sched.c

Linux内核源代码 为压缩文件 是<<Linux内核>>一书中的源代码

/*
 * linux/net/sunrpc/sched.c
 *
 * Scheduling for synchronous and asynchronous RPC requests.
 *
 * Copyright (C) 1996 Olaf Kirch, <[e-mail]>
 * 
 * TCP NFS related read + write fixes
 * (C) 1999 Dave Airlie, University of Limerick, Ireland <[e-mail]>
 */

#include <linux/module.h>

#define __KERNEL_SYSCALLS__
#include <linux/sched.h>
#include <linux/interrupt.h>
#include <linux/malloc.h>
#include <linux/unistd.h>
#include <linux/smp.h>
#include <linux/smp_lock.h>
#include <linux/spinlock.h>

#include <linux/sunrpc/clnt.h>

#ifdef RPC_DEBUG
#define RPCDBG_FACILITY		RPCDBG_SCHED
static int			rpc_task_id;
#endif

/*
 * We give RPC the same get_free_pages priority as NFS
 */
#define GFP_RPC			GFP_NFS

static void			__rpc_default_timer(struct rpc_task *task);
static void			rpciod_killall(void);

/*
 * When an asynchronous RPC task is activated within a bottom half
 * handler, or while executing another RPC task, it is put on
 * schedq, and rpciod is woken up.
 */
static struct rpc_wait_queue	schedq = RPC_INIT_WAITQ("schedq");

/*
 * RPC tasks that create another task (e.g. for contacting the portmapper)
 * will wait on this queue for their child's completion
 */
static struct rpc_wait_queue	childq = RPC_INIT_WAITQ("childq");

/*
 * RPC tasks sit here while waiting for conditions to improve.
 */
static struct rpc_wait_queue	delay_queue = RPC_INIT_WAITQ("delayq");

/*
 * All RPC tasks are linked into this list
 */
static struct rpc_task *	all_tasks;

/*
 * rpciod-related stuff
 */
static DECLARE_WAIT_QUEUE_HEAD(rpciod_idle);
static DECLARE_WAIT_QUEUE_HEAD(rpciod_killer);
static DECLARE_MUTEX(rpciod_sema);
static unsigned int		rpciod_users;
static pid_t			rpciod_pid;
static int			rpc_inhibit;

/*
 * Spinlock for wait queues. Access to the latter also has to be
 * interrupt-safe in order to allow timers to wake up sleeping tasks.
 */
spinlock_t rpc_queue_lock = SPIN_LOCK_UNLOCKED;
/*
 * Spinlock for other critical sections of code.
 */
spinlock_t rpc_sched_lock = SPIN_LOCK_UNLOCKED;

/*
 * This is the last-ditch buffer for NFS swap requests
 */
static u32			swap_buffer[PAGE_SIZE >> 2];
static long			swap_buffer_used;

/*
 * Make allocation of the swap_buffer SMP-safe
 */
static __inline__ int rpc_lock_swapbuf(void)
{
	return !test_and_set_bit(1, &swap_buffer_used);
}
static __inline__ void rpc_unlock_swapbuf(void)
{
	clear_bit(1, &swap_buffer_used);
}

/*
 * Disable the timer for a given RPC task. Should be called with
 * rpc_queue_lock and bh_disabled in order to avoid races within
 * rpc_run_timer().
 */
static inline void
__rpc_disable_timer(struct rpc_task *task)
{
	dprintk("RPC: %4d disabling timer\n", task->tk_pid);
	task->tk_timeout_fn = NULL;
	task->tk_timeout = 0;
}

/*
 * Run a timeout function.
 * We use the callback in order to allow __rpc_wake_up_task()
 * and friends to disable the timer synchronously on SMP systems
 * without calling del_timer_sync(). The latter could cause a
 * deadlock if called while we're holding spinlocks...
 */
static void
rpc_run_timer(struct rpc_task *task)
{
	void (*callback)(struct rpc_task *);

	spin_lock_bh(&rpc_queue_lock);
	callback = task->tk_timeout_fn;
	task->tk_timeout_fn = NULL;
	spin_unlock_bh(&rpc_queue_lock);
	if (callback) {
		dprintk("RPC: %4d running timer\n", task->tk_pid);
		callback(task);
	}
}

/*
 * Set up a timer for the current task.
 */
static inline void
__rpc_add_timer(struct rpc_task *task, rpc_action timer)
{
	if (!task->tk_timeout)
		return;

	dprintk("RPC: %4d setting alarm for %lu ms\n",
			task->tk_pid, task->tk_timeout * 1000 / HZ);

	if (timer)
		task->tk_timeout_fn = timer;
	else
		task->tk_timeout_fn = __rpc_default_timer;
	mod_timer(&task->tk_timer, jiffies + task->tk_timeout);
}

