2 * CFQ, or complete fairness queueing, disk scheduler.
4 * Based on ideas from a previously unfinished io
5 * scheduler (round robin per-process disk scheduling) and Andrea Arcangeli.
7 * Copyright (C) 2003 Jens Axboe <axboe@kernel.dk>
9 #include <linux/module.h>
10 #include <linux/blkdev.h>
11 #include <linux/elevator.h>
12 #include <linux/jiffies.h>
13 #include <linux/rbtree.h>
14 #include <linux/ioprio.h>
15 #include <linux/blktrace_api.h>
16 #include "blk-cgroup.h"
17 #include "cfq-iosched.h"
22 /* max queue in one round of service */
23 static const int cfq_quantum
= 4;
24 static const int cfq_fifo_expire
[2] = { HZ
/ 4, HZ
/ 8 };
25 /* maximum backwards seek, in KiB */
26 static const int cfq_back_max
= 16 * 1024;
27 /* penalty of a backwards seek */
28 static const int cfq_back_penalty
= 2;
29 static const int cfq_slice_sync
= HZ
/ 10;
30 static int cfq_slice_async
= HZ
/ 25;
31 static const int cfq_slice_async_rq
= 2;
32 static int cfq_slice_idle
= HZ
/ 125;
33 static const int cfq_target_latency
= HZ
* 3/10; /* 300 ms */
34 static const int cfq_hist_divisor
= 4;
37 * offset from end of service tree
39 #define CFQ_IDLE_DELAY (HZ / 5)
42 * below this threshold, we consider thinktime immediate
44 #define CFQ_MIN_TT (2)
47 * Allow merged cfqqs to perform this amount of seeky I/O before
48 * deciding to break the queues up again.
50 #define CFQQ_COOP_TOUT (HZ)
52 #define CFQ_SLICE_SCALE (5)
53 #define CFQ_HW_QUEUE_MIN (5)
54 #define CFQ_SERVICE_SHIFT 12
57 ((struct cfq_io_context *) (rq)->elevator_private)
58 #define RQ_CFQQ(rq) (struct cfq_queue *) ((rq)->elevator_private2)
60 static struct kmem_cache
*cfq_pool
;
61 static struct kmem_cache
*cfq_ioc_pool
;
63 static DEFINE_PER_CPU(unsigned long, cfq_ioc_count
);
64 static struct completion
*ioc_gone
;
65 static DEFINE_SPINLOCK(ioc_gone_lock
);
67 #define CFQ_PRIO_LISTS IOPRIO_BE_NR
68 #define cfq_class_idle(cfqq) ((cfqq)->ioprio_class == IOPRIO_CLASS_IDLE)
69 #define cfq_class_rt(cfqq) ((cfqq)->ioprio_class == IOPRIO_CLASS_RT)
71 #define sample_valid(samples) ((samples) > 80)
72 #define rb_entry_cfqg(node) rb_entry((node), struct cfq_group, rb_node)
75 * Most of our rbtree usage is for sorting with min extraction, so
76 * if we cache the leftmost node we don't have to walk down the tree
77 * to find it. Idea borrowed from Ingo Molnars CFS scheduler. We should
78 * move this into the elevator for the rq sorting as well.
85 struct rb_node
*active
;
86 unsigned total_weight
;
88 #define CFQ_RB_ROOT (struct cfq_rb_root) { RB_ROOT, NULL, 0, 0, }
91 * Per process-grouping structure
96 /* various state flags, see below */
99 struct cfq_data
*cfqd
;
100 /* service_tree member */
101 struct rb_node rb_node
;
102 /* service_tree key */
103 unsigned long rb_key
;
104 /* prio tree member */
105 struct rb_node p_node
;
106 /* prio tree root we belong to, if any */
107 struct rb_root
*p_root
;
108 /* sorted list of pending requests */
109 struct rb_root sort_list
;
110 /* if fifo isn't expired, next request to serve */
111 struct request
*next_rq
;
112 /* requests queued in sort_list */
114 /* currently allocated requests */
116 /* fifo list of requests in sort_list */
117 struct list_head fifo
;
119 /* time when queue got scheduled in to dispatch first request. */
120 unsigned long dispatch_start
;
121 unsigned int allocated_slice
;
122 /* time when first request from queue completed and slice started. */
123 unsigned long slice_start
;
124 unsigned long slice_end
;
126 unsigned int slice_dispatch
;
128 /* pending metadata requests */
130 /* number of requests that are on the dispatch list or inside driver */
133 /* io prio of this group */
134 unsigned short ioprio
, org_ioprio
;
135 unsigned short ioprio_class
, org_ioprio_class
;
137 unsigned int seek_samples
;
140 sector_t last_request_pos
;
141 unsigned long seeky_start
;
145 struct cfq_rb_root
*service_tree
;
146 struct cfq_queue
*new_cfqq
;
147 struct cfq_group
*cfqg
;
148 struct cfq_group
*orig_cfqg
;
149 /* Sectors dispatched in current dispatch round */
150 unsigned long nr_sectors
;
154 * First index in the service_trees.
155 * IDLE is handled separately, so it has negative index
164 * Second index in the service_trees.
168 SYNC_NOIDLE_WORKLOAD
= 1,
172 /* This is per cgroup per device grouping structure */
174 /* group service_tree member */
175 struct rb_node rb_node
;
177 /* group service_tree key */
182 /* number of cfqq currently on this group */
185 /* Per group busy queus average. Useful for workload slice calc. */
186 unsigned int busy_queues_avg
[2];
188 * rr lists of queues with requests, onle rr for each priority class.
189 * Counts are embedded in the cfq_rb_root
191 struct cfq_rb_root service_trees
[2][3];
192 struct cfq_rb_root service_tree_idle
;
194 unsigned long saved_workload_slice
;
195 enum wl_type_t saved_workload
;
196 enum wl_prio_t saved_serving_prio
;
197 struct blkio_group blkg
;
198 #ifdef CONFIG_CFQ_GROUP_IOSCHED
199 struct hlist_node cfqd_node
;
205 * Per block device queue structure
208 struct request_queue
*queue
;
209 /* Root service tree for cfq_groups */
210 struct cfq_rb_root grp_service_tree
;
211 struct cfq_group root_group
;
212 /* Number of active cfq groups on group service tree */
216 * The priority currently being served
218 enum wl_prio_t serving_prio
;
219 enum wl_type_t serving_type
;
220 unsigned long workload_expires
;
221 struct cfq_group
*serving_group
;
222 bool noidle_tree_requires_idle
;
225 * Each priority tree is sorted by next_request position. These
226 * trees are used when determining if two or more queues are
227 * interleaving requests (see cfq_close_cooperator).
229 struct rb_root prio_trees
[CFQ_PRIO_LISTS
];
231 unsigned int busy_queues
;
237 * queue-depth detection
243 * -1 => indeterminate, (cfq will behave as if NCQ is present, to allow better detection)
244 * 1 => NCQ is present (hw_tag_est_depth is the estimated max depth)
247 int hw_tag_est_depth
;
248 unsigned int hw_tag_samples
;
251 * idle window management
253 struct timer_list idle_slice_timer
;
254 struct work_struct unplug_work
;
256 struct cfq_queue
*active_queue
;
257 struct cfq_io_context
*active_cic
;
260 * async queue for each priority case
262 struct cfq_queue
*async_cfqq
[2][IOPRIO_BE_NR
];
263 struct cfq_queue
*async_idle_cfqq
;
265 sector_t last_position
;
268 * tunables, see top of file
270 unsigned int cfq_quantum
;
271 unsigned int cfq_fifo_expire
[2];
272 unsigned int cfq_back_penalty
;
273 unsigned int cfq_back_max
;
274 unsigned int cfq_slice
[2];
275 unsigned int cfq_slice_async_rq
;
276 unsigned int cfq_slice_idle
;
277 unsigned int cfq_latency
;
278 unsigned int cfq_group_isolation
;
280 struct list_head cic_list
;
283 * Fallback dummy cfqq for extreme OOM conditions
285 struct cfq_queue oom_cfqq
;
287 unsigned long last_end_sync_rq
;
289 /* List of cfq groups being managed on this device*/
290 struct hlist_head cfqg_list
;
293 static struct cfq_group
*cfq_get_next_cfqg(struct cfq_data
*cfqd
);
295 static struct cfq_rb_root
*service_tree_for(struct cfq_group
*cfqg
,
298 struct cfq_data
*cfqd
)
303 if (prio
== IDLE_WORKLOAD
)
304 return &cfqg
->service_tree_idle
;
306 return &cfqg
->service_trees
[prio
][type
];
309 enum cfqq_state_flags
{
310 CFQ_CFQQ_FLAG_on_rr
= 0, /* on round-robin busy list */
311 CFQ_CFQQ_FLAG_wait_request
, /* waiting for a request */
312 CFQ_CFQQ_FLAG_must_dispatch
, /* must be allowed a dispatch */
313 CFQ_CFQQ_FLAG_must_alloc_slice
, /* per-slice must_alloc flag */
314 CFQ_CFQQ_FLAG_fifo_expire
, /* FIFO checked in this slice */
315 CFQ_CFQQ_FLAG_idle_window
, /* slice idling enabled */
316 CFQ_CFQQ_FLAG_prio_changed
, /* task priority has changed */
317 CFQ_CFQQ_FLAG_slice_new
, /* no requests dispatched in slice */
318 CFQ_CFQQ_FLAG_sync
, /* synchronous queue */
319 CFQ_CFQQ_FLAG_coop
, /* cfqq is shared */
320 CFQ_CFQQ_FLAG_deep
, /* sync cfqq experienced large depth */
321 CFQ_CFQQ_FLAG_wait_busy
, /* Waiting for next request */
322 CFQ_CFQQ_FLAG_wait_busy_done
, /* Got new request. Expire the queue */
325 #define CFQ_CFQQ_FNS(name) \
326 static inline void cfq_mark_cfqq_##name(struct cfq_queue *cfqq) \
328 (cfqq)->flags |= (1 << CFQ_CFQQ_FLAG_##name); \
330 static inline void cfq_clear_cfqq_##name(struct cfq_queue *cfqq) \
332 (cfqq)->flags &= ~(1 << CFQ_CFQQ_FLAG_##name); \
334 static inline int cfq_cfqq_##name(const struct cfq_queue *cfqq) \
336 return ((cfqq)->flags & (1 << CFQ_CFQQ_FLAG_##name)) != 0; \
340 CFQ_CFQQ_FNS(wait_request
);
341 CFQ_CFQQ_FNS(must_dispatch
);
342 CFQ_CFQQ_FNS(must_alloc_slice
);
343 CFQ_CFQQ_FNS(fifo_expire
);
344 CFQ_CFQQ_FNS(idle_window
);
345 CFQ_CFQQ_FNS(prio_changed
);
346 CFQ_CFQQ_FNS(slice_new
);
350 CFQ_CFQQ_FNS(wait_busy
);
351 CFQ_CFQQ_FNS(wait_busy_done
);
354 #ifdef CONFIG_DEBUG_CFQ_IOSCHED
355 #define cfq_log_cfqq(cfqd, cfqq, fmt, args...) \
356 blk_add_trace_msg((cfqd)->queue, "cfq%d%c %s " fmt, (cfqq)->pid, \
357 cfq_cfqq_sync((cfqq)) ? 'S' : 'A', \
