Linux 2.6.26-rc5
[linux-2.6/openmoko-kernel/knife-kernel.git] / block / deadline-iosched.c
blob342448c3d2ddf20432410b121ef3060859c998d3
1 /*
2 * Deadline i/o scheduler.
4 * Copyright (C) 2002 Jens Axboe <axboe@kernel.dk>
5 */
6 #include <linux/kernel.h>
7 #include <linux/fs.h>
8 #include <linux/blkdev.h>
9 #include <linux/elevator.h>
10 #include <linux/bio.h>
11 #include <linux/module.h>
12 #include <linux/slab.h>
13 #include <linux/init.h>
14 #include <linux/compiler.h>
15 #include <linux/rbtree.h>
18 * See Documentation/block/deadline-iosched.txt
20 static const int read_expire = HZ / 2; /* max time before a read is submitted. */
21 static const int write_expire = 5 * HZ; /* ditto for writes, these limits are SOFT! */
22 static const int writes_starved = 2; /* max times reads can starve a write */
23 static const int fifo_batch = 16; /* # of sequential requests treated as one
24 by the above parameters. For throughput. */
26 struct deadline_data {
28 * run time data
32 * requests (deadline_rq s) are present on both sort_list and fifo_list
34 struct rb_root sort_list[2];
35 struct list_head fifo_list[2];
38 * next in sort order. read, write or both are NULL
40 struct request *next_rq[2];
41 unsigned int batching; /* number of sequential requests made */
42 sector_t last_sector; /* head position */
43 unsigned int starved; /* times reads have starved writes */
46 * settings that change how the i/o scheduler behaves
48 int fifo_expire[2];
49 int fifo_batch;
50 int writes_starved;
51 int front_merges;
54 static void deadline_move_request(struct deadline_data *, struct request *);
56 #define RQ_RB_ROOT(dd, rq) (&(dd)->sort_list[rq_data_dir((rq))])
59 * get the request after `rq' in sector-sorted order
61 static inline struct request *
62 deadline_latter_request(struct request *rq)
64 struct rb_node *node = rb_next(&rq->rb_node);
66 if (node)
67 return rb_entry_rq(node);
69 return NULL;
72 static void
73 deadline_add_rq_rb(struct deadline_data *dd, struct request *rq)
75 struct rb_root *root = RQ_RB_ROOT(dd, rq);
76 struct request *__alias;
78 retry:
79 __alias = elv_rb_add(root, rq);
80 if (unlikely(__alias)) {
81 deadline_move_request(dd, __alias);
82 goto retry;
86 static inline void
87 deadline_del_rq_rb(struct deadline_data *dd, struct request *rq)
89 const int data_dir = rq_data_dir(rq);
91 if (dd->next_rq[data_dir] == rq)
92 dd->next_rq[data_dir] = deadline_latter_request(rq);
94 elv_rb_del(RQ_RB_ROOT(dd, rq), rq);
98 * add rq to rbtree and fifo
100 static void
101 deadline_add_request(struct request_queue *q, struct request *rq)
103 struct deadline_data *dd = q->elevator->elevator_data;
104 const int data_dir = rq_data_dir(rq);
106 deadline_add_rq_rb(dd, rq);
109 * set expire time (only used for reads) and add to fifo list
111 rq_set_fifo_time(rq, jiffies + dd->fifo_expire[data_dir]);
112 list_add_tail(&rq->queuelist, &dd->fifo_list[data_dir]);
116 * remove rq from rbtree and fifo.
118 static void deadline_remove_request(struct request_queue *q, struct request *rq)
120 struct deadline_data *dd = q->elevator->elevator_data;
122 rq_fifo_clear(rq);
123 deadline_del_rq_rb(dd, rq);
126 static int
127 deadline_merge(struct request_queue *q, struct request **req, struct bio *bio)
129 struct deadline_data *dd = q->elevator->elevator_data;
130 struct request *__rq;
131 int ret;
134 * check for front merge
136 if (dd->front_merges) {
137 sector_t sector = bio->bi_sector + bio_sectors(bio);
139 __rq = elv_rb_find(&dd->sort_list[bio_data_dir(bio)], sector);
140 if (__rq) {
141 BUG_ON(sector != __rq->sector);
143 if (elv_rq_merge_ok(__rq, bio)) {
144 ret = ELEVATOR_FRONT_MERGE;
145 goto out;
150 return ELEVATOR_NO_MERGE;
151 out:
152 *req = __rq;
153 return ret;
156 static void deadline_merged_request(struct request_queue *q,
157 struct request *req, int type)
159 struct deadline_data *dd = q->elevator->elevator_data;
162 * if the merge was a front merge, we need to reposition request
164 if (type == ELEVATOR_FRONT_MERGE) {
165 elv_rb_del(RQ_RB_ROOT(dd, req), req);
166 deadline_add_rq_rb(dd, req);
170 static void
171 deadline_merged_requests(struct request_queue *q, struct request *req,
172 struct request *next)
175 * if next expires before rq, assign its expire time to rq
176 * and move into next position (next will be deleted) in fifo
178 if (!list_empty(&req->queuelist) && !list_empty(&next->queuelist)) {
179 if (time_before(rq_fifo_time(next), rq_fifo_time(req))) {
180 list_move(&req->queuelist, &next->queuelist);
181 rq_set_fifo_time(req, rq_fifo_time(next));
186 * kill knowledge of next, this one is a goner
188 deadline_remove_request(q, next);
192 * move request from sort list to dispatch queue.
