iOS之深入解析GCD的底层原理

一、队列① 队列实现源码分析在源码中搜索 dispatch_queue_create 关键字，可以在 queue.c 中发现：dispatch_queue_tdispatch_queue_create(const char *label, dispatch_queue_attr_t attr){return _dispatch_lan...

一、队列

① 队列实现源码分析

• 在源码中搜索 dispatch_queue_create 关键字，可以在 queue.c 中发现：

	dispatch_queue_t
	dispatch_queue_create(const char *label, dispatch_queue_attr_t attr)
	{
		return _dispatch_lane_create_with_target(label, attr,
				DISPATCH_TARGET_QUEUE_DEFAULT, true);
	}

• 进入 _dispatch_lane_create_with_target 中：

	DISPATCH_NOINLINE
	static dispatch_queue_t
	_dispatch_lane_create_with_target(const char *label, dispatch_queue_attr_t dqa,
	        dispatch_queue_t tq, bool legacy)
	{
	    // 创建 dqai
	    dispatch_queue_attr_info_t dqai = _dispatch_queue_attr_to_info(dqa);
	    
	    // 规范化参数，例如qos, overcommit, tq 和 Initialize the queue 
		...
	    
	    // 拼接队列名称
	    const void *vtable;
	    dispatch_queue_flags_t dqf = legacy ? DQF_MUTABLE : 0;
	    if (dqai.dqai_concurrent) { // vtable表示类的类型
	        // OS_dispatch_queue_concurrent
	        vtable = DISPATCH_VTABLE(queue_concurrent);
	    } else {
	        vtable = DISPATCH_VTABLE(queue_serial);
	    }
	    
	    ....
	    
	    // 创建队列，并初始化
	    dispatch_lane_t dq = _dispatch_object_alloc(vtable,
	            sizeof(struct dispatch_lane_s)); // alloc
	    // 根据dqai.dqai_concurrent的值，就能判断队列 是 串行 还是并发
	    _dispatch_queue_init(dq, dqf, dqai.dqai_concurrent ?
	            DISPATCH_QUEUE_WIDTH_MAX : 1, DISPATCH_QUEUE_ROLE_INNER |
	            (dqai.dqai_inactive ? DISPATCH_QUEUE_INACTIVE : 0)); // init
	    // 设置队列label标识符
	    dq->dq_label = label;// label赋值
	    dq->dq_priority = _dispatch_priority_make((dispatch_qos_t)dqai.dqai_qos, dqai.dqai_relpri);// 优先级处理
	    
	    ...
	    
	    // 类似于类与元类的绑定，不是直接的继承关系，而是类似于模型与模板的关系
	    dq->do_targetq = tq;
	    _dispatch_object_debug(dq, "%s", __func__);
	    return _dispatch_trace_queue_create(dq)._dq; // 研究dq
	}

• 通过 _dispatch_queue_attr_to_info 方法传入 dqa（即队列类型，串行、并发等）创建 dispatch_queue_attr_info_t 类型的对象 dqai，用于存储队列的相关属性信息：

	dispatch_queue_attr_info_t
	_dispatch_queue_attr_to_info(dispatch_queue_attr_t dqa)
	{
		dispatch_queue_attr_info_t dqai = { };
	
		if (!dqa) return dqai;
	
	#if DISPATCH_VARIANT_STATIC
		if (dqa == &_dispatch_queue_attr_concurrent) {
			dqai.dqai_concurrent = true;
			return dqai;
		}
	#endif
	
		if (dqa < _dispatch_queue_attrs ||
				dqa >= &_dispatch_queue_attrs[DISPATCH_QUEUE_ATTR_COUNT]) {
			DISPATCH_CLIENT_CRASH(dqa->do_vtable, "Invalid queue attribute");
		}
	
		size_t idx = (size_t)(dqa - _dispatch_queue_attrs);
	
		dqai.dqai_inactive = (idx % DISPATCH_QUEUE_ATTR_INACTIVE_COUNT);
		idx /= DISPATCH_QUEUE_ATTR_INACTIVE_COUNT;
	
		dqai.dqai_concurrent = !(idx % DISPATCH_QUEUE_ATTR_CONCURRENCY_COUNT);
		idx /= DISPATCH_QUEUE_ATTR_CONCURRENCY_COUNT;
	
		dqai.dqai_relpri = -(int)(idx % DISPATCH_QUEUE_ATTR_PRIO_COUNT);
		idx /= DISPATCH_QUEUE_ATTR_PRIO_COUNT;
	
		dqai.dqai_qos = idx % DISPATCH_QUEUE_ATTR_QOS_COUNT;
		idx /= DISPATCH_QUEUE_ATTR_QOS_COUNT;
	
		dqai.dqai_autorelease_frequency =
				idx % DISPATCH_QUEUE_ATTR_AUTORELEASE_FREQUENCY_COUNT;
		idx /= DISPATCH_QUEUE_ATTR_AUTORELEASE_FREQUENCY_COUNT;
	
		dqai.dqai_overcommit = idx % DISPATCH_QUEUE_ATTR_OVERCOMMIT_COUNT;
		idx /= DISPATCH_QUEUE_ATTR_OVERCOMMIT_COUNT;
	
		return dqai;
	}

• 设置队列相关联的属性，qos, overcommit, tq 和 Initialize the queue 的实现如下：

	//
	// Step 1: Normalize arguments (qos, overcommit, tq)
	//

	dispatch_qos_t qos = dqai.dqai_qos;
#if !HAVE_PTHREAD_WORKQUEUE_QOS
	if (qos == DISPATCH_QOS_USER_INTERACTIVE) {
		dqai.dqai_qos = qos = DISPATCH_QOS_USER_INITIATED;
	}
	if (qos == DISPATCH_QOS_MAINTENANCE) {
		dqai.dqai_qos = qos = DISPATCH_QOS_BACKGROUND;
	}
#endif // !HAVE_PTHREAD_WORKQUEUE_QOS

	_dispatch_queue_attr_overcommit_t overcommit = dqai.dqai_overcommit;
	if (overcommit != _dispatch_queue_attr_overcommit_unspecified && tq) {
		if (tq->do_targetq) {
			DISPATCH_CLIENT_CRASH(tq, "Cannot specify both overcommit and "
					"a non-global target queue");
		}
	}

	if (tq && dx_type(tq) == DISPATCH_QUEUE_GLOBAL_ROOT_TYPE) {
		// Handle discrepancies between attr and target queue, attributes win
		if (overcommit == _dispatch_queue_attr_overcommit_unspecified) {
			if (tq->dq_priority & DISPATCH_PRIORITY_FLAG_OVERCOMMIT) {
				overcommit = _dispatch_queue_attr_overcommit_enabled;
			} else {
				overcommit = _dispatch_queue_attr_overcommit_disabled;
			}
		}
		if (qos == DISPATCH_QOS_UNSPECIFIED) {
			qos = _dispatch_priority_qos(tq->dq_priority);
		}
		tq = NULL;
	} else if (tq && !tq->do_targetq) {
		// target is a pthread or runloop root queue, setting QoS or overcommit
		// is disallowed
		if (overcommit != _dispatch_queue_attr_overcommit_unspecified) {
			DISPATCH_CLIENT_CRASH(tq, "Cannot specify an overcommit attribute "
					"and use this kind of target queue");
		}
	} else {
		if (overcommit == _dispatch_queue_attr_overcommit_unspecified) {
			// Serial queues default to overcommit!
			overcommit = dqai.dqai_concurrent ?
					_dispatch_queue_attr_overcommit_disabled :
					_dispatch_queue_attr_overcommit_enabled;
		}
	}
	if (!tq) {
		tq = _dispatch_get_root_queue(
				qos == DISPATCH_QOS_UNSPECIFIED ? DISPATCH_QOS_DEFAULT : qos,
				overcommit == _dispatch_queue_attr_overcommit_enabled)->_as_dq;
		if (unlikely(!tq)) {
			DISPATCH_CLIENT_CRASH(qos, "Invalid queue attribute");
		}
	}

	//
	// Step 2: Initialize the queue
	//

	if (legacy) {
		// if any of these attributes is specified, use non legacy classes
		if (dqai.dqai_inactive || dqai.dqai_autorelease_frequency) {
			legacy = false;
		}
	}

• 通过 DISPATCH_VTABLE 拼接队列名称，即 vtable，其中 DISPATCH_VTABLE 是宏定义，队列的类型是通过 OS_dispatch + 队列类型 queue_concurrent 拼接而成的；
• • 创建以下队列DISPATCH_QUEUE_SERIA、DISPATCH_QUEUE_CONCURRENT、dispatch_get_main_queue、dispatch_get_global_queue：

	// OS_dispatch_queue_serial
    dispatch_queue_t serial = dispatch_queue_create("YDW", DISPATCH_QUEUE_SERIAL);
    // OS_dispatch_queue_concurrent
    // OS_dispatch_queue_concurrent
    dispatch_queue_t conque = dispatch_queue_create("YDW", DISPATCH_QUEUE_CONCURRENT);
    // DISPATCH_QUEUE_SERIAL max && 1
    // queue 对象 alloc init class
    dispatch_queue_t mainQueue = dispatch_get_main_queue();
    
    // 多个集合
    dispatch_queue_t globQueue = dispatch_get_global_queue(0, 0);
    
    NSLog(@"%@-%@-%@-%@", serial, conque, mainQueue, globQueue);

• • 串行队列类型：OS_dispatch_queue_serial，调试结果如下：

	(lldb) po object_getClass(serial)
	OS_dispatch_queue_serial

• • 并发队列类型：OS_dispatch_queue_concurrent，调试结果如下：

	(lldb) po object_getClass(conque)
	OS_dispatch_queue_concurrent

• • DISPATCH 的定义如下：

	#define DISPATCH_VTABLE(name) DISPATCH_OBJC_CLASS(name)

	#define DISPATCH_OBJC_CLASS(name)   (&DISPATCH_CLASS_SYMBOL(name))

	#define DISPATCH_CLASS(name) OS_dispatch_##name

• 通过 alloc+init 初始化队列，即 dq，其中在 _dispatch_queue_init 传参中根据dqai.dqai_concurrent 的布尔值，就能判断队列是串行还是并发，我们知道 vtable 表示的是队列类型，因此也可以说明队列也是对象；
• • 进入 _dispatch_object_alloc -> _os_object_alloc_realized 方法中，可以看到设置了 isa 的指向，这里也可以验证队列也是对象：

	inline _os_object_t
	_os_object_alloc_realized(const void *cls, size_t size)
	{
		_os_object_t obj;
		dispatch_assert(size >= sizeof(struct _os_object_s));
		while (unlikely(!(obj = calloc(1u, size)))) {
			_dispatch_temporary_resource_shortage();
		}
		// 设置 isa 指向
		obj->os_obj_isa = cls;
		return obj;
	}