358 blkg_path(&(cfqq)->cfqg->blkg), ##args);
360 #define cfq_log_cfqg(cfqd, cfqg, fmt, args...) \
361 blk_add_trace_msg((cfqd)->queue, "%s " fmt, \
362 blkg_path(&(cfqg)->blkg), ##args); \
365 #define cfq_log_cfqq(cfqd, cfqq, fmt, args...) \
366 blk_add_trace_msg((cfqd)->queue, "cfq%d " fmt, (cfqq)->pid, ##args)
367 #define cfq_log_cfqg(cfqd, cfqg, fmt, args...) do {} while (0);
369 #define cfq_log(cfqd, fmt, args...) \
370 blk_add_trace_msg((cfqd)->queue, "cfq " fmt, ##args)
372 /* Traverses through cfq group service trees */
373 #define for_each_cfqg_st(cfqg, i, j, st) \
374 for (i = 0; i <= IDLE_WORKLOAD; i++) \
375 for (j = 0, st = i < IDLE_WORKLOAD ? &cfqg->service_trees[i][j]\
376 : &cfqg->service_tree_idle; \
377 (i < IDLE_WORKLOAD && j <= SYNC_WORKLOAD) || \
378 (i == IDLE_WORKLOAD && j == 0); \
379 j++, st = i < IDLE_WORKLOAD ? \
380 &cfqg->service_trees[i][j]: NULL) \
383 static inline enum wl_prio_t cfqq_prio(struct cfq_queue *cfqq)
385 if (cfq_class_idle(cfqq
))
386 return IDLE_WORKLOAD
;
387 if (cfq_class_rt(cfqq
))
393 static enum wl_type_t
cfqq_type(struct cfq_queue
*cfqq
)
395 if (!cfq_cfqq_sync(cfqq
))
396 return ASYNC_WORKLOAD
;
397 if (!cfq_cfqq_idle_window(cfqq
))
398 return SYNC_NOIDLE_WORKLOAD
;
399 return SYNC_WORKLOAD
;
402 static inline int cfq_group_busy_queues_wl(enum wl_prio_t wl
,
403 struct cfq_data
*cfqd
,
404 struct cfq_group
*cfqg
)
406 if (wl
== IDLE_WORKLOAD
)
407 return cfqg
->service_tree_idle
.count
;
409 return cfqg
->service_trees
[wl
][ASYNC_WORKLOAD
].count
410 + cfqg
->service_trees
[wl
][SYNC_NOIDLE_WORKLOAD
].count
411 + cfqg
->service_trees
[wl
][SYNC_WORKLOAD
].count
;
414 static inline int cfqg_busy_async_queues(struct cfq_data
*cfqd
,
415 struct cfq_group
*cfqg
)
417 return cfqg
->service_trees
[RT_WORKLOAD
][ASYNC_WORKLOAD
].count
418 + cfqg
->service_trees
[BE_WORKLOAD
][ASYNC_WORKLOAD
].count
;
421 static void cfq_dispatch_insert(struct request_queue
*, struct request
*);
422 static struct cfq_queue
*cfq_get_queue(struct cfq_data
*, bool,
423 struct io_context
*, gfp_t
);
424 static struct cfq_io_context
*cfq_cic_lookup(struct cfq_data
*,
425 struct io_context
*);
427 static inline int rq_in_driver(struct cfq_data
*cfqd
)
429 return cfqd
->rq_in_driver
[0] + cfqd
->rq_in_driver
[1];
432 static inline struct cfq_queue
*cic_to_cfqq(struct cfq_io_context
*cic
,
435 return cic
->cfqq
[is_sync
];
438 static inline void cic_set_cfqq(struct cfq_io_context
*cic
,
439 struct cfq_queue
*cfqq
, bool is_sync
)
441 cic
->cfqq
[is_sync
] = cfqq
;
445 * We regard a request as SYNC, if it's either a read or has the SYNC bit
446 * set (in which case it could also be direct WRITE).
448 static inline bool cfq_bio_sync(struct bio
*bio
)
450 return bio_data_dir(bio
) == READ
|| bio_rw_flagged(bio
, BIO_RW_SYNCIO
);
454 * scheduler run of queue, if there are requests pending and no one in the
455 * driver that will restart queueing
457 static inline void cfq_schedule_dispatch(struct cfq_data
*cfqd
)
459 if (cfqd
->busy_queues
) {
460 cfq_log(cfqd
, "schedule dispatch");
461 kblockd_schedule_work(cfqd
->queue
, &cfqd
->unplug_work
);
465 static int cfq_queue_empty(struct request_queue
*q
)
467 struct cfq_data
*cfqd
= q
->elevator
->elevator_data
;
469 return !cfqd
->rq_queued
;
473 * Scale schedule slice based on io priority. Use the sync time slice only
474 * if a queue is marked sync and has sync io queued. A sync queue with async
475 * io only, should not get full sync slice length.
477 static inline int cfq_prio_slice(struct cfq_data
*cfqd
, bool sync
,
480 const int base_slice
= cfqd
->cfq_slice
[sync
];
482 WARN_ON(prio
>= IOPRIO_BE_NR
);
484 return base_slice
+ (base_slice
/CFQ_SLICE_SCALE
* (4 - prio
));
488 cfq_prio_to_slice(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
)
490 return cfq_prio_slice(cfqd
, cfq_cfqq_sync(cfqq
), cfqq
->ioprio
);
493 static inline u64
cfq_scale_slice(unsigned long delta
, struct cfq_group
*cfqg
)
495 u64 d
= delta
<< CFQ_SERVICE_SHIFT
;
497 d
= d
* BLKIO_WEIGHT_DEFAULT
;
498 do_div(d
, cfqg
->weight
);
502 static inline u64
max_vdisktime(u64 min_vdisktime
, u64 vdisktime
)
504 s64 delta
= (s64
)(vdisktime
- min_vdisktime
);
506 min_vdisktime
= vdisktime
;
508 return min_vdisktime
;
511 static inline u64
min_vdisktime(u64 min_vdisktime
, u64 vdisktime
)
513 s64 delta
= (s64
)(vdisktime
- min_vdisktime
);
515 min_vdisktime
= vdisktime
;
517 return min_vdisktime
;
520 static void update_min_vdisktime(struct cfq_rb_root
*st
)
522 u64 vdisktime
= st
->min_vdisktime
;
523 struct cfq_group
*cfqg
;
526 cfqg
= rb_entry_cfqg(st
->active
);
527 vdisktime
= cfqg
->vdisktime
;
531 cfqg
= rb_entry_cfqg(st
->left
);
532 vdisktime
= min_vdisktime(vdisktime
, cfqg
->vdisktime
);
535 st
->min_vdisktime
= max_vdisktime(st
->min_vdisktime
, vdisktime
);
539 * get averaged number of queues of RT/BE priority.
540 * average is updated, with a formula that gives more weight to higher numbers,
541 * to quickly follows sudden increases and decrease slowly
544 static inline unsigned cfq_group_get_avg_queues(struct cfq_data
*cfqd
,
545 struct cfq_group
*cfqg
, bool rt
)
547 unsigned min_q
, max_q
;
548 unsigned mult
= cfq_hist_divisor
- 1;
549 unsigned round
= cfq_hist_divisor
/ 2;
550 unsigned busy
= cfq_group_busy_queues_wl(rt
, cfqd
, cfqg
);
552 min_q
= min(cfqg
->busy_queues_avg
[rt
], busy
);
553 max_q
= max(cfqg
->busy_queues_avg
[rt
], busy
);
554 cfqg
->busy_queues_avg
[rt
] = (mult
* max_q
+ min_q
+ round
) /
556 return cfqg
->busy_queues_avg
[rt
];
559 static inline unsigned
560 cfq_group_slice(struct cfq_data
*cfqd
, struct cfq_group
*cfqg
)
562 struct cfq_rb_root
*st
= &cfqd
->grp_service_tree
;
564 return cfq_target_latency
* cfqg
->weight
/ st
->total_weight
;
568 cfq_set_prio_slice(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
)
570 unsigned slice
= cfq_prio_to_slice(cfqd
, cfqq
);
571 if (cfqd
->cfq_latency
) {
573 * interested queues (we consider only the ones with the same
574 * priority class in the cfq group)
576 unsigned iq
= cfq_group_get_avg_queues(cfqd
, cfqq
->cfqg
,
578 unsigned sync_slice
= cfqd
->cfq_slice
[1];
579 unsigned expect_latency
= sync_slice
* iq
;
580 unsigned group_slice
= cfq_group_slice(cfqd
, cfqq
->cfqg
);
582 if (expect_latency
> group_slice
) {
583 unsigned base_low_slice
= 2 * cfqd
->cfq_slice_idle
;
584 /* scale low_slice according to IO priority
585 * and sync vs async */
587 min(slice
, base_low_slice
* slice
/ sync_slice
);
588 /* the adapted slice value is scaled to fit all iqs
589 * into the target latency */
590 slice
= max(slice
* group_slice
/ expect_latency
,
594 cfqq
->slice_start
= jiffies
;
595 cfqq
->slice_end
= jiffies
+ slice
;
596 cfqq
->allocated_slice
= slice
;
597 cfq_log_cfqq(cfqd
, cfqq
, "set_slice=%lu", cfqq
->slice_end
- jiffies
);
601 * We need to wrap this check in cfq_cfqq_slice_new(), since ->slice_end
602 * isn't valid until the first request from the dispatch is activated
603 * and the slice time set.
605 static inline bool cfq_slice_used(struct cfq_queue
*cfqq
)
607 if (cfq_cfqq_slice_new(cfqq
))
609 if (time_before(jiffies
, cfqq
->slice_end
))
616 * Lifted from AS - choose which of rq1 and rq2 that is best served now.
617 * We choose the request that is closest to the head right now. Distance
618 * behind the head is penalized and only allowed to a certain extent.
620 static struct request
*
621 cfq_choose_req(struct cfq_data
*cfqd
, struct request
*rq1
, struct request
*rq2
, sector_t last
)
623 sector_t s1
, s2
, d1
= 0, d2
= 0;
624 unsigned long back_max
;
625 #define CFQ_RQ1_WRAP 0x01 /* request 1 wraps */
626 #define CFQ_RQ2_WRAP 0x02 /* request 2 wraps */
627 unsigned wrap
= 0; /* bit mask: requests behind the disk head? */
629 if (rq1
== NULL
|| rq1
== rq2
)
634 if (rq_is_sync(rq1
) && !rq_is_sync(rq2
))
636 else if (rq_is_sync(rq2
) && !rq_is_sync(rq1
))
638 if (rq_is_meta(rq1
) && !rq_is_meta(rq2
))
640 else if (rq_is_meta(rq2
) && !rq_is_meta(rq1
))
643 s1
= blk_rq_pos(rq1
);
644 s2
= blk_rq_pos(rq2
);
647 * by definition, 1KiB is 2 sectors
649 back_max
= cfqd
->cfq_back_max
* 2;
652 * Strict one way elevator _except_ in the case where we allow
653 * short backward seeks which are biased as twice the cost of a
654 * similar forward seek.
658 else if (s1
+ back_max
>= last
)
659 d1
= (last
- s1
) * cfqd
->cfq_back_penalty
;
661 wrap
|= CFQ_RQ1_WRAP
;
665 else if (s2
+ back_max
>= last
)
666 d2
= (last
- s2
) * cfqd
->cfq_back_penalty
;
668 wrap
|= CFQ_RQ2_WRAP
;
670 /* Found required data */
673 * By doing switch() on the bit mask "wrap" we avoid having to
674 * check two variables for all permutations: --> faster!
677 case 0: /* common case for CFQ: rq1 and rq2 not wrapped */
693 case (CFQ_RQ1_WRAP
|CFQ_RQ2_WRAP
): /* both rqs wrapped */
696 * Since both rqs are wrapped,
697 * start with the one that's further behind head
698 * (--> only *one* back seek required),
699 * since back seek takes more time than forward.