194 static inline void
195 deadline_move_to_dispatch(struct deadline_data *dd, struct request *rq)
197 struct request_queue *q = rq->q;
199 deadline_remove_request(q, rq);
200 elv_dispatch_add_tail(q, rq);
204 * move an entry to dispatch queue
206 static void
207 deadline_move_request(struct deadline_data *dd, struct request *rq)
209 const int data_dir = rq_data_dir(rq);
211 dd->next_rq[READ] = NULL;
212 dd->next_rq[WRITE] = NULL;
213 dd->next_rq[data_dir] = deadline_latter_request(rq);
215 dd->last_sector = rq->sector + rq->nr_sectors;
218 * take it off the sort and fifo list, move
219 * to dispatch queue
221 deadline_move_to_dispatch(dd, rq);
225 * deadline_check_fifo returns 0 if there are no expired reads on the fifo,
226 * 1 otherwise. Requires !list_empty(&dd->fifo_list[data_dir])
228 static inline int deadline_check_fifo(struct deadline_data *dd, int ddir)
230 struct request *rq = rq_entry_fifo(dd->fifo_list[ddir].next);
233 * rq is expired!
235 if (time_after(jiffies, rq_fifo_time(rq)))
236 return 1;
238 return 0;
242 * deadline_dispatch_requests selects the best request according to
243 * read/write expire, fifo_batch, etc
245 static int deadline_dispatch_requests(struct request_queue *q, int force)
247 struct deadline_data *dd = q->elevator->elevator_data;
248 const int reads = !list_empty(&dd->fifo_list[READ]);
249 const int writes = !list_empty(&dd->fifo_list[WRITE]);
250 struct request *rq;
251 int data_dir;
254 * batches are currently reads XOR writes
256 if (dd->next_rq[WRITE])
257 rq = dd->next_rq[WRITE];
258 else
259 rq = dd->next_rq[READ];
261 if (rq) {
262 /* we have a "next request" */
264 if (dd->last_sector != rq->sector)
265 /* end the batch on a non sequential request */
266 dd->batching += dd->fifo_batch;
268 if (dd->batching < dd->fifo_batch)
269 /* we are still entitled to batch */
270 goto dispatch_request;
274 * at this point we are not running a batch. select the appropriate
275 * data direction (read / write)
278 if (reads) {
279 BUG_ON(RB_EMPTY_ROOT(&dd->sort_list[READ]));
281 if (writes && (dd->starved++ >= dd->writes_starved))
282 goto dispatch_writes;
284 data_dir = READ;
286 goto dispatch_find_request;
290 * there are either no reads or writes have been starved
293 if (writes) {
294 dispatch_writes:
295 BUG_ON(RB_EMPTY_ROOT(&dd->sort_list[WRITE]));
297 dd->starved = 0;
299 data_dir = WRITE;
301 goto dispatch_find_request;
304 return 0;
306 dispatch_find_request:
308 * we are not running a batch, find best request for selected data_dir
310 if (deadline_check_fifo(dd, data_dir) || !dd->next_rq[data_dir]) {
312 * A deadline has expired, the last request was in the other
313 * direction, or we have run out of higher-sectored requests.
314 * Start again from the request with the earliest expiry time.
316 rq = rq_entry_fifo(dd->fifo_list[data_dir].next);
317 } else {
319 * The last req was the same dir and we have a next request in
320 * sort order. No expired requests so continue on from here.
322 rq = dd->next_rq[data_dir];
325 dd->batching = 0;
327 dispatch_request:
329 * rq is the selected appropriate request.
331 dd->batching++;
332 deadline_move_request(dd, rq);
334 return 1;
337 static int deadline_queue_empty(struct request_queue *q)
339 struct deadline_data *dd = q->elevator->elevator_data;
341 return list_empty(&dd->fifo_list[WRITE])
342 && list_empty(&dd->fifo_list[READ]);
345 static void deadline_exit_queue(elevator_t *e)
347 struct deadline_data *dd = e->elevator_data;
349 BUG_ON(!list_empty(&dd->fifo_list[READ]));
350 BUG_ON(!list_empty(&dd->fifo_list[WRITE]));
352 kfree(dd);
356 * initialize elevator private data (deadline_data).
