val))
        {
          auto __res = __detail::__wait_impl(__addr, __args);
          __val = __args._M_on_wake(__addr, __vfn, __res);
        }
      // C++26 will return __val
    }
 
  // Wait on __addr while *__addr == __old is true.
  inline void
  __atomic_wait_address_v(const __detail::__platform_wait_t* __addr,
                          __detail::__platform_wait_t __old,
                          int __order, bool __bare_wait = false)
  {
    // This function must not be used if __wait_impl might use a proxy wait:
    __glibcxx_assert(__platform_wait_uses_type<__detail::__platform_wait_t>);
 
    __detail::__wait_args __args{ __addr, __old, __order, __bare_wait };
    // C++26 will not ignore the return value here
    __detail::__wait_impl(__addr, __args);
  }
 
  // Wait on __addr while __vfn() == __old is true.
  template<typename _Tp, typename _ValFn>
    void
    __atomic_wait_address_v(const _Tp* __addr, _Tp __old,
                            _ValFn __vfn) noexcept
    {
      auto __pfn = [&](const _Tp& __val)
          { return !__detail::__atomic_eq(__old, __val); };
      std::__atomic_wait_address(__addr, __pfn, forward<_ValFn>(__vfn));
    }
 
  template<typename _Tp>
    void
    __atomic_notify_address(const _Tp* __addr, bool __all,
                            bool __bare_wait = false) noexcept
    {
      __detail::__wait_args __args{ __addr, __bare_wait };
      __detail::__notify_impl(__addr, __all, __args);
    }
 
_GLIBCXX_END_NAMESPACE_VERSION
} // namespace std
#endif // __glibcxx_atomic_wait
#endif // _GLIBCXX_ATOMIC_WAIT_H