709 * The below is leftmost cache rbtree addon
711 static struct cfq_queue
*cfq_rb_first(struct cfq_rb_root
*root
)
713 /* Service tree is empty */
718 root
->left
= rb_first(&root
->rb
);
721 return rb_entry(root
->left
, struct cfq_queue
, rb_node
);
726 static struct cfq_group
*cfq_rb_first_group(struct cfq_rb_root
*root
)
729 root
->left
= rb_first(&root
->rb
);
732 return rb_entry_cfqg(root
->left
);
737 static void rb_erase_init(struct rb_node
*n
, struct rb_root
*root
)
743 static void cfq_rb_erase(struct rb_node
*n
, struct cfq_rb_root
*root
)
747 rb_erase_init(n
, &root
->rb
);
752 * would be nice to take fifo expire time into account as well
754 static struct request
*
755 cfq_find_next_rq(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
,
756 struct request
*last
)
758 struct rb_node
*rbnext
= rb_next(&last
->rb_node
);
759 struct rb_node
*rbprev
= rb_prev(&last
->rb_node
);
760 struct request
*next
= NULL
, *prev
= NULL
;
762 BUG_ON(RB_EMPTY_NODE(&last
->rb_node
));
765 prev
= rb_entry_rq(rbprev
);
768 next
= rb_entry_rq(rbnext
);
770 rbnext
= rb_first(&cfqq
->sort_list
);
771 if (rbnext
&& rbnext
!= &last
->rb_node
)
772 next
= rb_entry_rq(rbnext
);
775 return cfq_choose_req(cfqd
, next
, prev
, blk_rq_pos(last
));
778 static unsigned long cfq_slice_offset(struct cfq_data
*cfqd
,
779 struct cfq_queue
*cfqq
)
782 * just an approximation, should be ok.
784 return (cfqq
->cfqg
->nr_cfqq
- 1) * (cfq_prio_slice(cfqd
, 1, 0) -
785 cfq_prio_slice(cfqd
, cfq_cfqq_sync(cfqq
), cfqq
->ioprio
));
789 cfqg_key(struct cfq_rb_root
*st
, struct cfq_group
*cfqg
)
791 return cfqg
->vdisktime
- st
->min_vdisktime
;
795 __cfq_group_service_tree_add(struct cfq_rb_root
*st
, struct cfq_group
*cfqg
)
797 struct rb_node
**node
= &st
->rb
.rb_node
;
798 struct rb_node
*parent
= NULL
;
799 struct cfq_group
*__cfqg
;
800 s64 key
= cfqg_key(st
, cfqg
);
803 while (*node
!= NULL
) {
805 __cfqg
= rb_entry_cfqg(parent
);
807 if (key
< cfqg_key(st
, __cfqg
))
808 node
= &parent
->rb_left
;
810 node
= &parent
->rb_right
;
816 st
->left
= &cfqg
->rb_node
;
818 rb_link_node(&cfqg
->rb_node
, parent
, node
);
819 rb_insert_color(&cfqg
->rb_node
, &st
->rb
);
823 cfq_group_service_tree_add(struct cfq_data
*cfqd
, struct cfq_group
*cfqg
)
825 struct cfq_rb_root
*st
= &cfqd
->grp_service_tree
;
826 struct cfq_group
*__cfqg
;
834 * Currently put the group at the end. Later implement something
835 * so that groups get lesser vtime based on their weights, so that
836 * if group does not loose all if it was not continously backlogged.
838 n
= rb_last(&st
->rb
);
840 __cfqg
= rb_entry_cfqg(n
);
841 cfqg
->vdisktime
= __cfqg
->vdisktime
+ CFQ_IDLE_DELAY
;
843 cfqg
->vdisktime
= st
->min_vdisktime
;
845 __cfq_group_service_tree_add(st
, cfqg
);
848 st
->total_weight
+= cfqg
->weight
;
852 cfq_group_service_tree_del(struct cfq_data
*cfqd
, struct cfq_group
*cfqg
)
854 struct cfq_rb_root
*st
= &cfqd
->grp_service_tree
;
856 if (st
->active
== &cfqg
->rb_node
)
859 BUG_ON(cfqg
->nr_cfqq
< 1);
862 /* If there are other cfq queues under this group, don't delete it */
866 cfq_log_cfqg(cfqd
, cfqg
, "del_from_rr group");
869 st
->total_weight
-= cfqg
->weight
;
870 if (!RB_EMPTY_NODE(&cfqg
->rb_node
))
871 cfq_rb_erase(&cfqg
->rb_node
, st
);
872 cfqg
->saved_workload_slice
= 0;
873 blkiocg_update_blkio_group_dequeue_stats(&cfqg
->blkg
, 1);
876 static inline unsigned int cfq_cfqq_slice_usage(struct cfq_queue
*cfqq
)
878 unsigned int slice_used
;
881 * Queue got expired before even a single request completed or
882 * got expired immediately after first request completion.
884 if (!cfqq
->slice_start
|| cfqq
->slice_start
== jiffies
) {
886 * Also charge the seek time incurred to the group, otherwise
887 * if there are mutiple queues in the group, each can dispatch
888 * a single request on seeky media and cause lots of seek time
889 * and group will never know it.
891 slice_used
= max_t(unsigned, (jiffies
- cfqq
->dispatch_start
),
894 slice_used
= jiffies
- cfqq
->slice_start
;
895 if (slice_used
> cfqq
->allocated_slice
)
896 slice_used
= cfqq
->allocated_slice
;
899 cfq_log_cfqq(cfqq
->cfqd
, cfqq
, "sl_used=%u sect=%lu", slice_used
,
904 static void cfq_group_served(struct cfq_data
*cfqd
, struct cfq_group
*cfqg
,
905 struct cfq_queue
*cfqq
)
907 struct cfq_rb_root
*st
= &cfqd
->grp_service_tree
;
908 unsigned int used_sl
, charge_sl
;
909 int nr_sync
= cfqg
->nr_cfqq
- cfqg_busy_async_queues(cfqd
, cfqg
)
910 - cfqg
->service_tree_idle
.count
;
913 used_sl
= charge_sl
= cfq_cfqq_slice_usage(cfqq
);
915 if (!cfq_cfqq_sync(cfqq
) && !nr_sync
)
916 charge_sl
= cfqq
->allocated_slice
;
918 /* Can't update vdisktime while group is on service tree */
919 cfq_rb_erase(&cfqg
->rb_node
, st
);
920 cfqg
->vdisktime
+= cfq_scale_slice(charge_sl
, cfqg
);
921 __cfq_group_service_tree_add(st
, cfqg
);
923 /* This group is being expired. Save the context */
924 if (time_after(cfqd
->workload_expires
, jiffies
)) {
925 cfqg
->saved_workload_slice
= cfqd
->workload_expires
927 cfqg
->saved_workload
= cfqd
->serving_type
;
928 cfqg
->saved_serving_prio
= cfqd
->serving_prio
;
930 cfqg
->saved_workload_slice
= 0;
932 cfq_log_cfqg(cfqd
, cfqg
, "served: vt=%llu min_vt=%llu", cfqg
->vdisktime
,
934 blkiocg_update_blkio_group_stats(&cfqg
->blkg
, used_sl
,
938 #ifdef CONFIG_CFQ_GROUP_IOSCHED
939 static inline struct cfq_group
*cfqg_of_blkg(struct blkio_group
*blkg
)
942 return container_of(blkg
, struct cfq_group
, blkg
);
947 cfq_update_blkio_group_weight(struct blkio_group
*blkg
, unsigned int weight
)
949 cfqg_of_blkg(blkg
)->weight
= weight
;
952 static struct cfq_group
*
953 cfq_find_alloc_cfqg(struct cfq_data
*cfqd
, struct cgroup
*cgroup
, int create
)
955 struct blkio_cgroup
*blkcg
= cgroup_to_blkio_cgroup(cgroup
);
956 struct cfq_group
*cfqg
= NULL
;
959 struct cfq_rb_root
*st
;
960 struct backing_dev_info
*bdi
= &cfqd
->queue
->backing_dev_info
;
961 unsigned int major
, minor
;
963 /* Do we need to take this reference */
964 if (!css_tryget(&blkcg
->css
))
967 cfqg
= cfqg_of_blkg(blkiocg_lookup_group(blkcg
, key
));
971 cfqg
= kzalloc_node(sizeof(*cfqg
), GFP_ATOMIC
, cfqd
->queue
->node
);
975 cfqg
->weight
= blkcg
->weight
;
976 for_each_cfqg_st(cfqg
, i
, j
, st
)
978 RB_CLEAR_NODE(&cfqg
->rb_node
);
981 * Take the initial reference that will be released on destroy
982 * This can be thought of a joint reference by cgroup and
983 * elevator which will be dropped by either elevator exit
984 * or cgroup deletion path depending on who is exiting first.
986 atomic_set(&cfqg
->ref
, 1);
988 /* Add group onto cgroup list */
989 sscanf(dev_name(bdi
->dev
), "%u:%u", &major
, &minor
);
990 blkiocg_add_blkio_group(blkcg
, &cfqg
->blkg
, (void *)cfqd
,
991 MKDEV(major
, minor
));
993 /* Add group on cfqd list */
994 hlist_add_head(&cfqg
->cfqd_node
, &cfqd
->cfqg_list
);
997 css_put(&blkcg
->css
);
1002 * Search for the cfq group current task belongs to. If create = 1, then also
1003 * create the cfq group if it does not exist. request_queue lock must be held.
1005 static struct cfq_group
*cfq_get_cfqg(struct cfq_data
*cfqd
, int create
)
1007 struct cgroup
*cgroup
;
1008 struct cfq_group
*cfqg
= NULL
;
1011 cgroup
= task_cgroup(current
, blkio_subsys_id
);
1012 cfqg
= cfq_find_alloc_cfqg(cfqd
, cgroup
, create
);
1013 if (!cfqg
&& create
)
1014 cfqg
= &cfqd
->root_group
;
1019 static void cfq_link_cfqq_cfqg(struct cfq_queue
*cfqq
, struct cfq_group
*cfqg
)
1021 /* Currently, all async queues are mapped to root group */
1022 if (!cfq_cfqq_sync(cfqq
))
1023 cfqg
= &cfqq
->cfqd
->root_group
;
1026 /* cfqq reference on cfqg */
1027 atomic_inc(&cfqq
->cfqg
->ref
);
1030 static void cfq_put_cfqg(struct cfq_group
*cfqg
)
1032 struct cfq_rb_root
*st
;
1035 BUG_ON(atomic_read(&cfqg
->ref
) <= 0);
1036 if (!atomic_dec_and_test(&cfqg
->ref
))
1038 for_each_cfqg_st(cfqg
, i
, j
, st
)
1039 BUG_ON(!RB_EMPTY_ROOT(&st
->rb
) || st
->active
!= NULL
);
1043 static void cfq_destroy_cfqg(struct cfq_data
*cfqd
, struct cfq_group
*cfqg
)
1045 /* Something wrong if we are trying to remove same group twice */
1046 BUG_ON(hlist_unhashed(&cfqg
->cfqd_node
));
1048 hlist_del_init(&cfqg
->cfqd_node
);
1051 * Put the reference taken at the time of creation so that when all
1052 * queues are gone, group can be destroyed.
1057 static void cfq_release_cfq_groups(struct cfq_data
*cfqd
)
1059 struct hlist_node
*pos
, *n
;
1060 struct cfq_group
*cfqg
;
1062 hlist_for_each_entry_safe(cfqg
, pos
, n
, &cfqd
->cfqg_list
, cfqd_node
) {
1064 * If cgroup removal path got to blk_group first and removed
1065 * it from cgroup list, then it will take care of destroying
1068 if (!blkiocg_del_blkio_group(&cfqg
->blkg
))
1069 cfq_destroy_cfqg(cfqd
, cfqg
);
1074 * Blk cgroup controller notification saying that blkio_group object is being
1075 * delinked as associated cgroup object is going away. That also means that
1076 * no new IO will come in this group. So get rid of this group as soon as
1077 * any pending IO in the group is finished.
