path: root/net/sunrpc/sched.c

/*
 * linux/net/sunrpc/sched.c
 *
 * Scheduling for synchronous and asynchronous RPC requests.
 *
 * Copyright (C) 1996 Olaf Kirch, <okir@monad.swb.de>
 *
 * TCP NFS related read + write fixes
 * (C) 1999 Dave Airlie, University of Limerick, Ireland <airlied@linux.ie>
 */

#include <linux/module.h>

#include <linux/sched.h>
#include <linux/interrupt.h>
#include <linux/slab.h>
#include <linux/mempool.h>
#include <linux/smp.h>
#include <linux/spinlock.h>
#include <linux/mutex.h>
#include <linux/freezer.h>

#include <linux/sunrpc/clnt.h>

#include "sunrpc.h"

#ifdef RPC_DEBUG
#define RPCDBG_FACILITY		RPCDBG_SCHED
#endif

#define CREATE_TRACE_POINTS
#include <trace/events/sunrpc.h>

/*
 * RPC slabs and memory pools
 */
#define RPC_BUFFER_MAXSIZE	(2048)
#define RPC_BUFFER_POOLSIZE	(8)
#define RPC_TASK_POOLSIZE	(8)
static struct kmem_cache	*rpc_task_slabp __read_mostly;
static struct kmem_cache	*rpc_buffer_slabp __read_mostly;
static mempool_t	*rpc_task_mempool __read_mostly;
static mempool_t	*rpc_buffer_mempool __read_mostly;

static void			rpc_async_schedule(struct work_struct *);
static void			 rpc_release_task(struct rpc_task *task);
static void __rpc_queue_timer_fn(unsigned long ptr);

/*
 * RPC tasks sit here while waiting for conditions to improve.
 */
static struct rpc_wait_queue delay_queue;

/*
 * rpciod-related stuff
 */
struct workqueue_struct *rpciod_workqueue;

/*
 * Disable the timer for a given RPC task. Should be called with
 * queue->lock and bh_disabled in order to avoid races within
 * rpc_run_timer().
 */
static void
__rpc_disable_timer(struct rpc_wait_queue *queue, struct rpc_task *task)
{
	if (task->tk_timeout == 0)
		return;
	dprintk("RPC: %5u disabling timer\n", task->tk_pid);
	task->tk_timeout = 0;
	list_del(&task->u.tk_wait.timer_list);
	if (list_empty(&queue->timer_list.list))
		del_timer(&queue->timer_list.timer);
}

static void
rpc_set_queue_timer(struct rpc_wait_queue *queue, unsigned long expires)
{
	queue->timer_list.expires = expires;
	mod_timer(&queue->timer_list.timer, expires);
}

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

	dprintk("RPC: %5u setting alarm for %lu ms\n",
			task->tk_pid, task->tk_timeout * 1000 / HZ);

	task->u.tk_wait.expires = jiffies + task->tk_timeout;
	if (list_empty(&queue->timer_list.list) || time_before(task->u.tk_wait.expires, queue->timer_list.expires))
		rpc_set_queue_timer(queue, task->u.tk_wait.expires);
	list_add(&task->u.tk_wait.timer_list, &queue->timer_list.list);
}

static void rpc_set_waitqueue_priority(struct rpc_wait_queue *queue, int priority)
{
	queue->priority = priority;
}

static void rpc_set_waitqueue_owner(struct rpc_wait_queue *queue, pid_t pid)
{
	queue->owner = pid;
	queue->nr = RPC_BATCH_COUNT;
}

static void rpc_reset_waitqueue_priority(struct rpc_wait_queue *queue)
{
	rpc_set_waitqueue_priority(queue, queue->maxpriority);
	rpc_set_waitqueue_owner(queue, 0);
}

/*
 * Add new request to a priority queue.
 */
static void __rpc_add_wait_queue_priority(struct rpc_wait_queue *queue,
		struct rpc_task *task,
		unsigned char queue_priority)
{
	struct list_head *q;
	struct rpc_task *t;

	INIT_LIST_HEAD(&task->u.tk_wait.links);
	if (unlikely(queue_priority > queue->maxpriority))
		queue_priority = queue->maxpriority;
	if (queue_priority > queue->priority)
		rpc_set_waitqueue_priority(queue, queue_priority);
	q = &queue->tasks[queue_priority];
	list_for_each_entry(t, q, u