• 进入 _dispatch_queue_init 方法，队列类型是 dispatch_queue_t，并设置队列的相关属性：

	// Note to later developers: ensure that any initialization changes are
	// made for statically allocated queues (i.e. _dispatch_main_q).
	static inline dispatch_queue_class_t
	_dispatch_queue_init(dispatch_queue_class_t dqu, dispatch_queue_flags_t dqf,
			uint16_t width, uint64_t initial_state_bits)
	{
		uint64_t dq_state = DISPATCH_QUEUE_STATE_INIT_VALUE(width);
		dispatch_queue_t dq = dqu._dq;
	
		dispatch_assert((initial_state_bits & ~(DISPATCH_QUEUE_ROLE_MASK |
				DISPATCH_QUEUE_INACTIVE)) == 0);
	
		if (initial_state_bits & DISPATCH_QUEUE_INACTIVE) {
			dq->do_ref_cnt += 2; // rdar://8181908 see _dispatch_lane_resume
			if (dx_metatype(dq) == _DISPATCH_SOURCE_TYPE) {
				dq->do_ref_cnt++; // released when DSF_DELETED is set
			}
		}
	
		dq_state |= initial_state_bits;
		dq->do_next = DISPATCH_OBJECT_LISTLESS;
		// 串行还是并发
		dqf |= DQF_WIDTH(width);
		os_atomic_store2o(dq, dq_atomic_flags, dqf, relaxed);
		dq->dq_state = dq_state;
		dq->dq_serialnum =
				os_atomic_inc_orig(&_dispatch_queue_serial_numbers, relaxed);
		return dqu;
	}

• 通过 _dispatch_trace_queue_create 对创建的队列进行处理，其中_dispatch_trace_queue_create 是 _dispatch_introspection_queue_create 封装的宏定义，最后会返回处理过的_dq；

	dispatch_queue_class_t
	_dispatch_introspection_queue_create(dispatch_queue_t dq)
	{
		dispatch_queue_introspection_context_t dqic;
		size_t sz = sizeof(struct dispatch_queue_introspection_context_s);
	
		if (!_dispatch_introspection.debug_queue_inversions) {
			sz = offsetof(struct dispatch_queue_introspection_context_s,
					__dqic_no_queue_inversion);
		}
		dqic = _dispatch_calloc(1, sz);
		dqic->dqic_queue._dq = dq;
		if (_dispatch_introspection.debug_queue_inversions) {
			LIST_INIT(&dqic->dqic_order_top_head);
			LIST_INIT(&dqic->dqic_order_bottom_head);
		}
		dq->do_finalizer = dqic;
	
		_dispatch_unfair_lock_lock(&_dispatch_introspection.queues_lock);
		LIST_INSERT_HEAD(&_dispatch_introspection.queues, dqic, dqic_list);
		_dispatch_unfair_lock_unlock(&_dispatch_introspection.queues_lock);
	
		DISPATCH_INTROSPECTION_INTERPOSABLE_HOOK_CALLOUT(queue_create, dq);
		if (DISPATCH_INTROSPECTION_HOOK_ENABLED(queue_create)) {
			_dispatch_introspection_queue_create_hook(dq);
		}
		return upcast(dq)._dqu;
	}

• 进入 _dispatch_introspection_queue_create_hook -> dispatch_introspection_queue_get_info -> _dispatch_introspection_lane_get_info 中可以看出，与我们自定义的类还是有区别的，创建队列在底层的实现是通过模板创建的：

	DISPATCH_ALWAYS_INLINE
	static inline dispatch_introspection_queue_s
	_dispatch_introspection_lane_get_info(dispatch_lane_class_t dqu)
	{
		dispatch_lane_t dq = dqu._dl;
		bool global = _dispatch_object_is_global(dq);
		uint64_t dq_state = os_atomic_load2o(dq, dq_state, relaxed);
	
		dispatch_introspection_queue_s diq = {
			.queue = dq->_as_dq,
			.target_queue = dq->do_targetq,
			.label = dq->dq_label,
			.serialnum = dq->dq_serialnum,
			.width = dq->dq_width,
			.suspend_count = _dq_state_suspend_cnt(dq_state) + dq->dq_side_suspend_cnt,
			.enqueued = _dq_state_is_enqueued(dq_state) && !global,
			.barrier = _dq_state_is_in_barrier(dq_state) && !global,
			.draining = (dq->dq_items_head == (void*)~0ul) ||
					(!dq->dq_items_head && dq->dq_items_tail),
			.global = global,
			.main = dx_type(dq) == DISPATCH_QUEUE_MAIN_TYPE,
		};
		return diq;
	}

② 队列实现流程总结

• 队列创建方法 dispatch_queue_create 中的参数二（即队列类型），决定了下层中 max & 1（用于区分是串行还是并发），其中1表示串行；
• queue 也是一个对象，也需要底层通过 alloc + init 创建，并且在 alloc 中也有一个class，这个 class 是通过宏定义拼接而成，并且同时会指定 isa 的指向；
• 创建队列在底层的处理是通过模板创建的，其类型是 dispatch_introspection_queue_s 结构体。

③ 队列实现流程示意图

二、函数

① 异步函数

• 进入 dispatch_async 的源码实现，主要有两个函数：_dispatch_continuation_init 任务包装函数，_dispatch_continuation_async 并发处理函数，如下所示：

	#ifdef __BLOCKS__
	void
	dispatch_async(dispatch_queue_t dq, dispatch_block_t work)
	{
		dispatch_continuation_t dc = _dispatch_continuation_alloc();
		uintptr_t dc_flags = DC_FLAG_CONSUME;
		dispatch_qos_t qos;
	
		qos = _dispatch_continuation_init(dc, dq, work, 0, dc_flags);
		_dispatch_continuation_async(dq, dc, qos, dc->dc_flags);
	}
	#endif

• 进入 _dispatch_continuation_init 源码实现，可以看到，主要业务是包装任务，并设置线程的回程函数，相当于初始化：

	DISPATCH_ALWAYS_INLINE
	static inline dispatch_qos_t
	_dispatch_continuation_init(dispatch_continuation_t dc,
	        dispatch_queue_class_t dqu, dispatch_block_t work,
	        dispatch_block_flags_t flags, uintptr_t dc_flags)
	{
		// 拷贝任务
	    void *ctxt = _dispatch_Block_copy(work);
	
	    dc_flags |= DC_FLAG_BLOCK | DC_FLAG_ALLOCATED;
	    if (unlikely(_dispatch_block_has_private_data(work))) {
	        dc->dc_flags = dc_flags;
	        // 赋值
	        dc->dc_ctxt = ctxt;
	        // will initialize all fields but requires dc_flags & dc_ctxt to be set
	        return _dispatch_continuation_init_slow(dc, dqu, flags);
	    }
		// 封装work - 异步回调
	    dispatch_function_t func = _dispatch_Block_invoke(work);
	    if (dc_flags & DC_FLAG_CONSUME) {
	    	// 回调函数赋值 - 同步回调
	        func = _dispatch_call_block_and_release;
	    }
	    return _dispatch_continuation_init_f(dc, dqu, ctxt, func, flags, dc_flags);
	}

• 分析说明：
• • 通过 _dispatch_Block_copy 拷贝任务；
• • 通过 _dispatch_Block_invoke 封装任务，其中 _dispatch_Block_invoke 是个宏定义，根据以上分析得知是异步回调：

	#define _dispatch_Block_invoke(bb) \
        ((dispatch_function_t)((struct Block_layout *)bb)->invoke)

• • 如果是同步的，则回调函数赋值为 _dispatch_call_block_and_release；
• • 通过 _dispatch_continuation_init_f 方法将回调函数赋值，即 f 就是 func，将其保存在属性中：

	DISPATCH_ALWAYS_INLINE
	static inline dispatch_qos_t
	_dispatch_continuation_init_f(dispatch_continuation_t dc,
			dispatch_queue_class_t dqu, void *ctxt, dispatch_function_t f,
			dispatch_block_flags_t flags, uintptr_t dc_flags)
	{
		pthread_priority_t pp = 0;
		dc->dc_flags = dc_flags | DC_FLAG_ALLOCATED;
		dc->dc_func = f;
		dc->dc_ctxt = ctxt;
		// in this context DISPATCH_BLOCK_HAS_PRIORITY means that the priority
		// should not be propagated, only taken from the handler if it has one
		if (!(flags & DISPATCH_BLOCK_HAS_PRIORITY)) {
			pp = _dispatch_priority_propagate();
		}
		_dispatch_continuation_voucher_set(dc, flags);
		return _dispatch_continuation_priority_set(dc, dqu, pp, flags);
	}

• 进入 _dispatch_continuation_async 的源码实现，如下所示：

	DISPATCH_ALWAYS_INLINE
	static inline void
	_dispatch_continuation_async(dispatch_queue_class_t dqu,
	        dispatch_continuation_t dc, dispatch_qos_t qos, uintptr_t dc_flags)
	{
	#if DISPATCH_INTROSPECTION
	    if (!(dc_flags & DC_FLAG_NO_INTROSPECTION)) {
	        _dispatch_trace_item_push(dqu, dc); // 跟踪日志
	    }
	#else
	    (void)dc_flags;
	#endif
	    return dx_push(dqu._dq, dc, qos); // 与dx_invoke一样，都是宏
	}

• • 其中，最主要的是 dx_push(dqu._dq, dc, qos)，dx_push 是宏定义，如下所示：

	#define dx_push(x, y, z) dx_vtable(x)->dq_push(x, y, z)

• • dq_push 需要根据队列的类型，执行不同的函数：

	DISPATCH_VTABLE_INSTANCE(workloop,
		.do_type        = DISPATCH_WORKLOOP_TYPE,
		.do_dispose     = _dispatch_workloop_dispose,
		.do_debug       = _dispatch_queue_debug,
		.do_invoke      = _dispatch_workloop_invoke,
	
		.dq_activate    = _dispatch_queue_no_activate,
		.dq_wakeup      = _dispatch_workloop_wakeup,
		.dq_push        = _dispatch_workloop_push,
	);
	
	DISPATCH_VTABLE_SUBCLASS_INSTANCE(queue_serial, lane,
		.do_type        = DISPATCH_QUEUE_SERIAL_TYPE,
		.do_dispose     = _dispatch_lane_dispose,
		.do_debug       = _dispatch_queue_debug,
		.do_invoke      = _dispatch_lane_invoke,
	