1079 * This function is called under rcu_read_lock(). key is the rcu protected
1080 * pointer. That means "key" is a valid cfq_data pointer as long as we are rcu
1083 * "key" was fetched from blkio_group under blkio_cgroup->lock. That means
1084 * it should not be NULL as even if elevator was exiting, cgroup deltion
1085 * path got to it first.
1087 void cfq_unlink_blkio_group(void *key
, struct blkio_group
*blkg
)
1089 unsigned long flags
;
1090 struct cfq_data
*cfqd
= key
;
1092 spin_lock_irqsave(cfqd
->queue
->queue_lock
, flags
);
1093 cfq_destroy_cfqg(cfqd
, cfqg_of_blkg(blkg
));
1094 spin_unlock_irqrestore(cfqd
->queue
->queue_lock
, flags
);
1097 #else /* GROUP_IOSCHED */
1098 static struct cfq_group
*cfq_get_cfqg(struct cfq_data
*cfqd
, int create
)
1100 return &cfqd
->root_group
;
1103 cfq_link_cfqq_cfqg(struct cfq_queue
*cfqq
, struct cfq_group
*cfqg
) {
1107 static void cfq_release_cfq_groups(struct cfq_data
*cfqd
) {}
1108 static inline void cfq_put_cfqg(struct cfq_group
*cfqg
) {}
1110 #endif /* GROUP_IOSCHED */
1113 * The cfqd->service_trees holds all pending cfq_queue's that have
1114 * requests waiting to be processed. It is sorted in the order that
1115 * we will service the queues.
1117 static void cfq_service_tree_add(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
,
1120 struct rb_node
**p
, *parent
;
1121 struct cfq_queue
*__cfqq
;
1122 unsigned long rb_key
;
1123 struct cfq_rb_root
*service_tree
;
1126 int group_changed
= 0;
1128 #ifdef CONFIG_CFQ_GROUP_IOSCHED
1129 if (!cfqd
->cfq_group_isolation
1130 && cfqq_type(cfqq
) == SYNC_NOIDLE_WORKLOAD
1131 && cfqq
->cfqg
&& cfqq
->cfqg
!= &cfqd
->root_group
) {
1132 /* Move this cfq to root group */
1133 cfq_log_cfqq(cfqd
, cfqq
, "moving to root group");
1134 if (!RB_EMPTY_NODE(&cfqq
->rb_node
))
1135 cfq_group_service_tree_del(cfqd
, cfqq
->cfqg
);
1136 cfqq
->orig_cfqg
= cfqq
->cfqg
;
1137 cfqq
->cfqg
= &cfqd
->root_group
;
1138 atomic_inc(&cfqd
->root_group
.ref
);
1140 } else if (!cfqd
->cfq_group_isolation
1141 && cfqq_type(cfqq
) == SYNC_WORKLOAD
&& cfqq
->orig_cfqg
) {
1142 /* cfqq is sequential now needs to go to its original group */
1143 BUG_ON(cfqq
->cfqg
!= &cfqd
->root_group
);
1144 if (!RB_EMPTY_NODE(&cfqq
->rb_node
))
1145 cfq_group_service_tree_del(cfqd
, cfqq
->cfqg
);
1146 cfq_put_cfqg(cfqq
->cfqg
);
1147 cfqq
->cfqg
= cfqq
->orig_cfqg
;
1148 cfqq
->orig_cfqg
= NULL
;
1150 cfq_log_cfqq(cfqd
, cfqq
, "moved to origin group");
1154 service_tree
= service_tree_for(cfqq
->cfqg
, cfqq_prio(cfqq
),
1155 cfqq_type(cfqq
), cfqd
);
1156 if (cfq_class_idle(cfqq
)) {
1157 rb_key
= CFQ_IDLE_DELAY
;
1158 parent
= rb_last(&service_tree
->rb
);
1159 if (parent
&& parent
!= &cfqq
->rb_node
) {
1160 __cfqq
= rb_entry(parent
, struct cfq_queue
, rb_node
);
1161 rb_key
+= __cfqq
->rb_key
;
1164 } else if (!add_front
) {
1166 * Get our rb key offset. Subtract any residual slice
1167 * value carried from last service. A negative resid
1168 * count indicates slice overrun, and this should position
1169 * the next service time further away in the tree.
1171 rb_key
= cfq_slice_offset(cfqd
, cfqq
) + jiffies
;
1172 rb_key
-= cfqq
->slice_resid
;
1173 cfqq
->slice_resid
= 0;
1176 __cfqq
= cfq_rb_first(service_tree
);
1177 rb_key
+= __cfqq
? __cfqq
->rb_key
: jiffies
;
1180 if (!RB_EMPTY_NODE(&cfqq
->rb_node
)) {
1183 * same position, nothing more to do
1185 if (rb_key
== cfqq
->rb_key
&&
1186 cfqq
->service_tree
== service_tree
)
1189 cfq_rb_erase(&cfqq
->rb_node
, cfqq
->service_tree
);
1190 cfqq
->service_tree
= NULL
;
1195 cfqq
->service_tree
= service_tree
;
1196 p
= &service_tree
->rb
.rb_node
;
1201 __cfqq
= rb_entry(parent
, struct cfq_queue
, rb_node
);
1204 * sort by key, that represents service time.
1206 if (time_before(rb_key
, __cfqq
->rb_key
))
1209 n
= &(*p
)->rb_right
;
1217 service_tree
->left
= &cfqq
->rb_node
;
1219 cfqq
->rb_key
= rb_key
;
1220 rb_link_node(&cfqq
->rb_node
, parent
, p
);
1221 rb_insert_color(&cfqq
->rb_node
, &service_tree
->rb
);
1222 service_tree
->count
++;
1223 if ((add_front
|| !new_cfqq
) && !group_changed
)
1225 cfq_group_service_tree_add(cfqd
, cfqq
->cfqg
);
1228 static struct cfq_queue
*
1229 cfq_prio_tree_lookup(struct cfq_data
*cfqd
, struct rb_root
*root
,
1230 sector_t sector
, struct rb_node
**ret_parent
,
1231 struct rb_node
***rb_link
)
1233 struct rb_node
**p
, *parent
;
1234 struct cfq_queue
*cfqq
= NULL
;
1242 cfqq
= rb_entry(parent
, struct cfq_queue
, p_node
);
1245 * Sort strictly based on sector. Smallest to the left,
1246 * largest to the right.
1248 if (sector
> blk_rq_pos(cfqq
->next_rq
))
1249 n
= &(*p
)->rb_right
;
1250 else if (sector
< blk_rq_pos(cfqq
->next_rq
))
1258 *ret_parent
= parent
;
1264 static void cfq_prio_tree_add(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
)
1266 struct rb_node
**p
, *parent
;
1267 struct cfq_queue
*__cfqq
;
1270 rb_erase(&cfqq
->p_node
, cfqq
->p_root
);
1271 cfqq
->p_root
= NULL
;
1274 if (cfq_class_idle(cfqq
))
1279 cfqq
->p_root
= &cfqd
->prio_trees
[cfqq
->org_ioprio
];
1280 __cfqq
= cfq_prio_tree_lookup(cfqd
, cfqq
->p_root
,
1281 blk_rq_pos(cfqq
->next_rq
), &parent
, &p
);
1283 rb_link_node(&cfqq
->p_node
, parent
, p
);
1284 rb_insert_color(&cfqq
->p_node
, cfqq
->p_root
);
1286 cfqq
->p_root
= NULL
;
1290 * Update cfqq's position in the service tree.
1292 static void cfq_resort_rr_list(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
)
1295 * Resorting requires the cfqq to be on the RR list already.
1297 if (cfq_cfqq_on_rr(cfqq
)) {
1298 cfq_service_tree_add(cfqd
, cfqq
, 0);
1299 cfq_prio_tree_add(cfqd
, cfqq
);
1304 * add to busy list of queues for service, trying to be fair in ordering
1305 * the pending list according to last request service
1307 static void cfq_add_cfqq_rr(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
)
1309 cfq_log_cfqq(cfqd
, cfqq
, "add_to_rr");
1310 BUG_ON(cfq_cfqq_on_rr(cfqq
));
1311 cfq_mark_cfqq_on_rr(cfqq
);
1312 cfqd
->busy_queues
++;
1314 cfq_resort_rr_list(cfqd
, cfqq
);
1318 * Called when the cfqq no longer has requests pending, remove it from
1321 static void cfq_del_cfqq_rr(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
)
1323 cfq_log_cfqq(cfqd
, cfqq
, "del_from_rr");
1324 BUG_ON(!cfq_cfqq_on_rr(cfqq
));
1325 cfq_clear_cfqq_on_rr(cfqq
);
1327 if (!RB_EMPTY_NODE(&cfqq
->rb_node
)) {
1328 cfq_rb_erase(&cfqq
->rb_node
, cfqq
->service_tree
);
1329 cfqq
->service_tree
= NULL
;
1332 rb_erase(&cfqq
->p_node
, cfqq
->p_root
);
1333 cfqq
->p_root
= NULL
;
1336 cfq_group_service_tree_del(cfqd
, cfqq
->cfqg
);
1337 BUG_ON(!cfqd
->busy_queues
);
1338 cfqd
->busy_queues
--;
1342 * rb tree support functions
1344 static void cfq_del_rq_rb(struct request
*rq
)
1346 struct cfq_queue
*cfqq
= RQ_CFQQ(rq
);
1347 const int sync
= rq_is_sync(rq
);
1349 BUG_ON(!cfqq
->queued
[sync
]);
1350 cfqq
->queued
[sync
]--;
1352 elv_rb_del(&cfqq
->sort_list
, rq
);
1354 if (cfq_cfqq_on_rr(cfqq
) && RB_EMPTY_ROOT(&cfqq
->sort_list
)) {
1356 * Queue will be deleted from service tree when we actually
1357 * expire it later. Right now just remove it from prio tree
1361 rb_erase(&cfqq
->p_node
, cfqq
->p_root
);
1362 cfqq
->p_root
= NULL
;
1367 static void cfq_add_rq_rb(struct request
*rq
)
1369 struct cfq_queue
*cfqq
= RQ_CFQQ(rq
);
1370 struct cfq_data
*cfqd
= cfqq
->cfqd
;
1371 struct request
*__alias
, *prev
;
1373 cfqq
->queued
[rq_is_sync(rq
)]++;
1376 * looks a little odd, but the first insert might return an alias.
