linux deadline调度算法源码阅读记录

内核版本：3.13
deadline-iosched.c

#include <linux/kernel.h>
#include <linux/fs.h>
#include <linux/blkdev.h>
#include <linux/elevator.h>
#include <linux/bio.h>
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/compiler.h>
#include <linux/rbtree.h>

static const int read_expire = HZ / 2; /* 提交读操作前的最大时间 */
static const int write_expire = 5 * HZ; /* ditto for writes, these limits are SOFT! */
static const int writes_starved = 2; /* max times reads can starve a write */
//of sequential requests treated as one by the above parameters. For throughput.
static const int fifo_batch = 16;

//截止时间调度程序(1:batching 2:写饥饿 3:请求超时)
struct deadline_data { //相当于该算法的调度器
	//requests (deadline_rq s) are present on both sort_list and fifo_list
	struct rb_root sort_list[2];
	struct list_head fifo_list[2];
	//next in sort order. read, write or both are NULL
	struct request *next_rq[2]; //指向读/写方向上的下一个请求
	unsigned int batching;		/* number of sequential requests made */
	sector_t last_sector;		/* head position */
	unsigned int starved;		/* 写请求被饥饿的次数 */
	//settings that change how the i/o scheduler behaves
	int fifo_expire[2];
	int fifo_batch;		//批量发送的最大值 默认值16
	int writes_starved;	//写请求饥饿线，默认为2 
	int front_merges; //表示能否进行前向合并的检查 默认值给1
};

static void deadline_move_request(struct deadline_data *, struct request *);

//返回根节点
static inline struct rb_root * deadline_rb_root(struct deadline_data *dd, struct request *rq)
{
	return &dd->sort_list[rq_data_dir(rq)];
}
//返回rq的下一个请求
static inline struct request * deadline_latter_request(struct request *rq)
{
	struct rb_node *node = rb_next(&rq->rb_node);

	if (node)
		return rb_entry_rq(node); //返回该节点对应的结构体首地址
	return NULL; //如果是NULL说明已经是最后一个节点了
}
//将请求增加到红黑树里
static void deadline_add_rq_rb(struct deadline_data *dd, struct request *rq)
{
	struct rb_root *root = deadline_rb_root(dd, rq); //先找到根节点

	//将rq加入红黑树里
	elv_rb_add(root, rq);
}
//将请求从红黑树里删除
static inline void deadline_del_rq_rb(struct deadline_data *dd, struct request *rq)
{
	const int data_dir = rq_data_dir(rq);

	//如果要删除的节点等于下一个要处理的节点，那么就更新下一个要处理的节点
	if (dd->next_rq[data_dir] == rq)
		dd->next_rq[data_dir] = deadline_latter_request(rq);

	//将rq从红黑树里移除(第一个参数返回根节点，第二个参数是要删除的节点)
	elv_rb_del(deadline_rb_root(dd, rq), rq);
}

//将请求增加到红黑树里和链表里，并设置该请求的jiffies值
static void deadline_add_request(struct request_queue *q, struct request *rq)
{
	struct deadline_data *dd = q->elevator->elevator_data;
	const int data_dir = rq_data_dir(rq); //读/写方向

	deadline_add_rq_rb(dd, rq); //将rq加入红黑树里
	// 设置rq的jiffies的值
	rq_set_fifo_time(rq, jiffies + dd->fifo_expire[data_dir]);
	//后面的参数是头节点
	list_add_tail(&rq->queuelist, &dd->fifo_list[data_dir]); //将rq加入链表里
}