/*
 * Set up a timer for an already sleeping task.
 */
void rpc_add_timer(struct rpc_task *task, rpc_action timer)
{
	spin_lock_bh(&rpc_queue_lock);
	if (!(RPC_IS_RUNNING(task) || task->tk_wakeup))
		__rpc_add_timer(task, timer);
	spin_unlock_bh(&rpc_queue_lock);
}

/*
 * Delete any timer for the current task. Because we use del_timer_sync(),
 * this function should never be called while holding rpc_queue_lock.
 */
static inline void
rpc_delete_timer(struct rpc_task *task)
{
	if (timer_pending(&task->tk_timer)) {
		dprintk("RPC: %4d deleting timer\n", task->tk_pid);
		del_timer_sync(&task->tk_timer);
	}
}

/*
 * Add new request to wait queue.
 *
 * Swapper tasks always get inserted at the head of the queue.
 * This should avoid many nasty memory deadlocks and hopefully
 * improve overall performance.
 * Everyone else gets appended to the queue to ensure proper FIFO behavior.
 */
static inline int
__rpc_add_wait_queue(struct rpc_wait_queue *queue, struct rpc_task *task)
{
	if (task->tk_rpcwait == queue)
		return 0;

	if (task->tk_rpcwait) {
		printk(KERN_WARNING "RPC: doubly enqueued task!\n");
		return -EWOULDBLOCK;
	}
	if (RPC_IS_SWAPPER(task))
		rpc_insert_list(&queue->task, task);
	else
		rpc_append_list(&queue->task, task);
	task->tk_rpcwait = queue;

	dprintk("RPC: %4d added to queue %p \"%s\"\n",
				task->tk_pid, queue, rpc_qname(queue));

	return 0;
}

int
rpc_add_wait_queue(struct rpc_wait_queue *q, struct rpc_task *task)
{
	int		result;

	spin_lock_bh(&rpc_queue_lock);
	result = __rpc_add_wait_queue(q, task);
	spin_unlock_bh(&rpc_queue_lock);
	return result;
}

/*
 * Remove request from queue.
 * Note: must be called with spin lock held.
 */
static inline void
__rpc_remove_wait_queue(struct rpc_task *task)
{
	struct rpc_wait_queue *queue = task->tk_rpcwait;

	if (!queue)
		return;

	rpc_remove_list(&queue->task, task);
	task->tk_rpcwait = NULL;

	dprintk("RPC: %4d removed from queue %p \"%s\"\n",
				task->tk_pid, queue, rpc_qname(queue));
}

void
rpc_remove_wait_queue(struct rpc_task *task)
{
	if (!task->tk_rpcwait)
		return;
	spin_lock_bh(&rpc_queue_lock);
	__rpc_remove_wait_queue(task);
	spin_unlock_bh(&rpc_queue_lock);
}

/*
 * Make an RPC task runnable.
 *
 * Note: If the task is ASYNC, this must be called with 
 * the spinlock held to protect the wait queue operation.
 */
static inline void
rpc_make_runnable(struct rpc_task *task)
{
	if (task->tk_timeout_fn) {
		printk(KERN_ERR "RPC: task w/ running timer in rpc_make_runnable!!\n");
		return;
	}
	rpc_set_running(task);
	if (RPC_IS_ASYNC(task)) {
		if (RPC_IS_SLEEPING(task)) {
			int status;
			status = __rpc_add_wait_queue(&schedq, task);
			if (status < 0) {
				printk(KERN_WARNING "RPC: failed to add task to queue: error: %d!\n", status);
				task->tk_status = status;
				return;
			}
			rpc_clear_sleeping(task);
			if (waitqueue_active(&rpciod_idle))
				wake_up(&rpciod_idle);
		}
	} else {
		rpc_clear_sleeping(task);
		if (waitqueue_active(&task->tk_wait))
			wake_up(&task->tk_wait);
	}
}

/*
 * Place a newly initialized task on the schedq.
 */
static inline void
rpc_schedule_run(struct rpc_task *task)
{
	/* Don't run a child twice! */
	if (RPC_IS_ACTIVATED(task))
		return;
	task->tk_active = 1;
	rpc_set_sleeping(task);
	rpc_make_runnable(task);
}

/*
 *	For other people who may need to wake the I/O daemon
 *	but should (for now) know nothing about its innards
 */
void rpciod_wake_up(void)
{
	if(rpciod_pid==0)
		printk(KERN_ERR "rpciod: wot no daemon?\n");
	if (waitqueue_active(&rpciod_idle))
		wake_up(&rpciod_idle);
}