		.dq_activate    = _dispatch_lane_activate,
		.dq_wakeup      = _dispatch_lane_wakeup,
		.dq_push        = _dispatch_lane_push,
	);
	
	DISPATCH_VTABLE_SUBCLASS_INSTANCE(queue_concurrent, lane,
		.do_type        = DISPATCH_QUEUE_CONCURRENT_TYPE,
		.do_dispose     = _dispatch_lane_dispose,
		.do_debug       = _dispatch_queue_debug,
		.do_invoke      = _dispatch_lane_invoke,
	
		.dq_activate    = _dispatch_lane_activate,
		.dq_wakeup      = _dispatch_lane_wakeup,
		.dq_push        = _dispatch_lane_concurrent_push,
	);
	DISPATCH_VTABLE_SUBCLASS_INSTANCE(queue_global, lane,
		.do_type        = DISPATCH_QUEUE_GLOBAL_ROOT_TYPE,
		.do_dispose     = _dispatch_object_no_dispose,
		.do_debug       = _dispatch_queue_debug,
		.do_invoke      = _dispatch_object_no_invoke,
	
		.dq_activate    = _dispatch_queue_no_activate,
		.dq_wakeup      = _dispatch_root_queue_wakeup,
		.dq_push        = _dispatch_root_queue_push,
	);

	#if DISPATCH_USE_PTHREAD_ROOT_QUEUES
	DISPATCH_VTABLE_SUBCLASS_INSTANCE(queue_pthread_root, lane,
		.do_type        = DISPATCH_QUEUE_PTHREAD_ROOT_TYPE,
		.do_dispose     = _dispatch_pthread_root_queue_dispose,
		.do_debug       = _dispatch_queue_debug,
		.do_invoke      = _dispatch_object_no_invoke,
	
		.dq_activate    = _dispatch_queue_no_activate,
		.dq_wakeup      = _dispatch_root_queue_wakeup,
		.dq_push        = _dispatch_root_queue_push,
	);
	#endif // DISPATCH_USE_PTHREAD_ROOT_QUEUES

		DISPATCH_VTABLE_SUBCLASS_INSTANCE(queue_mgr, lane,
		.do_type        = DISPATCH_QUEUE_MGR_TYPE,
		.do_dispose     = _dispatch_object_no_dispose,
		.do_debug       = _dispatch_queue_debug,
	#if DISPATCH_USE_MGR_THREAD
		.do_invoke      = _dispatch_mgr_thread,
	#else
		.do_invoke      = _dispatch_object_no_invoke,
	#endif
	
		.dq_activate    = _dispatch_queue_no_activate,
		.dq_wakeup      = _dispatch_mgr_queue_wakeup,
		.dq_push        = _dispatch_mgr_queue_push,
	);
	
	DISPATCH_VTABLE_SUBCLASS_INSTANCE(queue_main, lane,
		.do_type        = DISPATCH_QUEUE_MAIN_TYPE,
		.do_dispose     = _dispatch_lane_dispose,
		.do_debug       = _dispatch_queue_debug,
		.do_invoke      = _dispatch_lane_invoke,
	
		.dq_activate    = _dispatch_queue_no_activate,
		.dq_wakeup      = _dispatch_main_queue_wakeup,
		.dq_push        = _dispatch_main_queue_push,
	);
	
	#if DISPATCH_COCOA_COMPAT
	DISPATCH_VTABLE_SUBCLASS_INSTANCE(queue_runloop, lane,
		.do_type        = DISPATCH_QUEUE_RUNLOOP_TYPE,
		.do_dispose     = _dispatch_runloop_queue_dispose,
		.do_debug       = _dispatch_queue_debug,
		.do_invoke      = _dispatch_lane_invoke,
	
		.dq_activate    = _dispatch_queue_no_activate,
		.dq_wakeup      = _dispatch_runloop_queue_wakeup,
		.dq_push        = _dispatch_lane_push,
	);
	#endif
	
	DISPATCH_VTABLE_INSTANCE(source,
		.do_type        = DISPATCH_SOURCE_KEVENT_TYPE,
		.do_dispose     = _dispatch_source_dispose,
		.do_debug       = _dispatch_source_debug,
		.do_invoke      = _dispatch_source_invoke,
	
		.dq_activate    = _dispatch_source_activate,
		.dq_wakeup      = _dispatch_source_wakeup,
		.dq_push        = _dispatch_lane_push,
	);
	
	DISPATCH_VTABLE_INSTANCE(channel,
		.do_type        = DISPATCH_CHANNEL_TYPE,
		.do_dispose     = _dispatch_channel_dispose,
		.do_debug       = _dispatch_channel_debug,
		.do_invoke      = _dispatch_channel_invoke,
	
		.dq_activate    = _dispatch_lane_activate,
		.dq_wakeup      = _dispatch_channel_wakeup,
		.dq_push        = _dispatch_lane_push,
	);
	
	#if HAVE_MACH
	DISPATCH_VTABLE_INSTANCE(mach,
		.do_type        = DISPATCH_MACH_CHANNEL_TYPE,
		.do_dispose     = _dispatch_mach_dispose,
		.do_debug       = _dispatch_mach_debug,
		.do_invoke      = _dispatch_mach_invoke,
	
		.dq_activate    = _dispatch_mach_activate,
		.dq_wakeup      = _dispatch_mach_wakeup,
		.dq_push        = _dispatch_lane_push,
	);
	#endif // HAVE_MACH

• 调试执行函数
• • 由于是并发队列，可以通过增加 _dispatch_lane_concurrent_push 符号断点，看看是否会执行：

	dispatch_queue_t conque = dispatch_queue_create("com.YDW.Queue", DISPATCH_QUEUE_CONCURRENT);
	dispatch_async(conque, ^{
	    NSLog(@"异步函数");
	});

• 运行可以看到，确实执行了 _dispatch_lane_concurrent_push；

• 我们进入 _dispatch_lane_concurrent_push 源码，可以发现有两步，继续通过符号断点 _dispatch_continuation_redirect_push 和 _dispatch_lane_push 调试，可以看到 执行了 _dispatch_continuation_redirect_push；

• 继续进入 _dispatch_continuation_redirect_push 源码，继续执行了 dx_push，即产生了“递归”，综合前面队列创建时可知，队列也是一个对象，有父类、根类，所以会递归执行到根类的方法：

	DISPATCH_NOINLINE
	static void
	_dispatch_continuation_redirect_push(dispatch_lane_t dl,
			dispatch_object_t dou, dispatch_qos_t qos)
	{
		if (likely(!_dispatch_object_is_redirection(dou))) {
			dou._dc = _dispatch_async_redirect_wrap(dl, dou);
		} else if (!dou._dc->dc_ctxt) {
			// find first queue in descending target queue order that has
			// an autorelease frequency set, and use that as the frequency for
			// this continuation.
			dou._dc->dc_ctxt = (void *)
			(uintptr_t)_dispatch_queue_autorelease_frequency(dl);
		}
	
		dispatch_queue_t dq = dl->do_targetq;
		if (!qos) qos = _dispatch_priority_qos(dq->dq_priority);
		// 递归
		dx_push(dq, dou, qos);
	}

• 接下来，通过根类的 _dispatch_root_queue_push 符号断点，来验证猜想是否正确，从运行结果看出，猜想完全正确。

• 继续进入 _dispatch_root_queue_push -> _dispatch_root_queue_push_inline ->_dispatch_root_queue_poke -> _dispatch_root_queue_poke_slow 源码，符号断点继续可以验证，查看该方法的源码实现，主要有两步：
• • 通过 _dispatch_root_queues_init 方法注册回调；
• • 通过 do-while 循环创建线程，使用 pthread_create 方法；

	DISPATCH_NOINLINE
	static void
	_dispatch_root_queue_poke_slow(dispatch_queue_global_t dq, int n, int floor)
	{
	    int remaining = n;
	    int r = ENOSYS;
	
	    _dispatch_root_queues_init();
	    
	    ...
	    // do-while循环创建线程
	    do {
	        _dispatch_retain(dq); // released in _dispatch_worker_thread
	        while ((r = pthread_create(pthr, attr, _dispatch_worker_thread, dq))) {
	            if (r != EAGAIN) {
	                (void)dispatch_assume_zero(r);
	            }
	            _dispatch_temporary_resource_shortage();
	        }
	    } while (--remaining);
	    
	    ...
	}

• 进入 _dispatch_root_queues_init 源码实现，发现它是一个 dispatch_once_f 单例，其中传入的 func 是 _dispatch_root_queues_init_once：

	DISPATCH_ALWAYS_INLINE
	static inline void
	_dispatch_root_queues_init(void)
	{
	    dispatch_once_f(&_dispatch_root_queues_pred, NULL, _dispatch_root_queues_init_once);
	}

• 进入 _dispatch_root_queues_init_once 的源码，其内部不同事务的调用句柄都是_dispatch_worker_thread2：

	#pragma clang diagnostic push
	#pragma clang diagnostic ignored "-Wunreachable-code"
		if (unlikely(!_dispatch_kevent_workqueue_enabled)) {
	#if DISPATCH_USE_KEVENT_SETUP
			cfg.workq_cb = _dispatch_worker_thread2;
			r = pthread_workqueue_setup(&cfg, sizeof(cfg));
	#else
			r = _pthread_workqueue_init(_dispatch_worker_thread2,
					offsetof(struct dispatch_queue_s, dq_serialnum), 0);
	#endif // DISPATCH_USE_KEVENT_SETUP
	#if DISPATCH_USE_KEVENT_WORKLOOP
		} else if (wq_supported & WORKQ_FEATURE_WORKLOOP) {
	#if DISPATCH_USE_KEVENT_SETUP
			cfg.workq_cb = _dispatch_worker_thread2;
			cfg.kevent_cb = (pthread_workqueue_function_kevent_t) _dispatch_kevent_worker_thread;
			cfg.workloop_cb = (pthread_workqueue_function_workloop_t) _dispatch_workloop_worker_thread;
			r = pthread_workqueue_setup(&cfg, sizeof(cfg));
	#else
			r = _pthread_workqueue_init_with_workloop(_dispatch_worker_thread2,
					(pthread_workqueue_function_kevent_t)
					_dispatch_kevent_worker_thread,
					(pthread_workqueue_function_workloop_t)
					_dispatch_workloop_worker_thread,
					offsetof(struct dispatch_queue_s, dq_serialnum), 0);
	#endif // DISPATCH_USE_KEVENT_SETUP
	#endif // DISPATCH_USE_KEVENT_WORKLOOP
	#if DISPATCH_USE_KEVENT_WORKQUEUE
		} else if (wq_supported & WORKQ_FEATURE_KEVENT) {
	#if DISPATCH_USE_KEVENT_SETUP
			cfg.workq_cb = _dispatch_worker_thread2;
			cfg.kevent_cb = (pthread_workqueue_function_kevent_t) _dispatch_kevent_worker_thread;
			r = pthread_workqueue_setup(&cfg, sizeof(cfg));
	#else
			r = _pthread_workqueue_init_with_kevent(_dispatch_worker_thread2,
					(pthread_workqueue_function_kevent_t)
					_dispatch_kevent_worker_thread,
					offsetof(struct dispatch_queue_s, dq_serialnum), 0);
	#endif // DISPATCH_USE_KEVENT_SETUP
	#endif
		} else {
			DISPATCH_INTERNAL_CRASH(wq_supported, "Missing Kevent WORKQ support");
		}
	#pragma clang diagnostic pop
	
		if (r != 0) {
			DISPATCH_INTERNAL_CRASH((r << 16) | wq_supported,
					"Root queue initialization failed");
		}
	#endif // DISPATCH_USE_INTERNAL_WORKQUEUE
	}