1377 * if that happens, put the alias on the dispatch list
1379 while ((__alias
= elv_rb_add(&cfqq
->sort_list
, rq
)) != NULL
)
1380 cfq_dispatch_insert(cfqd
->queue
, __alias
);
1382 if (!cfq_cfqq_on_rr(cfqq
))
1383 cfq_add_cfqq_rr(cfqd
, cfqq
);
1386 * check if this request is a better next-serve candidate
1388 prev
= cfqq
->next_rq
;
1389 cfqq
->next_rq
= cfq_choose_req(cfqd
, cfqq
->next_rq
, rq
, cfqd
->last_position
);
1392 * adjust priority tree position, if ->next_rq changes
1394 if (prev
!= cfqq
->next_rq
)
1395 cfq_prio_tree_add(cfqd
, cfqq
);
1397 BUG_ON(!cfqq
->next_rq
);
1400 static void cfq_reposition_rq_rb(struct cfq_queue
*cfqq
, struct request
*rq
)
1402 elv_rb_del(&cfqq
->sort_list
, rq
);
1403 cfqq
->queued
[rq_is_sync(rq
)]--;
1407 static struct request
*
1408 cfq_find_rq_fmerge(struct cfq_data
*cfqd
, struct bio
*bio
)
1410 struct task_struct
*tsk
= current
;
1411 struct cfq_io_context
*cic
;
1412 struct cfq_queue
*cfqq
;
1414 cic
= cfq_cic_lookup(cfqd
, tsk
->io_context
);
1418 cfqq
= cic_to_cfqq(cic
, cfq_bio_sync(bio
));
1420 sector_t sector
= bio
->bi_sector
+ bio_sectors(bio
);
1422 return elv_rb_find(&cfqq
->sort_list
, sector
);
1428 static void cfq_activate_request(struct request_queue
*q
, struct request
*rq
)
1430 struct cfq_data
*cfqd
= q
->elevator
->elevator_data
;
1432 cfqd
->rq_in_driver
[rq_is_sync(rq
)]++;
1433 cfq_log_cfqq(cfqd
, RQ_CFQQ(rq
), "activate rq, drv=%d",
1434 rq_in_driver(cfqd
));
1436 cfqd
->last_position
= blk_rq_pos(rq
) + blk_rq_sectors(rq
);
1439 static void cfq_deactivate_request(struct request_queue
*q
, struct request
*rq
)
1441 struct cfq_data
*cfqd
= q
->elevator
->elevator_data
;
1442 const int sync
= rq_is_sync(rq
);
1444 WARN_ON(!cfqd
->rq_in_driver
[sync
]);
1445 cfqd
->rq_in_driver
[sync
]--;
1446 cfq_log_cfqq(cfqd
, RQ_CFQQ(rq
), "deactivate rq, drv=%d",
1447 rq_in_driver(cfqd
));
1450 static void cfq_remove_request(struct request
*rq
)
1452 struct cfq_queue
*cfqq
= RQ_CFQQ(rq
);
1454 if (cfqq
->next_rq
== rq
)
1455 cfqq
->next_rq
= cfq_find_next_rq(cfqq
->cfqd
, cfqq
, rq
);
1457 list_del_init(&rq
->queuelist
);
1460 cfqq
->cfqd
->rq_queued
--;
1461 if (rq_is_meta(rq
)) {
1462 WARN_ON(!cfqq
->meta_pending
);
1463 cfqq
->meta_pending
--;
1467 static int cfq_merge(struct request_queue
*q
, struct request
**req
,
1470 struct cfq_data
*cfqd
= q
->elevator
->elevator_data
;
1471 struct request
*__rq
;
1473 __rq
= cfq_find_rq_fmerge(cfqd
, bio
);
1474 if (__rq
&& elv_rq_merge_ok(__rq
, bio
)) {
1476 return ELEVATOR_FRONT_MERGE
;
1479 return ELEVATOR_NO_MERGE
;
1482 static void cfq_merged_request(struct request_queue
*q
, struct request
*req
,
1485 if (type
== ELEVATOR_FRONT_MERGE
) {
1486 struct cfq_queue
*cfqq
= RQ_CFQQ(req
);
1488 cfq_reposition_rq_rb(cfqq
, req
);
1493 cfq_merged_requests(struct request_queue
*q
, struct request
*rq
,
1494 struct request
*next
)
1496 struct cfq_queue
*cfqq
= RQ_CFQQ(rq
);
1498 * reposition in fifo if next is older than rq
1500 if (!list_empty(&rq
->queuelist
) && !list_empty(&next
->queuelist
) &&
1501 time_before(rq_fifo_time(next
), rq_fifo_time(rq
))) {
1502 list_move(&rq
->queuelist
, &next
->queuelist
);
1503 rq_set_fifo_time(rq
, rq_fifo_time(next
));
1506 if (cfqq
->next_rq
== next
)
1508 cfq_remove_request(next
);
1511 static int cfq_allow_merge(struct request_queue
*q
, struct request
*rq
,
1514 struct cfq_data
*cfqd
= q
->elevator
->elevator_data
;
1515 struct cfq_io_context
*cic
;
1516 struct cfq_queue
*cfqq
;
1518 /* Deny merge if bio and rq don't belong to same cfq group */
1519 if ((RQ_CFQQ(rq
))->cfqg
!= cfq_get_cfqg(cfqd
, 0))
1522 * Disallow merge of a sync bio into an async request.
1524 if (cfq_bio_sync(bio
) && !rq_is_sync(rq
))
1528 * Lookup the cfqq that this bio will be queued with. Allow
1529 * merge only if rq is queued there.
1531 cic
= cfq_cic_lookup(cfqd
, current
->io_context
);
1535 cfqq
= cic_to_cfqq(cic
, cfq_bio_sync(bio
));
1536 return cfqq
== RQ_CFQQ(rq
);
1539 static void __cfq_set_active_queue(struct cfq_data
*cfqd
,
1540 struct cfq_queue
*cfqq
)
1543 cfq_log_cfqq(cfqd
, cfqq
, "set_active");
1544 cfqq
->slice_start
= 0;
1545 cfqq
->dispatch_start
= jiffies
;
1546 cfqq
->allocated_slice
= 0;
1547 cfqq
->slice_end
= 0;
1548 cfqq
->slice_dispatch
= 0;
1549 cfqq
->nr_sectors
= 0;
1551 cfq_clear_cfqq_wait_request(cfqq
);
1552 cfq_clear_cfqq_must_dispatch(cfqq
);
1553 cfq_clear_cfqq_must_alloc_slice(cfqq
);
1554 cfq_clear_cfqq_fifo_expire(cfqq
);
1555 cfq_mark_cfqq_slice_new(cfqq
);
1557 del_timer(&cfqd
->idle_slice_timer
);
1560 cfqd
->active_queue
= cfqq
;
1564 * current cfqq expired its slice (or was too idle), select new one
1567 __cfq_slice_expired(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
,
1570 cfq_log_cfqq(cfqd
, cfqq
, "slice expired t=%d", timed_out
);
1572 if (cfq_cfqq_wait_request(cfqq
))
1573 del_timer(&cfqd
->idle_slice_timer
);
1575 cfq_clear_cfqq_wait_request(cfqq
);
1576 cfq_clear_cfqq_wait_busy(cfqq
);
1577 cfq_clear_cfqq_wait_busy_done(cfqq
);
1580 * store what was left of this slice, if the queue idled/timed out
1582 if (timed_out
&& !cfq_cfqq_slice_new(cfqq
)) {
1583 cfqq
->slice_resid
= cfqq
->slice_end
- jiffies
;
1584 cfq_log_cfqq(cfqd
, cfqq
, "resid=%ld", cfqq
->slice_resid
);
1587 cfq_group_served(cfqd
, cfqq
->cfqg
, cfqq
);
1589 if (cfq_cfqq_on_rr(cfqq
) && RB_EMPTY_ROOT(&cfqq
->sort_list
))
1590 cfq_del_cfqq_rr(cfqd
, cfqq
);
1592 cfq_resort_rr_list(cfqd
, cfqq
);
1594 if (cfqq
== cfqd
->active_queue
)
1595 cfqd
->active_queue
= NULL
;
1597 if (&cfqq
->cfqg
->rb_node
== cfqd
->grp_service_tree
.active
)
1598 cfqd
->grp_service_tree
.active
= NULL
;
1600 if (cfqd
->active_cic
) {
1601 put_io_context(cfqd
->active_cic
->ioc
);
1602 cfqd
->active_cic
= NULL
;
1606 static inline void cfq_slice_expired(struct cfq_data
*cfqd
, bool timed_out
)
1608 struct cfq_queue
*cfqq
= cfqd
->active_queue
;
1611 __cfq_slice_expired(cfqd
, cfqq
, timed_out
);
1615 * Get next queue for service. Unless we have a queue preemption,
1616 * we'll simply select the first cfqq in the service tree.
1618 static struct cfq_queue
*cfq_get_next_queue(struct cfq_data
*cfqd
)
1620 struct cfq_rb_root
*service_tree
=
1621 service_tree_for(cfqd
->serving_group
, cfqd
->serving_prio
,
1622 cfqd
->serving_type
, cfqd
);
1624 if (!cfqd
->rq_queued
)
1627 /* There is nothing to dispatch */
1630 if (RB_EMPTY_ROOT(&service_tree
->rb
))
1632 return cfq_rb_first(service_tree
);
1635 static struct cfq_queue
*cfq_get_next_queue_forced(struct cfq_data
*cfqd
)
1637 struct cfq_group
*cfqg
;
1638 struct cfq_queue
*cfqq
;
1640 struct cfq_rb_root
*st
;
1642 if (!cfqd
->rq_queued
)
1645 cfqg
= cfq_get_next_cfqg(cfqd
);
1649 for_each_cfqg_st(cfqg
, i
, j
, st
)
1650 if ((cfqq
= cfq_rb_first(st
)) != NULL
)
1656 * Get and set a new active queue for service.
1658 static struct cfq_queue
*cfq_set_active_queue(struct cfq_data
*cfqd
,
1659 struct cfq_queue
*cfqq
)
1662 cfqq
= cfq_get_next_queue(cfqd
);
1664 __cfq_set_active_queue(cfqd
, cfqq
);
1668 static inline sector_t
cfq_dist_from_last(struct cfq_data
*cfqd
,
1671 if (blk_rq_pos(rq
) >= cfqd
->last_position
)
1672 return blk_rq_pos(rq
) - cfqd
->last_position
;
1674 return cfqd
->last_position
- blk_rq_pos(rq
);
1677 #define CFQQ_SEEK_THR 8 * 1024
1678 #define CFQQ_SEEKY(cfqq) ((cfqq)->seek_mean > CFQQ_SEEK_THR)
1680 static inline int cfq_rq_close(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
,
1683 sector_t sdist
= cfqq
->seek_mean
;
1685 if (!sample_valid(cfqq
->seek_samples
))
1686 sdist
= CFQQ_SEEK_THR
;
1688 return cfq_dist_from_last(cfqd
, rq
) <= sdist
;
1691 static struct cfq_queue
*cfqq_close(struct cfq_data
*cfqd
,
1692 struct cfq_queue
*cur_cfqq
)
1694 struct rb_root
*root
= &cfqd
->prio_trees
[cur_cfqq
->org_ioprio
];
1695 struct rb_node
*parent
, *node
;
1696 struct cfq_queue
*__cfqq
;
1697 sector_t sector
= cfqd
->last_position
;
1699 if (RB_EMPTY_ROOT(root
))
1703 * First, if we find a request starting at the end of the last
1704 * request, choose it.
1706 __cfqq
= cfq_prio_tree_lookup(cfqd
, root
, sector
, &parent
, NULL
);
1711 * If the exact sector wasn't found, the parent of the NULL leaf
1712 * will contain the closest sector.
1714 __cfqq
= rb_entry(parent
, struct cfq_queue
, p_node
);
1715 if (cfq_rq_close(cfqd
, cur_cfqq
, __cfqq
->next_rq
))
1718 if (blk_rq_pos(__cfqq
->next_rq
) < sector
)
1719 node
= rb_next(&__cfqq
->p_node
);
1721 node
= rb_prev(&__cfqq
->p_node
);
1725 __cfqq
= rb_entry(node
, struct cfq_queue
, p_node
);
1726 if (cfq_rq_close(cfqd
, cur_cfqq
, __cfqq
->next_rq
))
1734 * cur_cfqq - passed in so that we don't decide that the current queue is
1735 * closely cooperating with itself.