/*
 * Prepare for sleeping on a wait queue.
 * By always appending tasks to the list we ensure FIFO behavior.
 * NB: An RPC task will only receive interrupt-driven events as long
 * as it's on a wait queue.
 */
static void
__rpc_sleep_on(struct rpc_wait_queue *q, struct rpc_task *task,
			rpc_action action, rpc_action timer)
{
	int status;

	dprintk("RPC: %4d sleep_on(queue \"%s\" time %ld)\n", task->tk_pid,
				rpc_qname(q), jiffies);

	if (!RPC_IS_ASYNC(task) && !RPC_IS_ACTIVATED(task)) {
		printk(KERN_ERR "RPC: Inactive synchronous task put to sleep!\n");
		return;
	}

	/* Mark the task as being activated if so needed */
	if (!RPC_IS_ACTIVATED(task)) {
		task->tk_active = 1;
		rpc_set_sleeping(task);
	}

	status = __rpc_add_wait_queue(q, task);
	if (status) {
		printk(KERN_WARNING "RPC: failed to add task to queue: error: %d!\n", status);
		task->tk_status = status;
	} else {
		rpc_clear_running(task);
		if (task->tk_callback) {
			dprintk(KERN_ERR "RPC: %4d overwrites an active callback\n", task->tk_pid);
			BUG();
		}
		task->tk_callback = action;
		__rpc_add_timer(task, timer);
	}
}

void
rpc_sleep_on(struct rpc_wait_queue *q, struct rpc_task *task,
				rpc_action action, rpc_action timer)
{
	/*
	 * Protect the queue operations.
	 */
	spin_lock_bh(&rpc_queue_lock);
	__rpc_sleep_on(q, task, action, timer);
	spin_unlock_bh(&rpc_queue_lock);
}

void
rpc_sleep_locked(struct rpc_wait_queue *q, struct rpc_task *task,
		 rpc_action action, rpc_action timer)
{
	/*
	 * Protect the queue operations.
	 */
	spin_lock_bh(&rpc_queue_lock);
	__rpc_sleep_on(q, task, action, timer);
	__rpc_lock_task(task);
	spin_unlock_bh(&rpc_queue_lock);
}

/**
 * __rpc_wake_up_task - wake up a single rpc_task
 * @task: task to be woken up
 *
 * If the task is locked, it is merely removed from the queue, and
 * 'task->tk_wakeup' is set. rpc_unlock_task() will then ensure
 * that it is woken up as soon as the lock count goes to zero.
 *
 * Caller must hold rpc_queue_lock
 */
static void
__rpc_wake_up_task(struct rpc_task *task)
{
	dprintk("RPC: %4d __rpc_wake_up_task (now %ld inh %d)\n",
					task->tk_pid, jiffies, rpc_inhibit);

#ifdef RPC_DEBUG
	if (task->tk_magic != 0xf00baa) {
		printk(KERN_ERR "RPC: attempt to wake up non-existing task!\n");
		rpc_debug = ~0;
		rpc_show_tasks();
		return;
	}
#endif
	/* Has the task been executed yet? If not, we cannot wake it up! */
	if (!RPC_IS_ACTIVATED(task)) {
		printk(KERN_ERR "RPC: Inactive task (%p) being woken up!\n", task);
		return;
	}
	if (RPC_IS_RUNNING(task))
		return;

	__rpc_disable_timer(task);
	if (task->tk_rpcwait != &schedq)
		__rpc_remove_wait_queue(task);

	/* If the task has been locked, then set tk_wakeup so that
	 * rpc_unlock_task() wakes us up... */
	if (task->tk_lock) {
		task->tk_wakeup = 1;
		return;
	} else
		task->tk_wakeup = 0;

	rpc_make_runnable(task);

	dprintk("RPC:      __rpc_wake_up_task done\n");
}

/*
 * Default timeout handler if none specified by user
 */
static void
__rpc_default_timer(struct rpc_task *task)
{
	dprintk("RPC: %d timeout (default timer)\n", task->tk_pid);
	task->tk_status = -ETIMEDOUT;
	rpc_wake_up_task(task);
}

/*
 * Wake up the specified task
 */
void
rpc_wake_up_task(struct rpc_task *task)
{
	if (RPC_IS_RUNNING(task))
		return;
	spin_lock_bh(&rpc_queue_lock);
	__rpc_wake_up_task(task);
	spin_unlock_bh(&rpc_queue_lock);
}