• 其block回调执行的流程为：_dispatch_root_queues_init_once ->_dispatch_worker_thread2 -> _dispatch_root_queue_drain -> _dispatch_root_queue_drain -> _dispatch_continuation_pop_inline -> _dispatch_continuation_invoke_inline -> _dispatch_client_callout -> dispatch_call_block_and_release，通过 bt 打印堆栈信息，如下：

	(lldb) bt
	* thread #3, queue = 'com.YDW.Queue', stop reason = breakpoint 1.1
	  * frame #0: 0x00000001073d6367 函数与队列`__29-[ViewController viewDidLoad]_block_invoke(.block_descriptor=0x00000001073d9108) at ViewController.m:23:9
	    frame #1: 0x00000001076477ec libdispatch.dylib`_dispatch_call_block_and_release + 12
	    frame #2: 0x00000001076489c8 libdispatch.dylib`_dispatch_client_callout + 8
	    frame #3: 0x000000010764b316 libdispatch.dylib`_dispatch_continuation_pop + 557
	    frame #4: 0x000000010764a71c libdispatch.dylib`_dispatch_async_redirect_invoke + 779
	    frame #5: 0x000000010765a508 libdispatch.dylib`_dispatch_root_queue_drain + 351
	    frame #6: 0x000000010765ae6d libdispatch.dylib`_dispatch_worker_thread2 + 135
	    frame #7: 0x00007fff611639f7 libsystem_pthread.dylib`_pthread_wqthread + 220
	    frame #8: 0x00007fff61162b77 libsystem_pthread.dylib`start_wqthread + 15
	(lldb) 

• 特别说明：单例的 block 回调和异步函数的 block 回调是不同的：
  • 单例中，block 回调中的 func 是 _dispatch_Block_invoke(block)；
  • 而异步函数中，block 回调中的 func 是 dispatch_call_block_and_release。
• 综上所述，异步函数的底层实现：将异步任务拷贝并封装，并设置回调函数 func，再通过 dx_push 递归，会重定向到根队列，然后通过 pthread_creat 创建线程，最后通过 dx_invoke 执行 block 回调（注意 dx_push 和 dx_invoke 是成对的）。

② 同步函数

• 进入 dispatch_sync 源码实现，其底层的实现是通过栅栏函数实现的：

	DISPATCH_NOINLINE
	void
	dispatch_sync(dispatch_queue_t dq, dispatch_block_t work)
	{
	    uintptr_t dc_flags = DC_FLAG_BLOCK;
	    if (unlikely(_dispatch_block_has_private_data(work))) {
	        return _dispatch_sync_block_with_privdata(dq, work, dc_flags);
	    }
	    _dispatch_sync_f(dq, work, _dispatch_Block_invoke(work), dc_flags);
	}

• 进入 _dispatch_sync_f 源码：

	DISPATCH_NOINLINE
	static void
	_dispatch_sync_f(dispatch_queue_t dq, void *ctxt, dispatch_function_t func,
			uintptr_t dc_flags)
	{
		_dispatch_sync_f_inline(dq, ctxt, func, dc_flags);
	}

• 查看 _dispatch_sync_f_inline 源码，其中 width = 1 表示是串行队列；
  • 栅栏：_dispatch_barrier_sync_f，可以得出同步函数的底层实现其实是同步栅栏函数；
  • 死锁：_dispatch_sync_f_slow，如果存在相互等待的情况，就会造成死锁。

	DISPATCH_ALWAYS_INLINE
	static inline void
	_dispatch_sync_f_inline(dispatch_queue_t dq, void *ctxt,
	        dispatch_function_t func, uintptr_t dc_flags)
	{
	    if (likely(dq->dq_width == 1)) {// 表示是串行队列
	        return _dispatch_barrier_sync_f(dq, ctxt, func, dc_flags);//栅栏
	    }
	
	    if (unlikely(dx_metatype(dq) != _DISPATCH_LANE_TYPE)) {
	        DISPATCH_CLIENT_CRASH(0, "Queue type doesn't support dispatch_sync");
	    }
	
	    dispatch_lane_t dl = upcast(dq)._dl;
	    // Global concurrent queues and queues bound to non-dispatch threads
	    // always fall into the slow case, see DISPATCH_ROOT_QUEUE_STATE_INIT_VALUE
	    if (unlikely(!_dispatch_queue_try_reserve_sync_width(dl))) {
	        return _dispatch_sync_f_slow(dl, ctxt, func, 0, dl, dc_flags);//死锁
	    }
	
	    if (unlikely(dq->do_targetq->do_targetq)) {
	        return _dispatch_sync_recurse(dl, ctxt, func, dc_flags);
	    }
	    _dispatch_introspection_sync_begin(dl); // 处理当前信息
	    _dispatch_sync_invoke_and_complete(dl, ctxt, func DISPATCH_TRACE_ARG(
	            _dispatch_trace_item_sync_push_pop(dq, ctxt, func, dc_flags))); // block执行并释放
	}

• 进入 _dispatch_sync_f_slow，当前的主队列是挂起阻塞：

	DISPATCH_NOINLINE
	static void
	_dispatch_sync_f_slow(dispatch_queue_class_t top_dqu, void *ctxt,
			dispatch_function_t func, uintptr_t top_dc_flags,
			dispatch_queue_class_t dqu, uintptr_t dc_flags)
	{
		dispatch_queue_t top_dq = top_dqu._dq;
		dispatch_queue_t dq = dqu._dq;
		if (unlikely(!dq->do_targetq)) {
			return _dispatch_sync_function_invoke(dq, ctxt, func);
		}
	
		pthread_priority_t pp = _dispatch_get_priority();
		struct dispatch_sync_context_s dsc = {
			.dc_flags    = DC_FLAG_SYNC_WAITER | dc_flags,
			.dc_func     = _dispatch_async_and_wait_invoke,
			.dc_ctxt     = &dsc,
			.dc_other    = top_dq,
			.dc_priority = pp | _PTHREAD_PRIORITY_ENFORCE_FLAG,
			.dc_voucher  = _voucher_get(),
			.dsc_func    = func,
			.dsc_ctxt    = ctxt,
			.dsc_waiter  = _dispatch_tid_self(),
		};
	
		_dispatch_trace_item_push(top_dq, &dsc);
		__DISPATCH_WAIT_FOR_QUEUE__(&dsc, dq);
	
		if (dsc.dsc_func == NULL) {
			dispatch_queue_t stop_dq = dsc.dc_other;
			return _dispatch_sync_complete_recurse(top_dq, stop_dq, top_dc_flags);
		}
	
		_dispatch_introspection_sync_begin(top_dq);
		_dispatch_trace_item_pop(top_dq, &dsc);
		_dispatch_sync_invoke_and_complete_recurse(top_dq, ctxt, func,top_dc_flags
				DISPATCH_TRACE_ARG(&dsc));
	}

• 往一个队列中加入任务，会 push 加入主队列，进入 _dispatch_trace_item_push：

	DISPATCH_ALWAYS_INLINE
	static inline void
	_dispatch_trace_item_push(dispatch_queue_class_t dqu, dispatch_object_t _tail)
	{
		if (unlikely(DISPATCH_QUEUE_PUSH_ENABLED())) {
			_dispatch_trace_continuation(dqu._dq, _tail._do, DISPATCH_QUEUE_PUSH);
		}
	
		_dispatch_trace_item_push_inline(dqu._dq, _tail._do);
		_dispatch_introspection_queue_push(dqu, _tail);
	}

• 进入__DISPATCH_WAIT_FOR_QUEUE__，判断dq是否为正在等待的队列，然后给出一个状态state，然后将dq的状态和当前任务依赖的队列进行匹配：

	DISPATCH_NOINLINE
	static void
	__DISPATCH_WAIT_FOR_QUEUE__(dispatch_sync_context_t dsc, dispatch_queue_t dq)
	{
		// 判断dq是否为正在等待的队列
		uint64_t dq_state = _dispatch_wait_prepare(dq);
		if (unlikely(_dq_state_drain_locked_by(dq_state, dsc->dsc_waiter))) {
			DISPATCH_CLIENT_CRASH((uintptr_t)dq_state,
					"dispatch_sync called on queue "
					"already owned by current thread");
		}
	
		// Blocks submitted to the main thread MUST run on the main thread, and
		// dispatch_async_and_wait also executes on the remote context rather than
		// the current thread.
		//
		// For both these cases we need to save the frame linkage for the sake of
		// _dispatch_async_and_wait_invoke
		_dispatch_thread_frame_save_state(&dsc->dsc_dtf);
	
		if (_dq_state_is_suspended(dq_state) ||
				_dq_state_is_base_anon(dq_state)) {
			dsc->dc_data = DISPATCH_WLH_ANON;
		} else if (_dq_state_is_base_wlh(dq_state)) {
			dsc->dc_data = (dispatch_wlh_t)dq;
		} else {
			_dispatch_wait_compute_wlh(upcast(dq)._dl, dsc);
		}
	
		if (dsc->dc_data == DISPATCH_WLH_ANON) {
			dsc->dsc_override_qos_floor = dsc->dsc_override_qos =
					(uint8_t)_dispatch_get_basepri_override_qos_floor();
			_dispatch_thread_event_init(&dsc->dsc_event);
		}
		dx_push(dq, dsc, _dispatch_qos_from_pp(dsc->dc_priority));
		_dispatch_trace_runtime_event(sync_wait, dq, 0);
		if (dsc->dc_data == DISPATCH_WLH_ANON) {
			_dispatch_thread_event_wait(&dsc->dsc_event); // acquire
		} else {
			_dispatch_event_loop_wait_for_ownership(dsc);
		}
		if (dsc->dc_data == DISPATCH_WLH_ANON) {
			_dispatch_thread_event_destroy(&dsc->dsc_event);
			// If _dispatch_sync_waiter_wake() gave this thread an override,
			// ensure that the root queue sees it.
			if (dsc->dsc_override_qos > dsc->dsc_override_qos_floor) {
				_dispatch_set_basepri_override_qos(dsc->dsc_override_qos);
			}
		}
	}