//remove rq from rbtree and fifo.
static void deadline_remove_request(struct request_queue *q, struct request *rq)
{
	struct deadline_data *dd = q->elevator->elevator_data;

	rq_fifo_clear(rq);
	deadline_del_rq_rb(dd, rq);
}

//这个还不算形成一个完整的request(bio可能还要合并到request里来)
static int deadline_merge(struct request_queue *q, struct request **req, struct bio *bio)
{
	struct deadline_data *dd = q->elevator->elevator_data;
	struct request *__rq;
	int ret;

	if (dd->front_merges) { //向前合并
		sector_t sector = bio_end_sector(bio); //bio的末尾扇区号
		//根据bio的末尾扇区号寻找rq，看rq的起始扇区号是否等于bio的末尾扇区号，相等说明可以向前合并
		__rq = elv_rb_find(&dd->sort_list[bio_data_dir(bio)], sector);
		if (__rq) {
			BUG_ON(sector != blk_rq_pos(__rq));
			if (elv_rq_merge_ok(__rq, bio)) {
				ret = ELEVATOR_FRONT_MERGE;
				goto out;
			}
		}
	}
	return ELEVATOR_NO_MERGE;
out:
	*req = __rq; //返回找到的request(这个req可以和bio进行前向合并)
	return ret;
}

//这个是已经进行了合并以后的操作，req合并以后起始扇区号发生了改变，所以需要从
//红黑树里删除节点，然后在插入，这样就会根据合并后扇区的起始号将req插入到合适的位置
static void deadline_merged_request(struct request_queue *q, struct request *req, int type)
{
	//响应合并后的处理
	struct deadline_data *dd = q->elevator->elevator_data;

	if (type == ELEVATOR_FRONT_MERGE) {
		//先删除
		elv_rb_del(deadline_rb_root(dd, req), req);
		//再增加
		deadline_add_rq_rb(dd, req);
	}
}

static void deadline_merged_requests(struct request_queue *q, struct request *req, struct request *next)
{
	/*如果next在rq之前过期，则将其过期时间分配给rq，
		并在fifo中移动到下一个位置(next将被删除)
	*/
	if (!list_empty(&req->queuelist) && !list_empty(&next->queuelist)) {
		//如果next的时间比req的时间早
		if (time_before(rq_fifo_time(next), rq_fifo_time(req))) {
			//这两个节点的顺序要根据时间值对换下
			list_move(&req->queuelist, &next->queuelist);
			//那么 合并后的req的时间值就要改成next的时间值
			rq_set_fifo_time(req, rq_fifo_time(next));
		}
	}
	//因为已经合并了，所以next可以删除了
	deadline_remove_request(q, next);
}

// 将请求从排序列表移动到调度队列。
static inline void deadline_move_to_dispatch(struct deadline_data *dd, struct request *rq)
{
	struct request_queue *q = rq->q;

	//将请求rq从deadline调度器的红黑树以及fifo_list链表中删掉，并且移除其超时时间
	deadline_remove_request(q, rq);
	//将请求rq插入 到块设备的请求队列
	elv_dispatch_add_tail(q, rq);
}

//将条目移动到调度队列
static void deadline_move_request(struct deadline_data *dd, struct request *rq)
{
	const int data_dir = rq_data_dir(rq);

	dd->next_rq[READ] = NULL;
	dd->next_rq[WRITE] = NULL;
	//next_rq赋值的地方，是从红黑树里面取出的(第一次时为NULL)
	dd->next_rq[data_dir] = deadline_latter_request(rq);
	dd->last_sector = rq_end_sector(rq);
	//将其从排序和fifo列表中移除，移到调度队列中
	deadline_move_to_dispatch(dd, rq);
}
/*
	如果fifo上没有过期的请求，Deadline_check_fifo返回0，
	否则返回1。需要! list_empty (dd - > fifo_list [data_dir])
*/
static inline int deadline_check_fifo(struct deadline_data *dd, int ddir)
{
	struct request *rq = rq_entry_fifo(dd->fifo_list[ddir].next); //返回第一个节点