1737 * So, basically we're assuming that that cur_cfqq has dispatched at least
1738 * one request, and that cfqd->last_position reflects a position on the disk
1739 * associated with the I/O issued by cur_cfqq. I'm not sure this is a valid
1742 static struct cfq_queue
*cfq_close_cooperator(struct cfq_data
*cfqd
,
1743 struct cfq_queue
*cur_cfqq
)
1745 struct cfq_queue
*cfqq
;
1747 if (!cfq_cfqq_sync(cur_cfqq
))
1749 if (CFQQ_SEEKY(cur_cfqq
))
1753 * We should notice if some of the queues are cooperating, eg
1754 * working closely on the same area of the disk. In that case,
1755 * we can group them together and don't waste time idling.
1757 cfqq
= cfqq_close(cfqd
, cur_cfqq
);
1761 /* If new queue belongs to different cfq_group, don't choose it */
1762 if (cur_cfqq
->cfqg
!= cfqq
->cfqg
)
1766 * It only makes sense to merge sync queues.
1768 if (!cfq_cfqq_sync(cfqq
))
1770 if (CFQQ_SEEKY(cfqq
))
1774 * Do not merge queues of different priority classes
1776 if (cfq_class_rt(cfqq
) != cfq_class_rt(cur_cfqq
))
1783 * Determine whether we should enforce idle window for this queue.
1786 static bool cfq_should_idle(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
)
1788 enum wl_prio_t prio
= cfqq_prio(cfqq
);
1789 struct cfq_rb_root
*service_tree
= cfqq
->service_tree
;
1791 BUG_ON(!service_tree
);
1792 BUG_ON(!service_tree
->count
);
1794 /* We never do for idle class queues. */
1795 if (prio
== IDLE_WORKLOAD
)
1798 /* We do for queues that were marked with idle window flag. */
1799 if (cfq_cfqq_idle_window(cfqq
) &&
1800 !(blk_queue_nonrot(cfqd
->queue
) && cfqd
->hw_tag
))
1804 * Otherwise, we do only if they are the last ones
1805 * in their service tree.
1807 return service_tree
->count
== 1;
1810 static void cfq_arm_slice_timer(struct cfq_data
*cfqd
)
1812 struct cfq_queue
*cfqq
= cfqd
->active_queue
;
1813 struct cfq_io_context
*cic
;
1817 * SSD device without seek penalty, disable idling. But only do so
1818 * for devices that support queuing, otherwise we still have a problem
1819 * with sync vs async workloads.
1821 if (blk_queue_nonrot(cfqd
->queue
) && cfqd
->hw_tag
)
1824 WARN_ON(!RB_EMPTY_ROOT(&cfqq
->sort_list
));
1825 WARN_ON(cfq_cfqq_slice_new(cfqq
));
1828 * idle is disabled, either manually or by past process history
1830 if (!cfqd
->cfq_slice_idle
|| !cfq_should_idle(cfqd
, cfqq
))
1834 * still active requests from this queue, don't idle
1836 if (cfqq
->dispatched
)
1840 * task has exited, don't wait
1842 cic
= cfqd
->active_cic
;
1843 if (!cic
|| !atomic_read(&cic
->ioc
->nr_tasks
))
1847 * If our average think time is larger than the remaining time
1848 * slice, then don't idle. This avoids overrunning the allotted
1851 if (sample_valid(cic
->ttime_samples
) &&
1852 (cfqq
->slice_end
- jiffies
< cic
->ttime_mean
))
1855 cfq_mark_cfqq_wait_request(cfqq
);
1857 sl
= cfqd
->cfq_slice_idle
;
1859 mod_timer(&cfqd
->idle_slice_timer
, jiffies
+ sl
);
1860 cfq_log_cfqq(cfqd
, cfqq
, "arm_idle: %lu", sl
);
1864 * Move request from internal lists to the request queue dispatch list.
1866 static void cfq_dispatch_insert(struct request_queue
*q
, struct request
*rq
)
1868 struct cfq_data
*cfqd
= q
->elevator
->elevator_data
;
1869 struct cfq_queue
*cfqq
= RQ_CFQQ(rq
);
1871 cfq_log_cfqq(cfqd
, cfqq
, "dispatch_insert");
1873 cfqq
->next_rq
= cfq_find_next_rq(cfqd
, cfqq
, rq
);
1874 cfq_remove_request(rq
);
1876 elv_dispatch_sort(q
, rq
);
1878 if (cfq_cfqq_sync(cfqq
))
1879 cfqd
->sync_flight
++;
1880 cfqq
->nr_sectors
+= blk_rq_sectors(rq
);
1884 * return expired entry, or NULL to just start from scratch in rbtree
1886 static struct request
*cfq_check_fifo(struct cfq_queue
*cfqq
)
1888 struct request
*rq
= NULL
;
1890 if (cfq_cfqq_fifo_expire(cfqq
))
1893 cfq_mark_cfqq_fifo_expire(cfqq
);
1895 if (list_empty(&cfqq
->fifo
))
1898 rq
= rq_entry_fifo(cfqq
->fifo
.next
);
1899 if (time_before(jiffies
, rq_fifo_time(rq
)))
1902 cfq_log_cfqq(cfqq
->cfqd
, cfqq
, "fifo=%p", rq
);
1907 cfq_prio_to_maxrq(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
)
1909 const int base_rq
= cfqd
->cfq_slice_async_rq
;
1911 WARN_ON(cfqq
->ioprio
>= IOPRIO_BE_NR
);
1913 return 2 * (base_rq
+ base_rq
* (CFQ_PRIO_LISTS
- 1 - cfqq
->ioprio
));
1917 * Must be called with the queue_lock held.
1919 static int cfqq_process_refs(struct cfq_queue
*cfqq
)
1921 int process_refs
, io_refs
;
1923 io_refs
= cfqq
->allocated
[READ
] + cfqq
->allocated
[WRITE
];
1924 process_refs
= atomic_read(&cfqq
->ref
) - io_refs
;
1925 BUG_ON(process_refs
< 0);
1926 return process_refs
;
1929 static void cfq_setup_merge(struct cfq_queue
*cfqq
, struct cfq_queue
*new_cfqq
)
1931 int process_refs
, new_process_refs
;
1932 struct cfq_queue
*__cfqq
;
1934 /* Avoid a circular list and skip interim queue merges */
1935 while ((__cfqq
= new_cfqq
->new_cfqq
)) {
1941 process_refs
= cfqq_process_refs(cfqq
);
1943 * If the process for the cfqq has gone away, there is no
1944 * sense in merging the queues.
1946 if (process_refs
== 0)
1950 * Merge in the direction of the lesser amount of work.
1952 new_process_refs
= cfqq_process_refs(new_cfqq
);
1953 if (new_process_refs
>= process_refs
) {
1954 cfqq
->new_cfqq
= new_cfqq
;
1955 atomic_add(process_refs
, &new_cfqq
->ref
);
1957 new_cfqq
->new_cfqq
= cfqq
;
1958 atomic_add(new_process_refs
, &cfqq
->ref
);
1962 static enum wl_type_t
cfq_choose_wl(struct cfq_data
*cfqd
,
1963 struct cfq_group
*cfqg
, enum wl_prio_t prio
,
1966 struct cfq_queue
*queue
;
1968 bool key_valid
= false;
1969 unsigned long lowest_key
= 0;
1970 enum wl_type_t cur_best
= SYNC_NOIDLE_WORKLOAD
;
1974 * When priorities switched, we prefer starting
1975 * from SYNC_NOIDLE (first choice), or just SYNC
1978 if (service_tree_for(cfqg
, prio
, cur_best
, cfqd
)->count
)
1980 cur_best
= SYNC_WORKLOAD
;
1981 if (service_tree_for(cfqg
, prio
, cur_best
, cfqd
)->count
)
1984 return ASYNC_WORKLOAD
;
1987 for (i
= 0; i
< 3; ++i
) {
1988 /* otherwise, select the one with lowest rb_key */
1989 queue
= cfq_rb_first(service_tree_for(cfqg
, prio
, i
, cfqd
));
1991 (!key_valid
|| time_before(queue
->rb_key
, lowest_key
))) {
1992 lowest_key
= queue
->rb_key
;
2001 static void choose_service_tree(struct cfq_data
*cfqd
, struct cfq_group
*cfqg
)
2003 enum wl_prio_t previous_prio
= cfqd
->serving_prio
;
2007 struct cfq_rb_root
*st
;
2008 unsigned group_slice
;
2011 cfqd
->serving_prio
= IDLE_WORKLOAD
;
2012 cfqd
->workload_expires
= jiffies
+ 1;
2016 /* Choose next priority. RT > BE > IDLE */
2017 if (cfq_group_busy_queues_wl(RT_WORKLOAD
, cfqd
, cfqg
))
2018 cfqd
->serving_prio
= RT_WORKLOAD
;
2019 else if (cfq_group_busy_queues_wl(BE_WORKLOAD
, cfqd
, cfqg
))
2020 cfqd
->serving_prio
= BE_WORKLOAD
;
2022 cfqd
->serving_prio
= IDLE_WORKLOAD
;
2023 cfqd
->workload_expires
= jiffies
+ 1;
2028 * For RT and BE, we have to choose also the type
2029 * (SYNC, SYNC_NOIDLE, ASYNC), and to compute a workload
2032 prio_changed
= (cfqd
->serving_prio
!= previous_prio
);
2033 st
= service_tree_for(cfqg
, cfqd
->serving_prio
, cfqd
->serving_type
,
2038 * If priority didn't change, check workload expiration,
2039 * and that we still have other queues ready
2041 if (!prio_changed
&& count
&&
2042 !time_after(jiffies
, cfqd
->workload_expires
))
2045 /* otherwise select new workload type */
2046 cfqd
->serving_type
=
2047 cfq_choose_wl(cfqd
, cfqg
, cfqd
->serving_prio
, prio_changed
);
2048 st
= service_tree_for(cfqg
, cfqd
->serving_prio
, cfqd
->serving_type
,
2053 * the workload slice is computed as a fraction of target latency
2054 * proportional to the number of queues in that workload, over
2055 * all the queues in the same priority class
2057 group_slice
= cfq_group_slice(cfqd
, cfqg
);
2059 slice
= group_slice
* count
/
2060 max_t(unsigned, cfqg
->busy_queues_avg
[cfqd
->serving_prio
],
2061 cfq_group_busy_queues_wl(cfqd
->serving_prio
, cfqd
, cfqg
));
2063 if (cfqd
->serving_type
== ASYNC_WORKLOAD
) {
2067 * Async queues are currently system wide. Just taking
2068 * proportion of queues with-in same group will lead to higher
2069 * async ratio system wide as generally root group is going
2070 * to have higher weight. A more accurate thing would be to
2071 * calculate system wide asnc/sync ratio.
2073 tmp
= cfq_target_latency
* cfqg_busy_async_queues(cfqd
, cfqg
);
2074 tmp
= tmp
/cfqd
->busy_queues
;
2075 slice
= min_t(unsigned, slice
, tmp
);
2077 /* async workload slice is scaled down according to
2078 * the sync/async slice ratio. */
2079 slice
= slice
* cfqd
->cfq_slice
[0] / cfqd
->cfq_slice
[1];
2081 /* sync workload slice is at least 2 * cfq_slice_idle */
2082 slice
= max(slice
, 2 * cfqd
->cfq_slice_idle
);
2084 slice
= max_t(unsigned, slice
, CFQ_MIN_TT
);
2085 cfqd
->workload_expires
= jiffies
+ slice
;
2086 cfqd
->noidle_tree_requires_idle
= false;
2089 static struct cfq_group
*cfq_get_next_cfqg(struct cfq_data
*cfqd
)
2091 struct cfq_rb_root
*st
= &cfqd
->grp_service_tree
;
2092 struct cfq_group
*cfqg
;
2094 if (RB_EMPTY_ROOT(&st
->rb
))
2096 cfqg
= cfq_rb_first_group(st
);
2097 st
->active
= &cfqg
->rb_node
;
2098 update_min_vdisktime(st
);
2102 static void cfq_choose_cfqg(struct cfq_data
*cfqd
)
2104 struct cfq_group
*cfqg
= cfq_get_next_cfqg(cfqd
);
2106 cfqd
->serving_group
= cfqg
;
2108 /* Restore the workload type data */
2109 if (cfqg
->saved_workload_slice
) {
2110 cfqd
->workload_expires
= jiffies
+ cfqg
->saved_workload_slice
;
2111 cfqd
->serving_type
= cfqg
->saved_workload
;
2112 cfqd
->serving_prio
= cfqg
->saved_serving_prio
;
2114 choose_service_tree(cfqd
, cfqg
);
2118 * Select a queue for service. If we have a current active queue,
2119 * check whether to continue servicing it, or retrieve and set a new one.