• 继续进入 _dq_state_drain_locked_by -> _dispatch_lock_is_locked_by 源码，如果当前等待的和正在执行的是同一个队列，即判断线程ID是否相等，如果相等，则会造成死锁，如下：

	DISPATCH_ALWAYS_INLINE
	static inline bool
	_dispatch_lock_is_locked_by(dispatch_lock lock_value, dispatch_tid tid)
	{
	    // equivalent to _dispatch_lock_owner(lock_value) == tid
	    // 异或操作：相同为0，不同为1，如果相同，则为0，0 &任何数都为0
	    // 即判断 当前要等待的任务 和 正在执行的任务是否一样，通俗的意思就是执行和等待的是否在同一队列
	    return ((lock_value ^ tid) & DLOCK_OWNER_MASK) == 0;
	}

• 综上所述，同步函数的底层实现实际是同步栅栏函数；同步函数中如果当前正在执行的队列和等待的是同一个队列，形成相互等待的局面，则会造成死锁。
• 同步函数的底层实现流程如下：

三、单例

① 单例的使用

• 在日常开发中，我们一般使用 GCD 的 dispatch_once 来创建单例，如下所示：

	static dispatch_once_t onceToken;
	dispatch_once(&onceToken, ^{
	    NSLog(@"这是一个单例！！");
	});

• 那么，单例的流程只执行一次，底层是如何控制的，为什么只能执行一次？单例的 block 是在什么时候进行调用的？

② 源码分析

• 进入 dispatch_once 源码，可以看到一个 dispatch_once_f 方法实现：

	#ifdef __BLOCKS__
	void
	// 参数1：onceToken，它是一个静态变量，由于不同位置定义的静态变量是不同的，所以静态变量具有唯一性
	// 参数2：block回调
	dispatch_once(dispatch_once_t *val, dispatch_block_t block)
	{
		dispatch_once_f(val, block, _dispatch_Block_invoke(block));
	}
	#endif

• 继续进入 dispatch_once_f 源码，其中的 val 是外界传入的 onceToken 静态变量，而 func 是 _dispatch_Block_invoke(block)，其底层实现如下所示：

	DISPATCH_NOINLINE
	void
	dispatch_once_f(dispatch_once_t *val, void *ctxt, dispatch_function_t func)
	{
	    dispatch_once_gate_t l = (dispatch_once_gate_t)val;
	
	#if !DISPATCH_ONCE_INLINE_FASTPATH || DISPATCH_ONCE_USE_QUIESCENT_COUNTER
	    uintptr_t v = os_atomic_load(&l->dgo_once, acquire);//load
	    if (likely(v == DLOCK_ONCE_DONE)) {	   // 已经执行过了，直接返回
	        return;
	    }
	#if DISPATCH_ONCE_USE_QUIESCENT_COUNTER
	    if (likely(DISPATCH_ONCE_IS_GEN(v))) {
	        return _dispatch_once_mark_done_if_quiesced(l, v);
	    }
	#endif
	#endif
	    if (_dispatch_once_gate_tryenter(l)) { // 尝试进入
	        return _dispatch_once_callout(l, ctxt, func);
	    }
	    return _dispatch_once_wait(l);		   // 无限次等待
	}

• dispatch_once_f 实现源码，分析如下：
  • 将val，也即为静态变量转换为 dispatch_once_gate_t 类型的变量 l；
  • 通过 os_atomic_load 获取此时的任务的标识符 v；
    • 如果 v 等于 DLOCK_ONCE_DONE，表示任务已经被执行，直接 return；
    • 如果任务执行，加锁失败，则执行 _dispatch_once_mark_done_if_quiesced 函数，再次进行存储，将标识符置为 DLOCK_ONCE_DONE；
    • 否则，则通过 _dispatch_once_gate_tryenter 尝试进入任务，即解锁，然后执行 _dispatch_once_callout 执行 block 回调；
  • 如果此时有任务正在执行，再次进来一个任务2，则通过 _dispatch_once_wait 函数让任务2进入无限次等待。
• 进入 _dispatch_once_gate_tryenter 解锁方法实现，可以看到它是通过 os_atomic_cmpxchg 方法进行对比，如果比较没有问题，则进行加锁，即任务的标识符置为 DLOCK_ONCE_UNLOCKED：

	DISPATCH_ALWAYS_INLINE
	static inline bool
	_dispatch_once_gate_tryenter(dispatch_once_gate_t l)
	{
		// 首先对比，然后进行改变
	    return os_atomic_cmpxchg(&l->dgo_once, DLOCK_ONCE_UNLOCKED,
	            (uintptr_t)_dispatch_lock_value_for_self(), relaxed); 
	}

• 进入 _dispatch_once_callout 回调，它主要通过 _dispatch_client_callout：block 回调执行，以及 _dispatch_once_gate_broadcast：进行广播：

	DISPATCH_NOINLINE
	static void
	_dispatch_once_callout(dispatch_once_gate_t l, void *ctxt,
			dispatch_function_t func)
	{
		// block调用执行
		_dispatch_client_callout(ctxt, func);
		// 进行广播：告诉别人已有归属，不要再找寻
		_dispatch_once_gate_broadcast(l);
	}

• 进入 _dispatch_client_callout 源码，主要就是执行 block 回调，其中的 f 等于 _dispatch_Block_invoke(block)，即异步回调：

	#undef _dispatch_client_callout
	void
	_dispatch_client_callout(void *ctxt, dispatch_function_t f)
	{
	    @try {
	        return f(ctxt);
	    }
	    @catch (...) {
	        objc_terminate();
	    }
	}

• 进入 _dispatch_once_gate_broadcast -> _dispatch_once_mark_done 源码，它主要就是给 dgo->dgo_once 赋一个值，然后将任务的标识符为DLOCK_ONCE_DONE，即解锁；

	DISPATCH_ALWAYS_INLINE
	static inline uintptr_t
	_dispatch_once_mark_done(dispatch_once_gate_t dgo)
	{
	    // 如果不相同，直接改为相同，然后上锁 -- DLOCK_ONCE_DONE
	    return os_atomic_xchg(&dgo->dgo_once, DLOCK_ONCE_DONE, release);
	}

• 看完单例的底层源码实现，我们就可以解释上面的问题：
• • 单例只执行一次的原理：GCD 单例中，有两个重要参数，onceToken 和 block，其中 onceToken 是静态变量，具有唯一性，在底层被封装成了 dispatch_once_gate_t 类型的变量 l，l 主要是用来获取底层原子封装性的关联，即变量 v，通过 v 来查询任务的状态，如果此时 v 等于 DLOCK_ONCE_DONE，说明任务已经处理过一次了，则不会再继续执行，直接 return；
• • block 调用时机：如果此时任务没有执行过，则会在底层通过 C++ 函数的比较，将任务进行加锁，即任务状态置为 DLOCK_ONCE_UNLOCK（目的是为了保证当前任务执行的唯一性，防止在其它地方有多次定义），加锁之后进行block回调函数的执行，执行完成后，将当前任务解锁，将当前的任务状态置为 DLOCK_ONCE_DONE，在下次进来时，就不会在执行，会直接返回；
• • 多线程影响：如果在当前任务执行期间，有其它任务进来，会进入无限次等待，原因是当前任务已经获取了锁，进行了加锁，那么其它任务是无法获取锁的。
• 单例的底层分析流程如下：

四、栅栏函数

① 栅栏函数说明

• GCD 中常用的栅栏函数，主要有两种：同步栅栏函数dispatch_barrier_sync（在主线程中执行）：前面的任务执行完毕才会执行栅栏函数之后的任务，但是同步栅栏函数会堵塞线程，影响后面的任务执行；异步栅栏函数 dispatch_barrier_async：前面的任务执行完毕才会继续执行栅栏函数的之后的任务。
• 栅栏函数最直接的作用就是控制任务执行顺序，使同步执行。
• 栅栏函数只能控制同一并发队列；
• 同步栅栏添加进入队列的时候，当前线程会被锁死，直到同步栅栏之前的任务和同步栅栏任务本身执行完毕时，当前线程才会打开然后继续执行下一句代码。
• 在使用栅栏函数时，使用自定义队列才有意义，如果用的是串行队列或者系统提供的全局并发队列，这个栅栏函数的作用等同于一个同步函数的作用，没有任何意义。

② 栅栏函数使用

• 现在总共有4个任务，其中前2个任务有依赖关系，即任务1执行完，执行任务2，此时可以使用栅栏函数；
• 异步栅栏函数：
• • 使用异步栅栏函数如下所示：

	dispatch_queue_t concurrentQueue = 	dispatch_queue_create("com.YDW.queue", DISPATCH_QUEUE_CONCURRENT);
    // 异步函数
    dispatch_async(concurrentQueue, ^{
        NSLog(@"任务1");
    });
    // 栅栏函数
    dispatch_barrier_async(concurrentQueue, ^{
        NSLog(@"任务2");
    });
    // 异步函数
    dispatch_async(concurrentQueue, ^{
        NSLog(@"任务3");
    });
    NSLog(@"任务4");

• • 执行结果如下所示：

	2021-03-31 19:52:08.355734+0800 GCD[98653:3157789] 任务4
	2021-03-31 19:52:08.355882+0800 GCD[98653:3157142] 任务1
	2021-03-31 19:52:08.356001+0800 GCD[98653:3157142] 任务2
	2021-03-31 19:52:08.356011+0800 GCD[98653:3157789] 任务3