358 static void *deadline_init_queue(struct request_queue *q)
360 struct deadline_data *dd;
362 dd = kmalloc_node(sizeof(*dd), GFP_KERNEL | __GFP_ZERO, q->node);
363 if (!dd)
364 return NULL;
366 INIT_LIST_HEAD(&dd->fifo_list[READ]);
367 INIT_LIST_HEAD(&dd->fifo_list[WRITE]);
368 dd->sort_list[READ] = RB_ROOT;
369 dd->sort_list[WRITE] = RB_ROOT;
370 dd->fifo_expire[READ] = read_expire;
371 dd->fifo_expire[WRITE] = write_expire;
372 dd->writes_starved = writes_starved;
373 dd->front_merges = 1;
374 dd->fifo_batch = fifo_batch;
375 return dd;
379 * sysfs parts below
382 static ssize_t
383 deadline_var_show(int var, char *page)
385 return sprintf(page, "%d\n", var);
388 static ssize_t
389 deadline_var_store(int *var, const char *page, size_t count)
391 char *p = (char *) page;
393 *var = simple_strtol(p, &p, 10);
394 return count;
397 #define SHOW_FUNCTION(__FUNC, __VAR, __CONV) \
398 static ssize_t __FUNC(elevator_t *e, char *page) \
400 struct deadline_data *dd = e->elevator_data; \
401 int __data = __VAR; \
402 if (__CONV) \
403 __data = jiffies_to_msecs(__data); \
404 return deadline_var_show(__data, (page)); \
406 SHOW_FUNCTION(deadline_read_expire_show, dd->fifo_expire[READ], 1);
407 SHOW_FUNCTION(deadline_write_expire_show, dd->fifo_expire[WRITE], 1);
408 SHOW_FUNCTION(deadline_writes_starved_show, dd->writes_starved, 0);
409 SHOW_FUNCTION(deadline_front_merges_show, dd->front_merges, 0);
410 SHOW_FUNCTION(deadline_fifo_batch_show, dd->fifo_batch, 0);
411 #undef SHOW_FUNCTION
413 #define STORE_FUNCTION(__FUNC, __PTR, MIN, MAX, __CONV) \
414 static ssize_t __FUNC(elevator_t *e, const char *page, size_t count) \
416 struct deadline_data *dd = e->elevator_data; \
417 int __data; \
418 int ret = deadline_var_store(&__data, (page), count); \
419 if (__data < (MIN)) \
420 __data = (MIN); \
421 else if (__data > (MAX)) \
422 __data = (MAX); \
423 if (__CONV) \
424 *(__PTR) = msecs_to_jiffies(__data); \
425 else \
426 *(__PTR) = __data; \
427 return ret; \
429 STORE_FUNCTION(deadline_read_expire_store, &dd->fifo_expire[READ], 0, INT_MAX, 1);
430 STORE_FUNCTION(deadline_write_expire_store, &dd->fifo_expire[WRITE], 0, INT_MAX, 1);
431 STORE_FUNCTION(deadline_writes_starved_store, &dd->writes_starved, INT_MIN, INT_MAX, 0);
432 STORE_FUNCTION(deadline_front_merges_store, &dd->front_merges, 0, 1, 0);
433 STORE_FUNCTION(deadline_fifo_batch_store, &dd->fifo_batch, 0, INT_MAX, 0);
434 #undef STORE_FUNCTION
436 #define DD_ATTR(name) \
437 __ATTR(name, S_IRUGO|S_IWUSR, deadline_##name##_show, \
438 deadline_##name##_store)
440 static struct elv_fs_entry deadline_attrs[] = {
441 DD_ATTR(read_expire),
442 DD_ATTR(write_expire),
443 DD_ATTR(writes_starved),
444 DD_ATTR(front_merges),
445 DD_ATTR(fifo_batch),
446 __ATTR_NULL
449 static struct elevator_type iosched_deadline = {
450 .ops = {
451 .elevator_merge_fn = deadline_merge,
452 .elevator_merged_fn = deadline_merged_request,
453 .elevator_merge_req_fn = deadline_merged_requests,
454 .elevator_dispatch_fn = deadline_dispatch_requests,
455 .elevator_add_req_fn = deadline_add_request,
456 .elevator_queue_empty_fn = deadline_queue_empty,
457 .elevator_former_req_fn = elv_rb_former_request,
458 .elevator_latter_req_fn = elv_rb_latter_request,
459 .elevator_init_fn = deadline_init_queue,
460 .elevator_exit_fn = deadline_exit_queue,
463 .elevator_attrs = deadline_attrs,
464 .elevator_name = "deadline",
465 .elevator_owner = THIS_MODULE,
468 static int __init deadline_init(void)
470 elv_register(&iosched_deadline);
472 return 0;
475 static void __exit deadline_exit(void)
477 elv_unregister(&iosched_deadline);
480 module_init(deadline_init);
481 module_exit(deadline_exit);
483 MODULE_AUTHOR("Jens Axboe");
484 MODULE_LICENSE("GPL");
485 MODULE_DESCRIPTION("deadline IO scheduler");