Scheduler 学习之二：主调度器_schedule函数

1. Summary
   1.1 _schedule函数是又被称作为主调度器，负责选择新的task run到当前的cpu当中
   1.2 _scheduler将选核的功能交给各调度类完成
   1.3 pick_next_task提供了一个优化功能 ：假定fair class中的Task数量与rq 中一致，刚直接从fair class中选即可
   1.4 主调度器起作用的时间主要是在task切换的时候，例如，抢锁抢不到，或者初始开启抢占，以及从kernel space回到user space
2. Flow
    

   static void __sched notrace __schedule(bool preempt)
   {
   	struct task_struct *prev, *next;
   	unsigned long *switch_count;
   	struct rq_flags rf;
   	struct rq *rq;
   	int cpu;
   
   	cpu = smp_processor_id();
   	rq = cpu_rq(cpu);
   	prev = rq->curr;//现在的将变成过往
   
   	schedule_debug(prev, preempt);//此函数检查当前调用schedule是否合适,如不能在原子上下文(即抢占被关闭的情况下)中调用schedule,栈崩溃等.
   
   	if (sched_feat(HRTICK))
   		hrtick_clear(rq);
   
   	local_irq_disable();
   	rcu_note_context_switch(preempt);//通知RCU的相关操作
   
   	/*
   	 * Make sure that signal_pending_state()->signal_pending() below
   	 * can't be reordered with __set_current_state(TASK_INTERRUPTIBLE)
   	 * done by the caller to avoid the race with signal_wake_up().
   	 *
   	 * The membarrier system call requires a full memory barrier
   	 * after coming from user-space, before storing to rq->curr.
   	 */
   	rq_lock(rq, &rf);
   	smp_mb__after_spinlock();
   
   	/* Promote REQ to ACT */
   	rq->clock_update_flags <<= 1;//为什么要修改clock_update_flags呢?
   	update_rq_clock(rq);
   
   	switch_count = &prev->nivcsw; //根据上下文,switch_count此时指向的是非自愿切换计数器
   	if (!preempt && prev->state) {//如果不是被抢占的,且task处于非running状态
   		if (signal_pending_state(prev->state, prev)) {//如果有signal pending,则重新将prev task置为running状态
   			prev->state = TASK_RUNNING;
   		} else { //否则做dequeue的操作.
   			deactivate_task(rq, prev, DEQUEUE_