• • 由此，异步栅栏函数不会阻塞主线程 ，异步堵塞的是队列；
• 同步栅栏函数：
• • 使用同步栅栏函数如下所示：

	dispatch_queue_t concurrentQueue = 	dispatch_queue_create("com.YDW.queue", DISPATCH_QUEUE_CONCURRENT);
    // 异步函数
    dispatch_async(concurrentQueue, ^{
        NSLog(@"任务1");
    });
    // 栅栏函数
    dispatch_barrier_sync(concurrentQueue, ^{
        NSLog(@"任务2");
    });
    // 异步函数
    dispatch_async(concurrentQueue, ^{
        NSLog(@"任务3");
    });
    NSLog(@"任务4");

• • 执行结果如下：

	2021-03-31 19:52:08.355734+0800 GCD[98653:3157789] 任务1
	2021-03-31 19:52:08.355882+0800 GCD[98653:3157142] 任务2
	2021-03-31 19:52:08.356001+0800 GCD[98653:3157142] 任务4
	2021-03-31 19:52:08.356011+0800 GCD[98653:3157789] 任务3

• • 因此，同步栅栏函数会堵塞主线程，同步堵塞是当前的线程。
• 栅栏函数除了用于任务有依赖关系时，同时还可以用于数据安全。
• • 现有如下的代码，执行会出现什么问题？

	dispatch_queue_t concurrentQueue = dispatch_queue_create("com.YDW.queue", DISPATCH_QUEUE_CONCURRENT);
    NSMutableArray *array = [NSMutableArray array];
    for (int i = 0; i < 10000; i++) {
        dispatch_async(concurrentQueue, ^{
            [array addObject:[NSString stringWithFormat:@"%d",i]];
        });
    }

• 可以看到，程序出现了崩溃，崩溃原因如下：

	malloc: *** error for object 0x7fe0ae809400: pointer being freed was not allocated
	malloc: *** set a breakpoint in malloc_error_break to debug

• 这是因为数据在不断的 retain 和 release，在数据还没有 retain 完毕时，然后又已经开始了 release，相当于加了一个空数据，进行 release，从而闪退。那么怎么解决呢？
• • 使用栅栏函数就可以解决问题，如下所示：

	dispatch_queue_t concurrentQueue = dispatch_queue_create("com.YDW.queue", DISPATCH_QUEUE_CONCURRENT);
    NSMutableArray *array = [NSMutableArray array];
    for (int i = 0; i < 10000; i++) {
        dispatch_async(concurrentQueue, ^{
            dispatch_barrier_async(concurrentQueue, ^{
                [array addObject:[NSString stringWithFormat:@"%d",i]];
            });
        });
    }

• • 使用互斥锁 synchronized 也可以解决，如下：

	dispatch_queue_t concurrentQueue = dispatch_queue_create("com.YDW.queue", DISPATCH_QUEUE_CONCURRENT);
    NSMutableArray *array = [NSMutableArray array];
    for (int i = 0; i < 10000; i++) {
        dispatch_async(concurrentQueue, ^{
            @synchronized (self) {
                [array addObject:[NSString stringWithFormat:@"%d", i]];
            };
        });
    }

• • 这里需要注意：
    • 如果栅栏函数中使用全局队列， 运行会崩溃，原因是系统也在用全局并发队列，使用栅栏同时会拦截系统的，所以会出现崩溃；
    • 如果将自定义并发队列改为串行队列，即 serial ，串行队列本身就是有序同步，此时加栅栏会浪费性能；
    • 栅栏函数只会阻塞一次。

③ 底层分析

• 异步栅栏函数：进入 dispatch_barrier_async 源码，其底层的实现与 dispatch_async 类似：

	#ifdef __BLOCKS__
	void
	dispatch_barrier_async(dispatch_queue_t dq, dispatch_block_t work)
	{
	    dispatch_continuation_t dc = _dispatch_continuation_alloc();
	    uintptr_t dc_flags = DC_FLAG_CONSUME | DC_FLAG_BARRIER;
	    dispatch_qos_t qos;
	
	    qos = _dispatch_continuation_init(dc, dq, work, 0, dc_flags);
	    _dispatch_continuation_async(dq, dc, qos, dc_flags);
	}
	#endif

• 同步栅栏函数：
• • 进入 dispatch_barrier_sync 源码，如下：

	void
	dispatch_barrier_sync(dispatch_queue_t dq, dispatch_block_t work)
	{
	    uintptr_t dc_flags = DC_FLAG_BARRIER | DC_FLAG_BLOCK;
	    if (unlikely(_dispatch_block_has_private_data(work))) {
	        return _dispatch_sync_block_with_privdata(dq, work, dc_flags);
	    }
	    _dispatch_barrier_sync_f(dq, work, _dispatch_Block_invoke(work), dc_flags);
	}

• • 进入 _dispatch_barrier_sync_f -> _dispatch_barrier_sync_f_inline 源码：

	DISPATCH_ALWAYS_INLINE
	static inline void
	_dispatch_barrier_sync_f_inline(dispatch_queue_t dq, void *ctxt,
	        dispatch_function_t func, uintptr_t dc_flags)
	{
	    dispatch_tid tid = _dispatch_tid_self();// 获取线程的id，即线程的唯一标识
	    
	    ...
	    
	    // 判断线程状态，需不需要等待，是否回收
	    if (unlikely(!_dispatch_queue_try_acquire_barrier_sync(dl, tid))) { // 栅栏函数死锁
	        return _dispatch_sync_f_slow(dl, ctxt, func, DC_FLAG_BARRIER, dl, // 没有回收
	                DC_FLAG_BARRIER | dc_flags);
	    }
	    // 验证target是否存在，如果存在，加入栅栏函数的递归查找 是否等待
	    if (unlikely(dl->do_targetq->do_targetq)) {
	        return _dispatch_sync_recurse(dl, ctxt, func,
	                DC_FLAG_BARRIER | dc_flags);
	    }
	    _dispatch_introspection_sync_begin(dl);
	    _dispatch_lane_barrier_sync_invoke_and_complete(dl, ctxt, func
	            DISPATCH_TRACE_ARG(_dispatch_trace_item_sync_push_pop(
	                    dq, ctxt, func, dc_flags | DC_FLAG_BARRIER))); // 执行
	}

• • 其中，_dispatch_queue_try_acquire_barrier_sync 实现如下：

	DISPATCH_ALWAYS_INLINE DISPATCH_WARN_RESULT
	static inline bool
	_dispatch_queue_try_acquire_barrier_sync(dispatch_queue_class_t dq, uint32_t tid)
	{
		return _dispatch_queue_try_acquire_barrier_sync_and_suspend(dq._dl, tid, 0);
	}

• 进入 _dispatch_queue_try_acquire_barrier_sync_and_suspend：

	DISPATCH_ALWAYS_INLINE DISPATCH_WARN_RESULT
	static inline bool
	_dispatch_queue_try_acquire_barrier_sync_and_suspend(dispatch_lane_t dq,
			uint32_t tid, uint64_t suspend_count)
	{
		uint64_t init  = DISPATCH_QUEUE_STATE_INIT_VALUE(dq->dq_width);
		uint64_t value = DISPATCH_QUEUE_WIDTH_FULL_BIT | DISPATCH_QUEUE_IN_BARRIER |
				_dispatch_lock_value_from_tid(tid) |
				(suspend_count * DISPATCH_QUEUE_SUSPEND_INTERVAL);
		uint64_t old_state, new_state;
	
		return os_atomic_rmw_loop2o(dq, dq_state, old_state, new_state, acquire, {
			uint64_t role = old_state & DISPATCH_QUEUE_ROLE_MASK;
			if (old_state != (init | role)) {
				os_atomic_rmw_loop_give_up(break);
			}
			new_state = value | role;
		});
	}

• • 通过 _dispatch_introspection_sync_begin 对向前信息进行处理：

	DISPATCH_ALWAYS_INLINE
	static inline void
	_dispatch_introspection_sync_begin(dispatch_queue_class_t dq)
	{
		if (!_dispatch_introspection.debug_queue_inversions) return;
		_dispatch_introspection_order_record(dq._dq);
	}

• • 通过 _dispatch_lane_barrier_sync_invoke_and_complete 执行 block 并释放：

	DISPATCH_NOINLINE
	static void
	_dispatch_lane_barrier_sync_invoke_and_complete(dispatch_lane_t dq,
			void *ctxt, dispatch_function_t func DISPATCH_TRACE_ARG(void *dc))
	{
		_dispatch_sync_function_invoke_inline(dq, ctxt, func);
		_dispatch_trace_item_complete(dc);
		if (unlikely(dq->dq_items_tail || dq->dq_width > 1)) {
			return _dispatch_lane_barrier_complete(dq, 0, 0);
		}
	
		// Presence of any of these bits requires more work that only
		// _dispatch_*_barrier_complete() handles properly
		//
		// Note: testing for RECEIVED_OVERRIDE or RECEIVED_SYNC_WAIT without
		// checking the role is sloppy, but is a super fast check, and neither of
		// these bits should be set if the lock was never contended/discovered.
		const uint64_t fail_unlock_mask = DISPATCH_QUEUE_SUSPEND_BITS_MASK |
				DISPATCH_QUEUE_ENQUEUED | DISPATCH_QUEUE_DIRTY |
				DISPATCH_QUEUE_RECEIVED_OVERRIDE | DISPATCH_QUEUE_SYNC_TRANSFER |
				DISPATCH_QUEUE_RECEIVED_SYNC_WAIT;
		uint64_t old_state, new_state;
	
		// similar to _dispatch_queue_drain_try_unlock
		// 对下层状态的释放
		os_atomic_rmw_loop2o(dq, dq_state, old_state, new_state, release, {
			new_state  = old_state - DISPATCH_QUEUE_SERIAL_DRAIN_OWNED;
			new_state &= ~DISPATCH_QUEUE_DRAIN_UNLOCK_MASK;
			new_state &= ~DISPATCH_QUEUE_MAX_QOS_MASK;
			if (unlikely(old_state & fail_unlock_mask)) {
				os_atomic_rmw_loop_give_up({
					// 通知 barrier 执行完毕
					return _dispatch_lane_barrier_complete(dq, 0, 0);
				});
			}
		});
		if (_dq_state_is_base_wlh(old_state)) {
			_dispatch_event_loop_assert_not_owned((dispatch_wlh_t)dq);
		}
	}

• • 同步栅栏函数底层流程总结：
    • 通过 _dispatch_tid_self 获取线程ID；
    • 通过 _dispatch_queue_try_acquire_barrier_sync 判断线程状态；
    • 通过 _dispatch_queue_try_acquire_barrier_sync_and_suspend 进行释放；
    • 通过 _dispatch_sync_recurse 递归查找栅栏函数的 target；
    • 通过 _dispatch_introspection_sync_begin 对向前信息进行处理；
    • 通过 _dispatch_lane_barrier_sync_invoke_and_complete 执行 block 并释放。

