Gecko [ v3.cesa.co.za ]

Name	Size	Permission	Date
backward	[ DIR ]	drwxr-xr-x	2025-10-28 07:06
bits	[ DIR ]	drwxr-xr-x	2025-10-28 07:06
debug	[ DIR ]	drwxr-xr-x	2025-10-28 07:06
decimal	[ DIR ]	drwxr-xr-x	2025-10-28 07:06
experimental	[ DIR ]	drwxr-xr-x	2025-10-28 07:06
ext	[ DIR ]	drwxr-xr-x	2025-10-28 07:06
parallel	[ DIR ]	drwxr-xr-x	2025-10-28 07:06
pstl	[ DIR ]	drwxr-xr-x	2025-10-28 07:06
tr1	[ DIR ]	drwxr-xr-x	2025-10-28 07:06
tr2	[ DIR ]	drwxr-xr-x	2025-10-28 07:06
algorithm	2.94 KB	-rw-r--r--	2025-04-07 11:26
any	18.38 KB	-rw-r--r--	2025-04-07 11:26
array	14.61 KB	-rw-r--r--	2025-04-07 11:26
atomic	46.67 KB	-rw-r--r--	2025-04-07 11:26
barrier	7.57 KB	-rw-r--r--	2025-04-07 11:26
bit	13.89 KB	-rw-r--r--	2025-04-07 11:26
bitset	44.92 KB	-rw-r--r--	2025-04-07 11:26
cassert	1.61 KB	-rw-r--r--	2025-04-07 11:26
ccomplex	1.3 KB	-rw-r--r--	2025-04-07 11:26
cctype	2.35 KB	-rw-r--r--	2025-04-07 11:26
cerrno	1.73 KB	-rw-r--r--	2025-04-07 11:26
cfenv	2 KB	-rw-r--r--	2025-04-07 11:26
cfloat	1.84 KB	-rw-r--r--	2025-04-07 11:26
charconv	20.49 KB	-rw-r--r--	2025-04-07 11:26
chrono	58.31 KB	-rw-r--r--	2025-04-07 11:26
cinttypes	2.11 KB	-rw-r--r--	2025-04-07 11:26
ciso646	1.43 KB	-rw-r--r--	2025-04-07 11:26
climits	1.87 KB	-rw-r--r--	2025-04-07 11:26
clocale	1.86 KB	-rw-r--r--	2025-04-07 11:26
cmath	48.23 KB	-rw-r--r--	2025-04-07 11:26
codecvt	5.15 KB	-rw-r--r--	2025-04-07 11:26
compare	36.77 KB	-rw-r--r--	2025-04-07 11:26
complex	54.63 KB	-rw-r--r--	2025-04-07 11:26
complex.h	1.56 KB	-rw-r--r--	2025-04-07 11:26
concepts	11.94 KB	-rw-r--r--	2025-04-07 11:26
condition_variable	12.48 KB	-rw-r--r--	2025-04-07 11:26
coroutine	9.58 KB	-rw-r--r--	2025-04-07 11:26
csetjmp	1.9 KB	-rw-r--r--	2025-04-07 11:26
csignal	1.81 KB	-rw-r--r--	2025-04-07 11:26
cstdalign	1.37 KB	-rw-r--r--	2025-04-07 11:26
cstdarg	1.82 KB	-rw-r--r--	2025-04-07 11:26
cstdbool	1.37 KB	-rw-r--r--	2025-04-07 11:26
cstddef	6.11 KB	-rw-r--r--	2025-04-07 11:26
cstdint	2.28 KB	-rw-r--r--	2025-04-07 11:26
cstdio	4.33 KB	-rw-r--r--	2025-04-07 11:26
cstdlib	6.18 KB	-rw-r--r--	2025-04-07 11:26
cstring	3.08 KB	-rw-r--r--	2025-04-07 11:26
ctgmath	1.33 KB	-rw-r--r--	2025-04-07 11:26
ctime	2.24 KB	-rw-r--r--	2025-04-07 11:26
cuchar	2.84 KB	-rw-r--r--	2025-04-07 11:26
cwchar	6.39 KB	-rw-r--r--	2025-04-07 11:26
cwctype	2.73 KB	-rw-r--r--	2025-04-07 11:26
cxxabi.h	21.76 KB	-rw-r--r--	2025-04-07 11:26
deque	4.35 KB	-rw-r--r--	2025-04-07 11:26
exception	5.29 KB	-rw-r--r--	2025-04-07 11:26
execution	1.76 KB	-rw-r--r--	2025-04-07 11:26
expected	33.11 KB	-rw-r--r--	2025-04-07 11:26
fenv.h	1.97 KB	-rw-r--r--	2025-04-07 11:26
filesystem	1.61 KB	-rw-r--r--	2025-04-07 11:26
forward_list	2.63 KB	-rw-r--r--	2025-04-07 11:26
fstream	39.73 KB	-rw-r--r--	2025-04-07 11:26