2121 static struct cfq_queue
*cfq_select_queue(struct cfq_data
*cfqd
)
2123 struct cfq_queue
*cfqq
, *new_cfqq
= NULL
;
2125 cfqq
= cfqd
->active_queue
;
2129 if (!cfqd
->rq_queued
)
2132 * The active queue has run out of time, expire it and select new.
2134 if ((cfq_slice_used(cfqq
) || cfq_cfqq_wait_busy_done(cfqq
))
2135 && !cfq_cfqq_must_dispatch(cfqq
))
2139 * The active queue has requests and isn't expired, allow it to
2142 if (!RB_EMPTY_ROOT(&cfqq
->sort_list
))
2146 * If another queue has a request waiting within our mean seek
2147 * distance, let it run. The expire code will check for close
2148 * cooperators and put the close queue at the front of the service
2149 * tree. If possible, merge the expiring queue with the new cfqq.
2151 new_cfqq
= cfq_close_cooperator(cfqd
, cfqq
);
2153 if (!cfqq
->new_cfqq
)
2154 cfq_setup_merge(cfqq
, new_cfqq
);
2159 * No requests pending. If the active queue still has requests in
2160 * flight or is idling for a new request, allow either of these
2161 * conditions to happen (or time out) before selecting a new queue.
2163 if (timer_pending(&cfqd
->idle_slice_timer
) ||
2164 (cfqq
->dispatched
&& cfq_should_idle(cfqd
, cfqq
))) {
2170 cfq_slice_expired(cfqd
, 0);
2173 * Current queue expired. Check if we have to switch to a new
2177 cfq_choose_cfqg(cfqd
);
2179 cfqq
= cfq_set_active_queue(cfqd
, new_cfqq
);
2184 static int __cfq_forced_dispatch_cfqq(struct cfq_queue
*cfqq
)
2188 while (cfqq
->next_rq
) {
2189 cfq_dispatch_insert(cfqq
->cfqd
->queue
, cfqq
->next_rq
);
2193 BUG_ON(!list_empty(&cfqq
->fifo
));
2195 /* By default cfqq is not expired if it is empty. Do it explicitly */
2196 __cfq_slice_expired(cfqq
->cfqd
, cfqq
, 0);
2201 * Drain our current requests. Used for barriers and when switching
2202 * io schedulers on-the-fly.
2204 static int cfq_forced_dispatch(struct cfq_data
*cfqd
)
2206 struct cfq_queue
*cfqq
;
2209 while ((cfqq
= cfq_get_next_queue_forced(cfqd
)) != NULL
)
2210 dispatched
+= __cfq_forced_dispatch_cfqq(cfqq
);
2212 cfq_slice_expired(cfqd
, 0);
2213 BUG_ON(cfqd
->busy_queues
);
2215 cfq_log(cfqd
, "forced_dispatch=%d", dispatched
);
2219 static bool cfq_may_dispatch(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
)
2221 unsigned int max_dispatch
;
2224 * Drain async requests before we start sync IO
2226 if (cfq_should_idle(cfqd
, cfqq
) && cfqd
->rq_in_driver
[BLK_RW_ASYNC
])
2230 * If this is an async queue and we have sync IO in flight, let it wait
2232 if (cfqd
->sync_flight
&& !cfq_cfqq_sync(cfqq
))
2235 max_dispatch
= cfqd
->cfq_quantum
;
2236 if (cfq_class_idle(cfqq
))
2240 * Does this cfqq already have too much IO in flight?
2242 if (cfqq
->dispatched
>= max_dispatch
) {
2244 * idle queue must always only have a single IO in flight
2246 if (cfq_class_idle(cfqq
))
2250 * We have other queues, don't allow more IO from this one
2252 if (cfqd
->busy_queues
> 1)
2256 * Sole queue user, no limit
2262 * Async queues must wait a bit before being allowed dispatch.
2263 * We also ramp up the dispatch depth gradually for async IO,
2264 * based on the last sync IO we serviced
2266 if (!cfq_cfqq_sync(cfqq
) && cfqd
->cfq_latency
) {
2267 unsigned long last_sync
= jiffies
- cfqd
->last_end_sync_rq
;
2270 depth
= last_sync
/ cfqd
->cfq_slice
[1];
2271 if (!depth
&& !cfqq
->dispatched
)
2273 if (depth
< max_dispatch
)
2274 max_dispatch
= depth
;
2278 * If we're below the current max, allow a dispatch
2280 return cfqq
->dispatched
< max_dispatch
;
2284 * Dispatch a request from cfqq, moving them to the request queue
2287 static bool cfq_dispatch_request(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
)
2291 BUG_ON(RB_EMPTY_ROOT(&cfqq
->sort_list
));
2293 if (!cfq_may_dispatch(cfqd
, cfqq
))
2297 * follow expired path, else get first next available
2299 rq
= cfq_check_fifo(cfqq
);
2304 * insert request into driver dispatch list
2306 cfq_dispatch_insert(cfqd
->queue
, rq
);
2308 if (!cfqd
->active_cic
) {
2309 struct cfq_io_context
*cic
= RQ_CIC(rq
);
2311 atomic_long_inc(&cic
->ioc
->refcount
);
2312 cfqd
->active_cic
= cic
;
2319 * Find the cfqq that we need to service and move a request from that to the
2322 static int cfq_dispatch_requests(struct request_queue
*q
, int force
)
2324 struct cfq_data
*cfqd
= q
->elevator
->elevator_data
;
2325 struct cfq_queue
*cfqq
;
2327 if (!cfqd
->busy_queues
)
2330 if (unlikely(force
))
2331 return cfq_forced_dispatch(cfqd
);
2333 cfqq
= cfq_select_queue(cfqd
);
2338 * Dispatch a request from this cfqq, if it is allowed
2340 if (!cfq_dispatch_request(cfqd
, cfqq
))
2343 cfqq
->slice_dispatch
++;
2344 cfq_clear_cfqq_must_dispatch(cfqq
);
2347 * expire an async queue immediately if it has used up its slice. idle
2348 * queue always expire after 1 dispatch round.
2350 if (cfqd
->busy_queues
> 1 && ((!cfq_cfqq_sync(cfqq
) &&
2351 cfqq
->slice_dispatch
>= cfq_prio_to_maxrq(cfqd
, cfqq
)) ||
2352 cfq_class_idle(cfqq
))) {
2353 cfqq
->slice_end
= jiffies
+ 1;
2354 cfq_slice_expired(cfqd
, 0);
2357 cfq_log_cfqq(cfqd
, cfqq
, "dispatched a request");
2362 * task holds one reference to the queue, dropped when task exits. each rq
2363 * in-flight on this queue also holds a reference, dropped when rq is freed.
2365 * Each cfq queue took a reference on the parent group. Drop it now.
2366 * queue lock must be held here.
2368 static void cfq_put_queue(struct cfq_queue
*cfqq
)
2370 struct cfq_data
*cfqd
= cfqq
->cfqd
;
2371 struct cfq_group
*cfqg
;
2373 BUG_ON(atomic_read(&cfqq
->ref
) <= 0);
2375 if (!atomic_dec_and_test(&cfqq
->ref
))
2378 cfq_log_cfqq(cfqd
, cfqq
, "put_queue");
2379 BUG_ON(rb_first(&cfqq
->sort_list
));
2380 BUG_ON(cfqq
->allocated
[READ
] + cfqq
->allocated
[WRITE
]);
2383 if (unlikely(cfqd
->active_queue
== cfqq
)) {
2384 __cfq_slice_expired(cfqd
, cfqq
, 0);
2385 cfq_schedule_dispatch(cfqd
);
2388 BUG_ON(cfq_cfqq_on_rr(cfqq
));
2389 kmem_cache_free(cfq_pool
, cfqq
);
2391 if (cfqq
->orig_cfqg
)
2392 cfq_put_cfqg(cfqq
->orig_cfqg
);
2396 * Must always be called with the rcu_read_lock() held
2399 __call_for_each_cic(struct io_context
*ioc
,
2400 void (*func
)(struct io_context
*, struct cfq_io_context
*))
2402 struct cfq_io_context
*cic
;
2403 struct hlist_node
*n
;
2405 hlist_for_each_entry_rcu(cic
, n
, &ioc
->cic_list
, cic_list
)
2410 * Call func for each cic attached to this ioc.
2413 call_for_each_cic(struct io_context
*ioc
,
2414 void (*func
)(struct io_context
*, struct cfq_io_context
*))
2417 __call_for_each_cic(ioc
, func
);
2421 static void cfq_cic_free_rcu(struct rcu_head
*head
)
2423 struct cfq_io_context
*cic
;
2425 cic
= container_of(head
, struct cfq_io_context
, rcu_head
);
2427 kmem_cache_free(cfq_ioc_pool
, cic
);
2428 elv_ioc_count_dec(cfq_ioc_count
);
2432 * CFQ scheduler is exiting, grab exit lock and check
2433 * the pending io context count. If it hits zero,
2434 * complete ioc_gone and set it back to NULL
2436 spin_lock(&ioc_gone_lock
);
2437 if (ioc_gone
&& !elv_ioc_count_read(cfq_ioc_count
)) {
2441 spin_unlock(&ioc_gone_lock
);
2445 static void cfq_cic_free(struct cfq_io_context
*cic
)
2447 call_rcu(&cic
->rcu_head
, cfq_cic_free_rcu
);
2450 static void cic_free_func(struct io_context
*ioc
, struct cfq_io_context
*cic
)
2452 unsigned long flags
;
2454 BUG_ON(!cic
->dead_key
);
2456 spin_lock_irqsave(&ioc
->lock
, flags
);
2457 radix_tree_delete(&ioc
->radix_root
, cic
->dead_key
);
2458 hlist_del_rcu(&cic
->cic_list
);
2459 spin_unlock_irqrestore(&ioc
->lock
, flags
);
2465 * Must be called with rcu_read_lock() held or preemption otherwise disabled.
2466 * Only two callers of this - ->dtor() which is called with the rcu_read_lock(),
2467 * and ->trim() which is called with the task lock held
2469 static void cfq_free_io_context(struct io_context
*ioc
)
2472 * ioc->refcount is zero here, or we are called from elv_unregister(),
2473 * so no more cic's are allowed to be linked into this ioc. So it
2474 * should be ok to iterate over the known list, we will see all cic's
2475 * since no new ones are added.
2477 __call_for_each_cic(ioc
, cic_free_func
);
2480 static void cfq_exit_cfqq(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
)
2482 struct cfq_queue
*__cfqq
, *next
;
2484 if (unlikely(cfqq
== cfqd
->active_queue
)) {
2485 __cfq_slice_expired(cfqd
, cfqq
, 0);
2486 cfq_schedule_dispatch(cfqd
);
2490 * If this queue was scheduled to merge with another queue, be
2491 * sure to drop the reference taken on that queue (and others in
2492 * the merge chain). See cfq_setup_merge and cfq_merge_cfqqs.