五、信号量

① dispatch_semaphore_create 创建

• 信号量的作用一般是用来使任务同步执行，类似于互斥锁，用户可以根据需要控制GCD最大并发数，一般使用如下：

	// 信号量
	dispatch_semaphore_t sem = dispatch_semaphore_create(1);
	dispatch_semaphore_wait(sem, DISPATCH_TIME_FOREVER);
	dispatch_semaphore_signal(sem);

• 创建的底层实现，主要是初始化信号量，并设置GCD的最大并发数，其最大并发数必须大于0，如下所示：

	dispatch_semaphore_t
	dispatch_semaphore_create(long value)
	{
	    dispatch_semaphore_t dsema;
	
	    // If the internal value is negative, then the absolute of the value is
	    // equal to the number of waiting threads. Therefore it is bogus to
	    // initialize the semaphore with a negative value.
	    if (value < 0) {
	        return DISPATCH_BAD_INPUT;
	    }
	
	    dsema = _dispatch_object_alloc(DISPATCH_VTABLE(semaphore),
	            sizeof(struct dispatch_semaphore_s));
	    dsema->do_next = DISPATCH_OBJECT_LISTLESS;
	    dsema->do_targetq = _dispatch_get_default_queue(false);
	    dsema->dsema_value = value;
	    _dispatch_sema4_init(&dsema->dsema_sema, _DSEMA4_POLICY_FIFO);
	    dsema->dsema_orig = value;
	    return dsema;
	}

② dispatch_semaphore_wait 加锁

• 该函数的实现，其主要作用是对信号量 dsema 通过 os_atomic_dec2o 进行了–操作，其内部是执行的 C++ 的 atomic_fetch_sub_explicit 方法：
  • 如果value 大于等于0，表示操作无效，即执行成功；
  • 如果value 等于LONG_MIN，系统会抛出一个 crash；
  • 如果value 小于0，则进入长等待。

	long
	dispatch_semaphore_wait(dispatch_semaphore_t dsema, dispatch_time_t timeout)
	{
	    // dsema_value 进行 -- 操作
	    long value = os_atomic_dec2o(dsema, dsema_value, acquire);
	    if (likely(value >= 0)) { // 表示执行操作无效，即执行成功
	        return 0;
	    }
	    return _dispatch_semaphore_wait_slow(dsema, timeout); // 长等待
	}

• 其中，os_atomic_dec2o 的宏定义转换如下：

	os_atomic_inc2o(p, f, m) 
	os_atomic_sub2o(p, f, 1, m)
	_os_atomic_c11_op((p), (v), m, sub, -)
	_os_atomic_c11_op((p), (v), m, add, +)
	({ _os_atomic_basetypeof(p) _v = (v), _r = \
	        atomic_fetch_##o##_explicit(_os_atomic_c11_atomic(p), _v, \
	        memory_order_##m); (__typeof__(_r))(_r op _v); })

	// 传值代入
	os_atomic_dec2o(dsema, dsema_value, acquire);
	os_atomic_sub2o(dsema, dsema_value, 1, m)
	os_atomic_sub(dsema->dsema_value, 1, m)
	_os_atomic_c11_op(dsema->dsema_value, 1, m, sub, -)
	_r = atomic_fetch_sub_explicit(dsema->dsema_value, 1),
	等价于 dsema->dsema_value - 1

• 进入 _dispatch_semaphore_wait_slow 的源码实现，当 value 小于 0 时，根据等待事件 timeout 做出不同操作：

	DISPATCH_NOINLINE
	static long
	_dispatch_semaphore_wait_slow(dispatch_semaphore_t dsema,
			dispatch_time_t timeout)
	{
		long orig;
	
		_dispatch_sema4_create(&dsema->dsema_sema, _DSEMA4_POLICY_FIFO);
		switch (timeout) {
		default:
			if (!_dispatch_sema4_timedwait(&dsema->dsema_sema, timeout)) {
				break;
			}
			// Fall through and try to undo what the fast path did to
			// dsema->dsema_value
		case DISPATCH_TIME_NOW:
			orig = dsema->dsema_value;
			while (orig < 0) {
				if (os_atomic_cmpxchgvw2o(dsema, dsema_value, orig, orig + 1,
						&orig, relaxed)) {
					return _DSEMA4_TIMEOUT();
				}
			}
			// Another thread called semaphore_signal().
			// Fall through and drain the wakeup.
		case DISPATCH_TIME_FOREVER:
			_dispatch_sema4_wait(&dsema->dsema_sema);
			break;
		}
		return 0;
	}

③ dispatch_semaphore_signal 解锁

• dispatch_semaphore_signal 实现原理，核心是通过 os_atomic_inc2o 函数对 value 进行 ++ 操作，os_atomic_inc2o 内部是通过 C++ 的 atomic_fetch_add_explicit，如果value 大于 0，表示操作无效，即执行成功；如果value 等于0，则进入长等待：

	long
	dispatch_semaphore_signal(dispatch_semaphore_t dsema)
	{
	    // signal 对 value 是 ++
	    long value = os_atomic_inc2o(dsema, dsema_value, release);
	    if (likely(value > 0)) { // 返回0，表示当前的执行操作无效，相当于执行成功
	        return 0;
	    }
	    if (unlikely(value == LONG_MIN)) {
	        DISPATCH_CLIENT_CRASH(value,
	                "Unbalanced call to dispatch_semaphore_signal()");
	    }
	    return _dispatch_semaphore_signal_slow(dsema); // 进入长等待
	}

• os_atomic_dec2o 的宏定义转换如下：

	os_atomic_inc2o(p, f, m) 
	os_atomic_add2o(p, f, 1, m)
	os_atomic_add(&(p)->f, (v), m)
	_os_atomic_c11_op((p), (v), m, add, +)
	({ _os_atomic_basetypeof(p) _v = (v), _r = \
	        atomic_fetch_##o##_explicit(_os_atomic_c11_atomic(p), _v, \
	        memory_order_##m); (__typeof__(_r))(_r op _v); })

	// 传值如下：
	os_atomic_inc2o(dsema, dsema_value, release);
	os_atomic_add2o(dsema, dsema_value, 1, m) 
	os_atomic_add(&(dsema)->dsema_value, (1), m)
	_os_atomic_c11_op((dsema->dsema_value), (1), m, add, +)
	_r = atomic_fetch_add_explicit(dsema->dsema_value, 1), // 等价于 dsema->dsema_value + 1

④ dispatch_semaphore 底层实现流程

• dispatch_semaphore_create 主要初始化信号量；
• dispatch_semaphore_wait 是对信号量的 value 进行–，即加锁操作；
• dispatch_semaphore_signal 是对信号量的 value 进行++，即解锁操作。

六、调度组

① dispatch_group_create 创建组

• 进入 dispatch_group_create 源码：

	dispatch_group_t
	dispatch_group_create(void)
	{
	    return _dispatch_group_create_with_count(0);
	}

• 进入 _dispatch_group_create_with_count 源码，主要是对 group 对象属性赋值，并返回 group 对象，n 等于 0；

	DISPATCH_ALWAYS_INLINE
	static inline dispatch_group_t
	_dispatch_group_create_with_count(uint32_t n)
	{
	    // 创建group对象,类型为OS_dispatch_group
	    dispatch_group_t dg = _dispatch_object_alloc(DISPATCH_VTABLE(group),
	            sizeof(struct dispatch_group_s));
	    // group对象赋值
	    dg->do_next = DISPATCH_OBJECT_LISTLESS;
	    dg->do_targetq = _dispatch_get_default_queue(false);
	    if (n) {
	        os_atomic_store2o(dg, dg_bits,
	                (uint32_t)-n * DISPATCH_GROUP_VALUE_INTERVAL, relaxed);
	        os_atomic_store2o(dg, do_ref_cnt, 1, relaxed);
	    }
	    return dg;
	}

② dispatch_group_enter 进组

• 进入 dispatch_group_enter 源码，通过 os_atomic_sub_orig2o 对 dg->dg.bits 作 – 操作，对数值进行处理：

	void
	dispatch_group_enter(dispatch_group_t dg)
	{
	    // The value is decremented on a 32bits wide atomic so that the carry
	    // for the 0 -> -1 transition is not propagated to the upper 32bits.
	    uint32_t old_bits = os_atomic_sub_orig2o(dg, dg_bits, // 原子递减 0 -> -1
	            DISPATCH_GROUP_VALUE_INTERVAL, acquire);
	    uint32_t old_value = old_bits & DISPATCH_GROUP_VALUE_MASK;
	    if (unlikely(old_value == 0)) { // 如果old_value
	        _dispatch_retain(dg);
	    }
	    if (unlikely(old_value == DISPATCH_GROUP_VALUE_MAX)) { // 到达临界值，会报crash
	        DISPATCH_CLIENT_CRASH(old_bits,
	                "Too many nested calls to dispatch_group_enter()");
	    }
	}

• 其中当 old_value == DISPATCH_GROUP_VALUE_MAX，到达临界值，会报crash， DISPATCH_GROUP_VALUE_MAX 的定义如下：

	#define DISPATCH_GROUP_GEN_MASK         0xffffffff00000000ULL
	#define DISPATCH_GROUP_VALUE_MASK       0x00000000fffffffcULL
	#define DISPATCH_GROUP_VALUE_INTERVAL   0x0000000000000004ULL
	#define DISPATCH_GROUP_VALUE_1          DISPATCH_GROUP_VALUE_MASK
	#define DISPATCH_GROUP_VALUE_MAX        DISPATCH_GROUP_VALUE_INTERVAL
	#define DISPATCH_GROUP_HAS_NOTIFS       0x0000000000000002ULL
	#define DISPATCH_GROUP_HAS_WAITERS      0x0000000000000001ULL