functional	42.3 KB	-rw-r--r--	2025-04-07 11:26
future	51.12 KB	-rw-r--r--	2025-04-07 11:26
initializer_list	2.97 KB	-rw-r--r--	2025-04-07 11:26
iomanip	16.16 KB	-rw-r--r--	2025-04-07 11:26
ios	1.56 KB	-rw-r--r--	2025-04-07 11:26
iosfwd	8.16 KB	-rw-r--r--	2025-04-07 11:26
iostream	2.63 KB	-rw-r--r--	2025-04-07 11:26
istream	33.3 KB	-rw-r--r--	2025-04-07 11:26
iterator	2.69 KB	-rw-r--r--	2025-04-07 11:26
latch	2.67 KB	-rw-r--r--	2025-04-07 11:26
limits	70.04 KB	-rw-r--r--	2025-04-07 11:26
list	3.57 KB	-rw-r--r--	2025-04-07 11:26
locale	1.45 KB	-rw-r--r--	2025-04-07 11:26
map	4.04 KB	-rw-r--r--	2025-04-07 11:26
math.h	4.47 KB	-rw-r--r--	2025-04-07 11:26
memory	4.44 KB	-rw-r--r--	2025-04-07 11:26
memory_resource	25.26 KB	-rw-r--r--	2025-04-07 11:26
mutex	25.4 KB	-rw-r--r--	2025-04-07 11:26
new	8.45 KB	-rw-r--r--	2025-04-07 11:26
numbers	6.1 KB	-rw-r--r--	2025-04-07 11:26
numeric	25.37 KB	-rw-r--r--	2025-04-07 11:26
optional	43.4 KB	-rw-r--r--	2025-04-07 11:26
ostream	25.31 KB	-rw-r--r--	2025-04-07 11:26
queue	2.41 KB	-rw-r--r--	2025-04-07 11:26
random	1.65 KB	-rw-r--r--	2025-04-07 11:26
ranges	117.31 KB	-rw-r--r--	2025-04-07 11:26
ratio	19.81 KB	-rw-r--r--	2025-04-07 11:26
regex	3.04 KB	-rw-r--r--	2025-04-07 11:26
scoped_allocator	17.24 KB	-rw-r--r--	2025-04-07 11:26
semaphore	2.93 KB	-rw-r--r--	2025-04-07 11:26
set	3.9 KB	-rw-r--r--	2025-04-07 11:26
shared_mutex	24.02 KB	-rw-r--r--	2025-04-07 11:26
source_location	2.68 KB	-rw-r--r--	2025-04-07 11:26
span	12.82 KB	-rw-r--r--	2025-04-07 11:26
spanstream	12.07 KB	-rw-r--r--	2025-04-07 11:26
sstream	38.31 KB	-rw-r--r--	2025-04-07 11:26
stack	2.33 KB	-rw-r--r--	2025-04-07 11:26
stacktrace	21.02 KB	-rw-r--r--	2025-04-07 11:26
stdatomic.h	3.97 KB	-rw-r--r--	2025-04-07 11:26
stdexcept	9.65 KB	-rw-r--r--	2025-04-07 11:26
stdlib.h	2.2 KB	-rw-r--r--	2025-04-07 11:26
stop_token	15.61 KB	-rw-r--r--	2025-04-07 11:26
streambuf	29.19 KB	-rw-r--r--	2025-04-07 11:26
string	4.69 KB	-rw-r--r--	2025-04-07 11:26
string_view	25.55 KB	-rw-r--r--	2025-04-07 11:26
syncstream	8.13 KB	-rw-r--r--	2025-04-07 11:26
system_error	14.65 KB	-rw-r--r--	2025-04-07 11:26
tgmath.h	1.33 KB	-rw-r--r--	2025-04-07 11:26
thread	6.68 KB	-rw-r--r--	2025-04-07 11:26
tuple	61.37 KB	-rw-r--r--	2025-04-07 11:26
type_traits	107.81 KB	-rw-r--r--	2025-04-07 11:26
typeindex	3.43 KB	-rw-r--r--	2025-04-07 11:26
typeinfo	8.11 KB	-rw-r--r--	2025-04-07 11:26
unordered_map	3.62 KB	-rw-r--r--	2025-04-07 11:26
unordered_set	3.48 KB	-rw-r--r--	2025-04-07 11:26
utility	6.89 KB	-rw-r--r--	2025-04-07 11:26
valarray	39.82 KB	-rw-r--r--	2025-04-07 11:26
variant	63.95 KB	-rw-r--r--	2025-04-07 11:26
vector	4.7 KB	-rw-r--r--	2025-04-07 11:26
version	10.98 KB	-rw-r--r--	2025-04-07 11:26