/*
 * Wake up the next task on the wait queue.
 */
struct rpc_task *
rpc_wake_up_next(struct rpc_wait_queue *queue)
{
	struct rpc_task	*task;

	dprintk("RPC:      wake_up_next(%p \"%s\")\n", queue, rpc_qname(queue));
	spin_lock_bh(&rpc_queue_lock);
	if ((task = queue->task) != 0)
		__rpc_wake_up_task(task);
	spin_unlock_bh(&rpc_queue_lock);

	return task;
}

/**
 * rpc_wake_up - wake up all rpc_tasks
 * @queue: rpc_wait_queue on which the tasks are sleeping
 *
 * Grabs rpc_queue_lock
 */
void
rpc_wake_up(struct rpc_wait_queue *queue)
{
	spin_lock_bh(&rpc_queue_lock);
	while (queue->task)
		__rpc_wake_up_task(queue->task);
	spin_unlock_bh(&rpc_queue_lock);
}

/**
 * rpc_wake_up_status - wake up all rpc_tasks and set their status value.
 * @queue: rpc_wait_queue on which the tasks are sleeping
 * @status: status value to set
 *
 * Grabs rpc_queue_lock
 */
void
rpc_wake_up_status(struct rpc_wait_queue *queue, int status)
{
	struct rpc_task	*task;

	spin_lock_bh(&rpc_queue_lock);
	while ((task = queue->task) != NULL) {
		task->tk_status = status;
		__rpc_wake_up_task(task);
	}
	spin_unlock_bh(&rpc_queue_lock);
}

/*
 * Lock down a sleeping task to prevent it from waking up
 * and disappearing from beneath us.
 *
 * This function should always be called with the
 * rpc_queue_lock held.
 */
int
__rpc_lock_task(struct rpc_task *task)
{
	if (!RPC_IS_RUNNING(task))
		return ++task->tk_lock;
	return 0;
}

void
rpc_unlock_task(struct rpc_task *task)
{
	spin_lock_bh(&rpc_queue_lock);
	if (task->tk_lock && !--task->tk_lock && task->tk_wakeup)
		__rpc_wake_up_task(task);
	spin_unlock_bh(&rpc_queue_lock);
}

/*
 * Run a task at a later time
 */
static void	__rpc_atrun(struct rpc_task *);
void
rpc_delay(struct rpc_task *task, unsigned long delay)
{
	task->tk_timeout = delay;
	rpc_sleep_on(&delay_queue, task, NULL, __rpc_atrun);
}

static void
__rpc_atrun(struct rpc_task *task)
{
	task->tk_status = 0;
	rpc_wake_up_task(task);
}

/*
 * This is the RPC `scheduler' (or rather, the finite state machine).
 */
static int
__rpc_execute(struct rpc_task *task)
{
	int		status = 0;

	dprintk("RPC: %4d rpc_execute flgs %x\n",
				task->tk_pid, task->tk_flags);

	if (!RPC_IS_RUNNING(task)) {
		printk(KERN_WARNING "RPC: rpc_execute called for sleeping task!!\n");
		return 0;
	}

 restarted:
	while (1) {
		/*
		 * Execute any pending callback.
		 */
		if (RPC_DO_CALLBACK(task)) {
			/* Define a callback save pointer */
			void (*save_callback)(struct rpc_task *);
	
			/* 
			 * If a callback exists, save it, reset it,
			 * call it.
			 * The save is needed to stop from resetting
			 * another callback set within the callback handler
			 * - Dave
			 */
			save_callback=task->tk_callback;
			task->tk_callback=NULL;
			save_callback(task);
		}

		/*
		 * Perform the next FSM step.
		 * tk_action may be NULL when the task has been killed
		 * by someone else.
		 */
		if (RPC_IS_RUNNING(task)) {
			/*
			 * Garbage collection of pending timers...
			 */
			rpc_delete_timer(task);
			if (!task->tk_action)
				break;
			task->tk_action(task);
		}

		/*
		 * Check whether task is sleeping.
		 */
		spin_lock_bh(&rpc_queue_lock);
		if (!RPC_IS_RUNNING(task)) {
			rpc_set_sleeping(task);
			if (RPC_IS_ASYNC(task)) {
				spin_unlock_bh(&rpc_queue_lock);
				return 0;
			}
		}
		spin_unlock_bh(&rpc_queue_lock);

		while (RPC_IS_SLEEPING(task)) {
			/* sync task: sleep here */
			dprintk("RPC: %4d sync task going to sleep\n",
							task->tk_pid);
			if (current->pid == rpciod_pid)
				printk(KERN_ERR "RPC: rpciod waiting on sync task!\n");

			__wait_event(task->tk_wait, !RPC_IS_SLEEPING(task));
			dprintk("RPC: %4d sync task resuming\n", task->tk_pid);

			/*
			 * When a sync task receives a signal, it exits with
			 * -ERESTARTSYS. In order to catch any callbacks that
			 * clean up after sleeping on some queue, we don't
			 * break the loop here, but go around once more.