③ dispatch_group_leave 出组

• 进入 dispatch_group_leave 源码：

	void
	dispatch_group_leave(dispatch_group_t dg)
	{
	    // The value is incremented on a 64bits wide atomic so that the carry for
	    // the -1 -> 0 transition increments the generation atomically.
	    uint64_t new_state, old_state = os_atomic_add_orig2o(dg, dg_state,// 原子递增 ++
	            DISPATCH_GROUP_VALUE_INTERVAL, release);
	    uint32_t old_value = (uint32_t)(old_state & DISPATCH_GROUP_VALUE_MASK);
	    // 根据状态，唤醒
	    if (unlikely(old_value == DISPATCH_GROUP_VALUE_1)) {
	        old_state += DISPATCH_GROUP_VALUE_INTERVAL;
	        do {
	            new_state = old_state;
	            if ((old_state & DISPATCH_GROUP_VALUE_MASK) == 0) {
	                new_state &= ~DISPATCH_GROUP_HAS_WAITERS;
	                new_state &= ~DISPATCH_GROUP_HAS_NOTIFS;
	            } else {
	                // If the group was entered again since the atomic_add above,
	                // we can't clear the waiters bit anymore as we don't know for
	                // which generation the waiters are for
	                new_state &= ~DISPATCH_GROUP_HAS_NOTIFS;
	            }
	            if (old_state == new_state) break;
	        } while (unlikely(!os_atomic_cmpxchgv2o(dg, dg_state,
	                old_state, new_state, &old_state, relaxed)));
	        return _dispatch_group_wake(dg, old_state, true);//唤醒
	    }
	    //-1 -> 0, 0+1 -> 1，即多次leave，会报crash，简单来说就是enter-leave不平衡
	    if (unlikely(old_value == 0)) {
	        DISPATCH_CLIENT_CRASH((uintptr_t)old_value,
	                "Unbalanced call to dispatch_group_leave()");
	    }
	}

• 分析说明：
• • 1 到 0，即++操作；
• • 根据状态，do-while 循环，唤醒执行 block 任务；
• • 如果0 + 1 = 1，enter-leave 不平衡，即 leave 多次调用，会 crash。
• 进入 _dispatch_group_wake 源码，do-while 循环进行异步命中，调用 _dispatch_continuation_async 执行：

	DISPATCH_NOINLINE
	static void
	_dispatch_group_wake(dispatch_group_t dg, uint64_t dg_state, bool needs_release)
	{
	    uint16_t refs = needs_release ? 1 : 0; 
	
	    if (dg_state & DISPATCH_GROUP_HAS_NOTIFS) {
	        dispatch_continuation_t dc, next_dc, tail;
	
	        // Snapshot before anything is notified/woken <rdar://problem/8554546>
	        dc = os_mpsc_capture_snapshot(os_mpsc(dg, dg_notify), &tail);
	        do {
	            dispatch_queue_t dsn_queue = (dispatch_queue_t)dc->dc_data;
	            next_dc = os_mpsc_pop_snapshot_head(dc, tail, do_next);
	            _dispatch_continuation_async(dsn_queue, dc,
	                    _dispatch_qos_from_pp(dc->dc_priority), dc->dc_flags); // block任务执行
	            _dispatch_release(dsn_queue);
	        } while ((dc = next_dc)); // do-while循环，进行异步任务的命中
	
	        refs++;
	    }
	
	    if (dg_state & DISPATCH_GROUP_HAS_WAITERS) {
	        _dispatch_wake_by_address(&dg->dg_gen); // 地址释放
	    }
	
	    if (refs) _dispatch_release_n(dg, refs);  // 引用释放
	}

• 进入 _dispatch_continuation_async 源码：

	DISPATCH_ALWAYS_INLINE
	static inline void
	_dispatch_continuation_async(dispatch_queue_class_t dqu,
	        dispatch_continuation_t dc, dispatch_qos_t qos, uintptr_t dc_flags)
	{
	#if DISPATCH_INTROSPECTION
	    if (!(dc_flags & DC_FLAG_NO_INTROSPECTION)) {
	        _dispatch_trace_item_push(dqu, dc); // 跟踪日志
	    }
	#else
	    (void)dc_flags;
	#endif
	    return dx_push(dqu._dq, dc, qos); // 与dx_invoke一样，都是宏
	}

④ dispatch_group_notify 通知

• 进入 dispatch_group_notify 源码，如果 old_state 等于 0，就可以进行释放：

	DISPATCH_ALWAYS_INLINE
	static inline void
	_dispatch_group_notify(dispatch_group_t dg, dispatch_queue_t dq,
	        dispatch_continuation_t dsn)
	{
	    uint64_t old_state, new_state;
	    dispatch_continuation_t prev;
	
	    dsn->dc_data = dq;
	    _dispatch_retain(dq);
	    // 获取dg底层的状态标识码，通过os_atomic_store2o获取的值，即从dg的状态码 转成了 os底层的state
	    prev = os_mpsc_push_update_tail(os_mpsc(dg, dg_notify), dsn, do_next);
	    if (os_mpsc_push_was_empty(prev)) _dispatch_retain(dg);
	    os_mpsc_push_update_prev(os_mpsc(dg, dg_notify), prev, dsn, do_next);
	    if (os_mpsc_push_was_empty(prev)) {
	        os_atomic_rmw_loop2o(dg, dg_state, old_state, new_state, release, {
	            new_state = old_state | DISPATCH_GROUP_HAS_NOTIFS;
	            if ((uint32_t)old_state == 0) { // 如果等于0，则可以进行释放了
	                os_atomic_rmw_loop_give_up({
	                    return _dispatch_group_wake(dg, new_state, false); // 唤醒
	                });
	            }
	        });
	    }
	}

• 其中 os_mpsc_push_update_tail 定义如下，用于获取 dg 的状态码：

	#define os_mpsc_push_update_tail(Q, tail, _o_next)  ({ \
	    os_mpsc_node_type(Q) _tl = (tail); \
	    os_atomic_store2o(_tl, _o_next, NULL, relaxed); \
	    os_atomic_xchg(_os_mpsc_tail Q, _tl, release); \
	})

⑤ dispatch_group_async

• 进入 dispatch_group_async 源码，主要是包装任务和异步处理任务：

	#ifdef __BLOCKS__
	void
	dispatch_group_async(dispatch_group_t dg, dispatch_queue_t dq,
	        dispatch_block_t db)
	{
	    
	    dispatch_continuation_t dc = _dispatch_continuation_alloc();
	    uintptr_t dc_flags = DC_FLAG_CONSUME | DC_FLAG_GROUP_ASYNC;
	    dispatch_qos_t qos;
	    // 任务包装器
	    qos = _dispatch_continuation_init(dc, dq, db, 0, dc_flags);
	    // 处理任务
	    _dispatch_continuation_group_async(dg, dq, dc, qos);
	}
	#endif

• 进入 _dispatch_continuation_group_async 源码，主要集成 dispatch_group_enter 进组操作：

	DISPATCH_ALWAYS_INLINE
	static inline void
	_dispatch_continuation_group_async(dispatch_group_t dg, dispatch_queue_t dq,
	        dispatch_continuation_t dc, dispatch_qos_t qos)
	{
	    dispatch_group_enter(dg); // 进组
	    dc->dc_data = dg;
	    _dispatch_continuation_async(dq, dc, qos, dc->dc_flags); // 异步操作
	}

• 进入 _dispatch_continuation_async 源码，执行常规的异步函数底层操作，既然有 enter，肯定有 leave，可以猜测 block 执行之后隐性的执行 leave。现有如下代码：

	dispatch_group_t group = dispatch_group_create();
    dispatch_queue_t queue = dispatch_get_global_queue(0, 0);
    dispatch_group_enter(group);
    dispatch_async(queue, ^{
        sleep(1);
        NSLog(@"任务1");
        dispatch_group_leave(group);
    });
    dispatch_group_enter(group);
    dispatch_async(queue, ^{
        NSLog(@"任务2");
        dispatch_group_leave(group);
    });
    dispatch_group_async(group, queue, ^{
        NSLog(@"任务5");
    });
    dispatch_group_enter(group);
    dispatch_async(queue, ^{
        NSLog(@"任务3");
        dispatch_group_leave(group);
    });
    dispatch_group_enter(group);
    dispatch_async(queue, ^{
        NSLog(@"任务4");
        dispatch_group_leave(group);
    });

• 打印堆栈信息：

	(lldb) bt
	* thread #6, queue = 'com.apple.root.default-qos', stop reason = breakpoint 1.1
	  * frame #0: 0x00000001053c7517 函数与队列`__29-[ViewController viewDidLoad]_block_invoke_2(.block_descriptor=0x00000001053ca128) at ViewController.m:35:9
	    frame #1: 0x00000001056387ec libdispatch.dylib`_dispatch_call_block_and_release + 12
	    frame #2: 0x00000001056399c8 libdispatch.dylib`_dispatch_client_callout + 8
	    frame #3: 0x000000010563bfe2 libdispatch.dylib`_dispatch_queue_override_invoke + 1444
	    frame #4: 0x000000010564b508 libdispatch.dylib`_dispatch_root_queue_drain + 351
	    frame #5: 0x000000010564be6d libdispatch.dylib`_dispatch_worker_thread2 + 135
	    frame #6: 0x00007fff611639f7 libsystem_pthread.dylib`_pthread_wqthread + 220
	    frame #7: 0x00007fff61162b77 libsystem_pthread.dylib`start_wqthread + 15
	(lldb) 

• 可以看到 libdispatch.dylib`_dispatch_client_callout + 8，我们查看 _dispatch_client_callout 的调用，可以看到在 _dispatch_continuation_with_group_invoke 中，因此可以验证 dispatch_group_async 底层是 enter-leave；

	DISPATCH_ALWAYS_INLINE
	static inline void
	_dispatch_continuation_with_group_invoke(dispatch_continuation_t dc)
	{
	    struct dispatch_object_s *dou = dc->dc_data;
	    unsigned long type = dx_type(dou);
	    if (type == DISPATCH_GROUP_TYPE) { // 如果是调度组类型
	        _dispatch_client_callout(dc->dc_ctxt, dc->dc_func); // block回调
	        _dispatch_trace_item_complete(dc);
	        dispatch_group_leave((dispatch_group_t)dou); // 出组
	    } else {
	        DISPATCH_INTERNAL_CRASH(dx_type(dou), "Unexpected object type");
	    }

⑥ dispatch_group 底层实现流程

• enter-leave 需要成对出现；
• dispatch_group_enter 在底层是通过 C++ 函数，对 group 的 value 进行 – 操作（即0 -> -1）；
• dispatch_group_leave 在底层是通过 C++ 函数，对 group 的 value 进行 ++ 操作（即-1 -> 0）；
• dispatch_group_notify 在底层主要是判断 group 的 state 是否等于0，当等于0时发出通知；
• block 任务的唤醒，可以通过 dispatch_group_leave，也可以通过 dispatch_group_notify；
• dispatch_group_async 等同于 enter - leave，其底层的实现就是 enter-leave。

七、源码下载

本文是在源码 libdispatch.dylib 中探索分析的，源码的下载地址：macOS 10.15 - Source。