Rename

// <barrier> -*- C++ -*-

// Copyright (C) 2020-2022 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.

// You should have received a copy of the GNU General Public License along
// with this library; see the file COPYING3.  If not see
// <http://www.gnu.org/licenses/>.

// This implementation is based on libcxx/include/barrier
//===-- barrier.h --------------------------------------------------===//
//
// 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
//
//===---------------------------------------------------------------===//

/** @file include/barrier
 *  This is a Standard C++ Library header.
 */

#ifndef _GLIBCXX_BARRIER
#define _GLIBCXX_BARRIER 1

#pragma GCC system_header

#if __cplusplus > 201703L
#include <bits/atomic_base.h>
#if __cpp_lib_atomic_wait && __cpp_aligned_new
#include <bits/std_thread.h>
#include <bits/unique_ptr.h>

#include <array>

#define __cpp_lib_barrier 201907L

namespace std _GLIBCXX_VISIBILITY(default)
{
_GLIBCXX_BEGIN_NAMESPACE_VERSION

struct __empty_completion
  {
    _GLIBCXX_ALWAYS_INLINE void
    operator()() noexcept
    { }
  };

The default implementation of __tree_barrier is a classic tree barrier.

It looks different from literature pseudocode for two main reasons:
 1. Threads that call into std::barrier functions do not provide indices,
    so a numbering step is added before the actual barrier algorithm,
    appearing as an N+1 round to the N rounds of the tree barrier.
 2. A great deal of attention has been paid to avoid cache line thrashing
    by flattening the tree structure into cache-line sized arrays, that
    are indexed in an efficient way.

enum class __barrier_phase_t : unsigned char { };

template<typename _CompletionF>
    class __tree_barrier
    {
      using __atomic_phase_ref_t = std::__atomic_ref<__barrier_phase_t>;
      using __atomic_phase_const_ref_t = std::__atomic_ref<const __barrier_phase_t>;
      static constexpr auto __phase_alignment =
		      __atomic_phase_ref_t::required_alignment;

using __tickets_t = std::array<__barrier_phase_t, 64>;
      struct alignas(64) /* naturally-align the heap state */ __state_t
      {
	alignas(__phase_alignment) __tickets_t __tickets;
      };

ptrdiff_t _M_expected;
      unique_ptr<__state_t[]> _M_state;
      __atomic_base<ptrdiff_t> _M_expected_adjustment;
      _CompletionF _M_completion;

alignas(__phase_alignment) __barrier_phase_t  _M_phase;

bool
      _M_arrive(__barrier_phase_t __old_phase, size_t __current)
      {
	const auto __old_phase_val = static_cast<unsigned char>(__old_phase);
	const auto __half_step =
			   static_cast<__barrier_phase_t>(__old_phase_val + 1);
	const auto __full_step =
			   static_cast<__barrier_phase_t>(__old_phase_val + 2);

size_t __current_expected = _M_expected;
	__current %= ((_M_expected + 1) >> 1);

