linux内核情景分析之内核中的互斥操作

信号量机制:

struct sempahore是其结构,定义如下

struct semaphore {
	atomic_t count;//资源数目
	int sleepers;//等待进程数目
	wait_queue_head_t wait;//等待队列
#if WAITQUEUE_DEBUG
	long __magic;
#endif
};

down操作成功(减后结果非负数)那就在标号1处结束down操作,转到临界区.

如果减为负数,跳转到2标号,并且调用call_down_failed,进入睡眠,一直要到唤醒并拿到资源才返回跳转到1标号,结束down操作进入临界区

/*
 * This is ugly, but we want the default case to fall through.
 * "__down_failed" is a special asm handler that calls the C
 * routine that actually waits. See arch/i386/kernel/semaphore.c
 */
static inline void down(struct semaphore * sem)
{
#if WAITQUEUE_DEBUG
	CHECK_MAGIC(sem->__magic);
#endif
	__asm__ __volatile__(
		"# atomic down operation
	"
		LOCK "decl %0
	"     /* --sem->count lock字段把总线锁住,防止其他cpu干扰*/ 
		"js 2f
"  /*如果小于0,那就跳转到2号*/  
		"1:
" /*成功拿到,从此处进入临界区*/
		".section .text.lock,"ax"
"
		"2:	call __down_failed
	" /*count--后为负值,休眠*/
		"jmp 1b
"/*失败睡眠,但经过一段时间被唤醒,并且进入临界区,就跳转到1*/
		".previous"
		:"=m" (sem->count)
		:"c" (sem)
		:"memory");
}

__down_failed源码,这里的目的只是为了调用__down函数

asm(
".align 4
"
".globl __down_failed
"
"__down_failed:
	"
	"pushl %eax
	"
	"pushl %edx
	"
	"pushl %ecx
	"
	"call __down
	"
	"popl %ecx
	"
	"popl %edx
	"
	"popl %eax
	"
	"ret"
);

__down将判断资源是否存在,不存在睡眠,如果被唤醒那就从等待队列删除,并且唤醒其他等待队列进程

void __down(struct semaphore * sem)
{
	struct task_struct *tsk = current;
	DECLARE_WAITQUEUE(wait, tsk);//wait代表tsk
	tsk->state = TASK_UNINTERRUPTIBLE;//设置为睡眠状态
	add_wait_queue_exclusive(&sem->wait, &wait);//把当前进程的等待队列元素wait加入到sem->wait等待队列队尾
	spin_lock_irq(&semaphore_lock);
	sem->sleepers++;//等待进程数目+1
	for (;;) {
		int sleepers = sem->sleepers;//禁止本地中断并获取指定的锁
		/*返回非0,表示进程需要等待
		 * 假设有2个进程,进程资源已经被占用,当前进程执行down失败,跳转到这里,等待调度
		 结郭前一个进程归还了资源,count变为0(之前down2次为-1),sleeper-1也为0,相加等于0,于是可以进入临界区
		 */
		if (!atomic_add_negative(sleepers - 1, &sem->count)) {//返回0表示可以进入临界区
			sem->sleepers = 0;//睡眠的进程为0,因为要唤醒这进程了
			break;
		}
		sem->sleepers = 1;	/* 没法到临界区,那就需要阻塞,执行到这,设置为1,那就只有us - see -1 above */
		spin_unlock_irq(&semaphore_lock);
		schedule();//调度
		tsk->state = TASK_UNINTERRUPTIBLE;//将当前进程设置为睡眠状态
		spin_lock_irq(&semaphore_lock);
	}
	spin_unlock_irq(&semaphore_lock);//解锁
	remove_wait_queue(&sem->wait, &wait);//当前进程可以进入临界区后,从wait队列移除
	tsk->state = TASK_RUNNING;//设置为可执行状态
	wake_up(&sem->wait);//唤醒等待队列(然而等待队列很多进程依旧无法进入临界区)
}

缺陷:优先级倒转,优先级高进程在外等待,在临界区的进程优先级很低,一旦优先级低的进程在临界区受阻睡眠,也得不到及时调度,优先级高进程会被拖累,解决办法:把高优先级借给进入临界区的进程

接下来分析up函数

/*
 * Note! This is subtle. We jump to wake people up only if
 * the semaphore was negative (== somebody was waiting on it).
 * The default case (no contention) will result in NO
 * jumps for both down() and up().
 */
static inline void up(struct semaphore * sem)
{
#if WAITQUEUE_DEBUG
	CHECK_MAGIC(sem->__magic);
#endif
	__asm__ __volatile__(
		"# atomic up operation
	"
		LOCK "incl %0
	"     /* ++sem->count */
		"jle 2f
"//如果资源充足(也就是递增结果为正数,直接从1:跳出临界区,不用唤醒阻塞进程(应该说没有阻塞临界区的进程)
		"1:
"
		".section .text.lock,"ax"
"
		"2:	call __up_wakeup
	" /*递增结果为负数或者非0值,就唤醒阻塞进程*/
		"jmp 1b
"
		".previous"
		:"=m" (sem->count)
		:"c" (sem)
		:"memory");
}

__up_wakup的目的也是调用__up

asm(
".align 4
"
".globl __up_wakeup
"
"__up_wakeup:
	"
	"pushl %eax
	"
	"pushl %edx
	"
	"pushl %ecx
	"
	"call __up
	"
	"popl %ecx
	"
	"popl %edx
	"
	"popl %eax
	"
	"ret"
);

/*
 * Semaphores are implemented using a two-way counter:
 * The "count" variable is decremented for each process
 * that tries to acquire the semaphore, while the "sleeping"
 * variable is a count of such acquires.
 *
 * Notably, the inline "up()" and "down()" functions can
 * efficiently test if they need to do any extra work (up
 * needs to do something only if count was negative before
 * the increment operation.
 *
 * "sleeping" and the contention routine ordering is
 * protected by the semaphore spinlock.
 *
 * Note that these functions are only called when there is
 * contention on the lock, and as such all this is the
 * "non-critical" part of the whole semaphore business. The
 * critical part is the inline stuff in <asm/semaphore.h>
 * where we want to avoid any extra jumps and calls.
 */
/*
 * Logic:
 *  - only on a boundary condition do we need to care. When we go
 *    from a negative count to a non-negative, we wake people up.
 *  - when we go from a non-negative count to a negative do we
 *    (a) synchronize with the "sleeper" count and (b) make sure
 *    that we're on the wakeup list before we synchronize so that
 *    we cannot lose wakeup events.
 */
void __up(struct semaphore *sem)
{
	wake_up(&sem->wait);//唤醒等待队列中的进程
}

#define wake_up(x)			__wake_up((x),TASK_UNINTERRUPTIBLE | TASK_INTERRUPTIBLE,WQ_FLAG_EXCLUSIVE)

void __wake_up(wait_queue_head_t *q, unsigned int mode, unsigned int wq_mode)
{	
    __wake_up_common(q, mode, wq_mode, 0);
}

posix信号量与内核信号量概念基本一样,不过posix信号量可以用于位于内核外临界区的不同进程.而内核信号量只可以用于临界区位于内核

互斥锁(也就是mutex.一般用于线程互斥),不过可以通过设置线程锁属性用于不同进程通信,为了达到多进程共享的需要，互斥锁对象需要创建在共享内存中

文件锁用于2个进程访问一个文件

自旋锁,读写锁只可用于线程互斥

信号量分为匿名信号量与有名信号量,前一个用于线程互斥,后一个用于进程同步

大内核锁也是用来保护临界区资源，避免出现多个处理器上的进程同时访问同一区域的

来自为知笔记(Wiz)