	//rq is expired!(当前的jiffies值大于rq的jiffies值)
	if (time_after_eq(jiffies, rq_fifo_time(rq)))
		return 1; //返回1说明必须要处理这个request了

	return 0;
}
//Deadline_dispatch_requests根据读/写过期、fif_batch等选择最佳请求
static int deadline_dispatch_requests(struct request_queue *q, int force)
{
	struct deadline_data *dd = q->elevator->elevator_data;
	const int reads = !list_empty(&dd->fifo_list[READ]); //链表为空 reads值为0
	const int writes = !list_empty(&dd->fifo_list[WRITE]); //链表为空 writes值为0
	struct request *rq;
	int data_dir;

	if (dd->next_rq[WRITE])
		rq = dd->next_rq[WRITE]; //先判断的写请求
	else
		rq = dd->next_rq[READ]; //有读请求

	if (rq && dd->batching < dd->fifo_batch) //fifo_batch默认值16
		goto dispatch_request;

	if (reads) { //为1说明链表不为空，有读请求
		BUG_ON(RB_EMPTY_ROOT(&dd->sort_list[READ]));
		//判断写请求是否已经饥饿，如果是则优先处理
		if (writes && (dd->starved++ >= dd->writes_starved))
			goto dispatch_writes;
		data_dir = READ;
		goto dispatch_find_request;
	}
	if (writes) { //为1说明链表不为空,有写请求
dispatch_writes:
		BUG_ON(RB_EMPTY_ROOT(&dd->sort_list[WRITE]));
		dd->starved = 0; //饥饿已经处理了，值更新为0
		data_dir = WRITE;
		goto dispatch_find_request;
	}
	return 0;
dispatch_find_request:
	/*
		1:查找是否有过期的读请求或者过期的写请求，如果有，那就取出
		2:next_rq为空的话req就从fifo_list里面取
	*/
	if (deadline_check_fifo(dd, data_dir) || !dd->next_rq[data_dir]) {
		rq = rq_entry_fifo(dd->fifo_list[data_dir].next); //这个req是从链表里取的
	} else {
		rq = dd->next_rq[data_dir]; //这个req是从红黑树里取的
	}
	dd->batching = 0; //这个值的初始化应该是0
dispatch_request:
	dd->batching++;
	//deadline_dispatch_requests -> deadline_move_request -> deadline_move_to_dispatch
	deadline_move_request(dd, rq);
	return 1;
}

static void deadline_exit_queue(struct elevator_queue *e)
{
	struct deadline_data *dd = e->elevator_data;

	BUG_ON(!list_empty(&dd->fifo_list[READ]));
	BUG_ON(!list_empty(&dd->fifo_list[WRITE]));
	kfree(dd);
}

static int deadline_init_queue(struct request_queue *q, struct elevator_type *e)
{
	struct deadline_data *dd;
	struct elevator_queue *eq;

	eq = elevator_alloc(q, e);
	if (!eq)
		return -ENOMEM;

	dd = kzalloc_node(sizeof(*dd), GFP_KERNEL, q->node);
	if (!dd) {
		kobject_put(&eq->kobj);
		return -ENOMEM;
	}
	eq->elevator_data = dd;
	//链表初始化
	INIT_LIST_HEAD(&dd->fifo_list[READ]);
	INIT_LIST_HEAD(&dd->fifo_list[WRITE]);
	//红黑树根节点初始化
	dd->sort_list[READ] = RB_ROOT;
	dd->sort_list[WRITE] = RB_ROOT;
	dd->fifo_expire[READ] = read_expire;
	dd->fifo_expire[WRITE] = write_expire;
	dd->writes_starved = writes_starved;
	dd->front_merges = 1;
	dd->fifo_batch = fifo_batch;
	spin_lock_irq(q->queue_lock);
	q->elevator = eq; //赋值给request_queue的对应变量
	spin_unlock_irq(q->queue_lock);
	return 0;
}

static ssize_t deadline_var_show(int var, char *page)
{
	return sprintf(page, "%d\n", var);
}

static ssize_t deadline_var_store(int *var, const char *page, size_t count)
{
	char *p = (char *) page;