for (int __round = 0; ; ++__round)
	  {
	    if (__current_expected <= 1)
		return true;
	    size_t const __end_node = ((__current_expected + 1) >> 1),
			 __last_node = __end_node - 1;
	    for ( ; ; ++__current)
	      {
		if (__current == __end_node)
		  __current = 0;
		auto __expect = __old_phase;
		__atomic_phase_ref_t __phase(_M_state[__current]
						.__tickets[__round]);
		if (__current == __last_node && (__current_expected & 1))
		  {
		    if (__phase.compare_exchange_strong(__expect, __full_step,
						        memory_order_acq_rel))
		      break;     // I'm 1 in 1, go to next __round
		  }
		else if (__phase.compare_exchange_strong(__expect, __half_step,
						         memory_order_acq_rel))
		  {
		    return false; // I'm 1 in 2, done with arrival
		  }
		else if (__expect == __half_step)
		  {
		    if (__phase.compare_exchange_strong(__expect, __full_step,
						        memory_order_acq_rel))
		      break;    // I'm 2 in 2, go to next __round
		  }
	      }
	    __current_expected = __last_node + 1;
	    __current >>= 1;
	  }
      }

public:
      using arrival_token = __barrier_phase_t;

static constexpr ptrdiff_t
      max() noexcept
      { return __PTRDIFF_MAX__; }

__tree_barrier(ptrdiff_t __expected, _CompletionF __completion)
	  : _M_expected(__expected), _M_expected_adjustment(0),
	    _M_completion(move(__completion)),
	    _M_phase(static_cast<__barrier_phase_t>(0))
      {
	size_t const __count = (_M_expected + 1) >> 1;

_M_state = std::make_unique<__state_t[]>(__count);
      }

[[nodiscard]] arrival_token
      arrive(ptrdiff_t __update)
      {
	std::hash<std::thread::id> __hasher;
	size_t __current = __hasher(std::this_thread::get_id());
	__atomic_phase_ref_t __phase(_M_phase);
	const auto __old_phase = __phase.load(memory_order_relaxed);
	const auto __cur = static_cast<unsigned char>(__old_phase);
	for(; __update; --__update)
	  {
	    if(_M_arrive(__old_phase, __current))
	      {
		_M_completion();
		_M_expected += _M_expected_adjustment.load(memory_order_relaxed);
		_M_expected_adjustment.store(0, memory_order_relaxed);
		auto __new_phase = static_cast<__barrier_phase_t>(__cur + 2);
		__phase.store(__new_phase, memory_order_release);
		__phase.notify_all();
	      }
	  }
	return __old_phase;
      }

void
      wait(arrival_token&& __old_phase) const
      {
	__atomic_phase_const_ref_t __phase(_M_phase);
	auto const __test_fn = [=]
	  {
	    return __phase.load(memory_order_acquire) != __old_phase;
	  };
	std::__atomic_wait_address(&_M_phase, __test_fn);
      }

void
      arrive_and_drop()
      {
	_M_expected_adjustment.fetch_sub(1, memory_order_relaxed);
	(void)arrive(1);
      }
    };

template<typename _CompletionF = __empty_completion>
    class barrier
    {
      // Note, we may introduce a "central" barrier algorithm at some point
      // for more space constrained targets
      using __algorithm_t = __tree_barrier<_CompletionF>;
      __algorithm_t _M_b;

public:
      class arrival_token final
      {
      public:
	arrival_token(arrival_token&&) = default;
	arrival_token& operator=(arrival_token&&) = default;
	~arrival_token() = default;

private:
	friend class barrier;
	using __token = typename __algorithm_t::arrival_token;
	explicit arrival_token(__token __tok) noexcept : _M_tok(__tok) { }
	__token _M_tok;
      };

static constexpr ptrdiff_t
      max() noexcept
      { return __algorithm_t::max(); }

explicit
      barrier(ptrdiff_t __count, _CompletionF __completion = _CompletionF())
      : _M_b(__count, std::move(__completion))
      { }

barrier(barrier const&) = delete;
      barrier& operator=(barrier const&) = delete;

[[nodiscard]] arrival_token
      arrive(ptrdiff_t __update = 1)
      { return arrival_token{_M_b.arrive(__update)}; }

void
      wait(arrival_token&& __phase) const
      { _M_b.wait(std::move(__phase._M_tok)); }

void
      arrive_and_wait()
      { wait(arrive()); }

void
      arrive_and_drop()
      { _M_b.arrive_and_drop(); }
    };

_GLIBCXX_END_NAMESPACE_VERSION
} // namespace
#endif // __cpp_lib_atomic_wait && __cpp_aligned_new
#endif // __cplusplus > 201703L
#endif // _GLIBCXX_BARRIER