	*var = simple_strtol(p, &p, 10);
	return count;
}

#define SHOW_FUNCTION(__FUNC, __VAR, __CONV)				\
static ssize_t __FUNC(struct elevator_queue *e, char *page)		\
{									\
	struct deadline_data *dd = e->elevator_data;			\
	int __data = __VAR;						\
	if (__CONV)							\
		__data = jiffies_to_msecs(__data);			\
	return deadline_var_show(__data, (page));			\
}
SHOW_FUNCTION(deadline_read_expire_show, dd->fifo_expire[READ], 1);
SHOW_FUNCTION(deadline_write_expire_show, dd->fifo_expire[WRITE], 1);
SHOW_FUNCTION(deadline_writes_starved_show, dd->writes_starved, 0);
SHOW_FUNCTION(deadline_front_merges_show, dd->front_merges, 0);
SHOW_FUNCTION(deadline_fifo_batch_show, dd->fifo_batch, 0);
#undef SHOW_FUNCTION

#define STORE_FUNCTION(__FUNC, __PTR, MIN, MAX, __CONV)			\
static ssize_t __FUNC(struct elevator_queue *e, const char *page, size_t count)	\
{									\
	struct deadline_data *dd = e->elevator_data;			\
	int __data;							\
	int ret = deadline_var_store(&__data, (page), count);		\
	if (__data < (MIN))						\
		__data = (MIN);						\
	else if (__data > (MAX))					\
		__data = (MAX);						\
	if (__CONV)							\
		*(__PTR) = msecs_to_jiffies(__data);			\
	else								\
		*(__PTR) = __data;					\
	return ret;							\
}
STORE_FUNCTION(deadline_read_expire_store, &dd->fifo_expire[READ], 0, INT_MAX, 1);
STORE_FUNCTION(deadline_write_expire_store, &dd->fifo_expire[WRITE], 0, INT_MAX, 1);
STORE_FUNCTION(deadline_writes_starved_store, &dd->writes_starved, INT_MIN, INT_MAX, 0);
STORE_FUNCTION(deadline_front_merges_store, &dd->front_merges, 0, 1, 0);
STORE_FUNCTION(deadline_fifo_batch_store, &dd->fifo_batch, 0, INT_MAX, 0);
#undef STORE_FUNCTION

#define DD_ATTR(name) \
	__ATTR(name, S_IRUGO|S_IWUSR, deadline_##name##_show, \
				      deadline_##name##_store)

static struct elv_fs_entry deadline_attrs[] = {
	DD_ATTR(read_expire),
	DD_ATTR(write_expire),
	DD_ATTR(writes_starved),
	DD_ATTR(front_merges),
	DD_ATTR(fifo_batch),
	__ATTR_NULL
};

static struct elevator_type iosched_deadline = {
	.ops = {
		.elevator_merge_fn = 		deadline_merge,
		.elevator_merged_fn =		deadline_merged_request,
		.elevator_merge_req_fn =	deadline_merged_requests,
		.elevator_dispatch_fn =		deadline_dispatch_requests,
		.elevator_add_req_fn =		deadline_add_request,
		.elevator_former_req_fn =	elv_rb_former_request,
		.elevator_latter_req_fn =	elv_rb_latter_request,
		.elevator_init_fn =		deadline_init_queue,
		.elevator_exit_fn =		deadline_exit_queue,
	},
	.elevator_attrs = deadline_attrs,
	.elevator_name = "deadline",
	.elevator_owner = THIS_MODULE,
};

static int __init deadline_init(void)
{
	return elv_register(&iosched_deadline);
}

static void __exit deadline_exit(void)
{
	elv_unregister(&iosched_deadline);
}

module_init(deadline_init);
module_exit(deadline_exit);

MODULE_AUTHOR("Jens Axboe");
MODULE_LICENSE("GPL");
MODULE_DESCRIPTION("deadline IO scheduler");

参考：
1：//https://www.bilibili.com/read/cv18210887/
2：//https://blog.csdn.net/moliyiye/article/details/49492929