2494 __cfqq
= cfqq
->new_cfqq
;
2496 if (__cfqq
== cfqq
) {
2497 WARN(1, "cfqq->new_cfqq loop detected\n");
2500 next
= __cfqq
->new_cfqq
;
2501 cfq_put_queue(__cfqq
);
2505 cfq_put_queue(cfqq
);
2508 static void __cfq_exit_single_io_context(struct cfq_data
*cfqd
,
2509 struct cfq_io_context
*cic
)
2511 struct io_context
*ioc
= cic
->ioc
;
2513 list_del_init(&cic
->queue_list
);
2516 * Make sure key == NULL is seen for dead queues
2519 cic
->dead_key
= (unsigned long) cic
->key
;
2522 if (ioc
->ioc_data
== cic
)
2523 rcu_assign_pointer(ioc
->ioc_data
, NULL
);
2525 if (cic
->cfqq
[BLK_RW_ASYNC
]) {
2526 cfq_exit_cfqq(cfqd
, cic
->cfqq
[BLK_RW_ASYNC
]);
2527 cic
->cfqq
[BLK_RW_ASYNC
] = NULL
;
2530 if (cic
->cfqq
[BLK_RW_SYNC
]) {
2531 cfq_exit_cfqq(cfqd
, cic
->cfqq
[BLK_RW_SYNC
]);
2532 cic
->cfqq
[BLK_RW_SYNC
] = NULL
;
2536 static void cfq_exit_single_io_context(struct io_context
*ioc
,
2537 struct cfq_io_context
*cic
)
2539 struct cfq_data
*cfqd
= cic
->key
;
2542 struct request_queue
*q
= cfqd
->queue
;
2543 unsigned long flags
;
2545 spin_lock_irqsave(q
->queue_lock
, flags
);
2548 * Ensure we get a fresh copy of the ->key to prevent
2549 * race between exiting task and queue
2551 smp_read_barrier_depends();
2553 __cfq_exit_single_io_context(cfqd
, cic
);
2555 spin_unlock_irqrestore(q
->queue_lock
, flags
);
2560 * The process that ioc belongs to has exited, we need to clean up
2561 * and put the internal structures we have that belongs to that process.
2563 static void cfq_exit_io_context(struct io_context
*ioc
)
2565 call_for_each_cic(ioc
, cfq_exit_single_io_context
);
2568 static struct cfq_io_context
*
2569 cfq_alloc_io_context(struct cfq_data
*cfqd
, gfp_t gfp_mask
)
2571 struct cfq_io_context
*cic
;
2573 cic
= kmem_cache_alloc_node(cfq_ioc_pool
, gfp_mask
| __GFP_ZERO
,
2576 cic
->last_end_request
= jiffies
;
2577 INIT_LIST_HEAD(&cic
->queue_list
);
2578 INIT_HLIST_NODE(&cic
->cic_list
);
2579 cic
->dtor
= cfq_free_io_context
;
2580 cic
->exit
= cfq_exit_io_context
;
2581 elv_ioc_count_inc(cfq_ioc_count
);
2587 static void cfq_init_prio_data(struct cfq_queue
*cfqq
, struct io_context
*ioc
)
2589 struct task_struct
*tsk
= current
;
2592 if (!cfq_cfqq_prio_changed(cfqq
))
2595 ioprio_class
= IOPRIO_PRIO_CLASS(ioc
->ioprio
);
2596 switch (ioprio_class
) {
2598 printk(KERN_ERR
"cfq: bad prio %x\n", ioprio_class
);
2599 case IOPRIO_CLASS_NONE
:
2601 * no prio set, inherit CPU scheduling settings
2603 cfqq
->ioprio
= task_nice_ioprio(tsk
);
2604 cfqq
->ioprio_class
= task_nice_ioclass(tsk
);
2606 case IOPRIO_CLASS_RT
:
2607 cfqq
->ioprio
= task_ioprio(ioc
);
2608 cfqq
->ioprio_class
= IOPRIO_CLASS_RT
;
2610 case IOPRIO_CLASS_BE
:
2611 cfqq
->ioprio
= task_ioprio(ioc
);
2612 cfqq
->ioprio_class
= IOPRIO_CLASS_BE
;
2614 case IOPRIO_CLASS_IDLE
:
2615 cfqq
->ioprio_class
= IOPRIO_CLASS_IDLE
;
2617 cfq_clear_cfqq_idle_window(cfqq
);
2622 * keep track of original prio settings in case we have to temporarily
2623 * elevate the priority of this queue
2625 cfqq
->org_ioprio
= cfqq
->ioprio
;
2626 cfqq
->org_ioprio_class
= cfqq
->ioprio_class
;
2627 cfq_clear_cfqq_prio_changed(cfqq
);
2630 static void changed_ioprio(struct io_context
*ioc
, struct cfq_io_context
*cic
)
2632 struct cfq_data
*cfqd
= cic
->key
;
2633 struct cfq_queue
*cfqq
;
2634 unsigned long flags
;
2636 if (unlikely(!cfqd
))
2639 spin_lock_irqsave(cfqd
->queue
->queue_lock
, flags
);
2641 cfqq
= cic
->cfqq
[BLK_RW_ASYNC
];
2643 struct cfq_queue
*new_cfqq
;
2644 new_cfqq
= cfq_get_queue(cfqd
, BLK_RW_ASYNC
, cic
->ioc
,
2647 cic
->cfqq
[BLK_RW_ASYNC
] = new_cfqq
;
2648 cfq_put_queue(cfqq
);
2652 cfqq
= cic
->cfqq
[BLK_RW_SYNC
];
2654 cfq_mark_cfqq_prio_changed(cfqq
);
2656 spin_unlock_irqrestore(cfqd
->queue
->queue_lock
, flags
);
2659 static void cfq_ioc_set_ioprio(struct io_context
*ioc
)
2661 call_for_each_cic(ioc
, changed_ioprio
);
2662 ioc
->ioprio_changed
= 0;
2665 static void cfq_init_cfqq(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
,
2666 pid_t pid
, bool is_sync
)
2668 RB_CLEAR_NODE(&cfqq
->rb_node
);
2669 RB_CLEAR_NODE(&cfqq
->p_node
);
2670 INIT_LIST_HEAD(&cfqq
->fifo
);
2672 atomic_set(&cfqq
->ref
, 0);
2675 cfq_mark_cfqq_prio_changed(cfqq
);
2678 if (!cfq_class_idle(cfqq
))
2679 cfq_mark_cfqq_idle_window(cfqq
);
2680 cfq_mark_cfqq_sync(cfqq
);
2685 #ifdef CONFIG_CFQ_GROUP_IOSCHED
2686 static void changed_cgroup(struct io_context
*ioc
, struct cfq_io_context
*cic
)
2688 struct cfq_queue
*sync_cfqq
= cic_to_cfqq(cic
, 1);
2689 struct cfq_data
*cfqd
= cic
->key
;
2690 unsigned long flags
;
2691 struct request_queue
*q
;
2693 if (unlikely(!cfqd
))
2698 spin_lock_irqsave(q
->queue_lock
, flags
);
2702 * Drop reference to sync queue. A new sync queue will be
2703 * assigned in new group upon arrival of a fresh request.
2705 cfq_log_cfqq(cfqd
, sync_cfqq
, "changed cgroup");
2706 cic_set_cfqq(cic
, NULL
, 1);
2707 cfq_put_queue(sync_cfqq
);
2710 spin_unlock_irqrestore(q
->queue_lock
, flags
);
2713 static void cfq_ioc_set_cgroup(struct io_context
*ioc
)
2715 call_for_each_cic(ioc
, changed_cgroup
);
2716 ioc
->cgroup_changed
= 0;
2718 #endif /* CONFIG_CFQ_GROUP_IOSCHED */
2720 static struct cfq_queue
*
2721 cfq_find_alloc_queue(struct cfq_data
*cfqd
, bool is_sync
,
2722 struct io_context
*ioc
, gfp_t gfp_mask
)
2724 struct cfq_queue
*cfqq
, *new_cfqq
= NULL
;
2725 struct cfq_io_context
*cic
;
2726 struct cfq_group
*cfqg
;
2729 cfqg
= cfq_get_cfqg(cfqd
, 1);
2730 cic
= cfq_cic_lookup(cfqd
, ioc
);
2731 /* cic always exists here */
2732 cfqq
= cic_to_cfqq(cic
, is_sync
);
2735 * Always try a new alloc if we fell back to the OOM cfqq
2736 * originally, since it should just be a temporary situation.
2738 if (!cfqq
|| cfqq
== &cfqd
->oom_cfqq
) {
2743 } else if (gfp_mask
& __GFP_WAIT
) {
2744 spin_unlock_irq(cfqd
->queue
->queue_lock
);
2745 new_cfqq
= kmem_cache_alloc_node(cfq_pool
,
2746 gfp_mask
| __GFP_ZERO
,
2748 spin_lock_irq(cfqd
->queue
->queue_lock
);
2752 cfqq
= kmem_cache_alloc_node(cfq_pool
,
2753 gfp_mask
| __GFP_ZERO
,
2758 cfq_init_cfqq(cfqd
, cfqq
, current
->pid
, is_sync
);
2759 cfq_init_prio_data(cfqq
, ioc
);
2760 cfq_link_cfqq_cfqg(cfqq
, cfqg
);
2761 cfq_log_cfqq(cfqd
, cfqq
, "alloced");
2763 cfqq
= &cfqd
->oom_cfqq
;
2767 kmem_cache_free(cfq_pool
, new_cfqq
);
2772 static struct cfq_queue
**
2773 cfq_async_queue_prio(struct cfq_data
*cfqd
, int ioprio_class
, int ioprio
)
2775 switch (ioprio_class
) {
2776 case IOPRIO_CLASS_RT
:
2777 return &cfqd
->async_cfqq
[0][ioprio
];
2778 case IOPRIO_CLASS_BE
:
2779 return &cfqd
->async_cfqq
[1][ioprio
];
2780 case IOPRIO_CLASS_IDLE
:
2781 return &cfqd
->async_idle_cfqq
;
2787 static struct cfq_queue
*
2788 cfq_get_queue(struct cfq_data
*cfqd
, bool is_sync
, struct io_context
*ioc
,
2791 const int ioprio
= task_ioprio(ioc
);
2792 const int ioprio_class
= task_ioprio_class(ioc
);
2793 struct cfq_queue
**async_cfqq
= NULL
;
2794 struct cfq_queue
*cfqq
= NULL
;
2797 async_cfqq
= cfq_async_queue_prio(cfqd
, ioprio_class
, ioprio
);
2802 cfqq
= cfq_find_alloc_queue(cfqd
, is_sync
, ioc
, gfp_mask
);
2805 * pin the queue now that it's allocated, scheduler exit will prune it
2807 if (!is_sync
&& !(*async_cfqq
)) {
2808 atomic_inc(&cfqq
->ref
);
2812 atomic_inc(&cfqq
->ref
);
2817 * We drop cfq io contexts lazily, so we may find a dead one.
2820 cfq_drop_dead_cic(struct cfq_data
*cfqd
, struct io_context
*ioc
,
2821 struct cfq_io_context
*cic
)
2823 unsigned long flags
;
2825 WARN_ON(!list_empty(&cic
->queue_list
));
2827 spin_lock_irqsave(&ioc
->lock
, flags
);
2829 BUG_ON(ioc
->ioc_data
== cic
);
2831 radix_tree_delete(&ioc
->radix_root
, (unsigned long) cfqd
);
2832 hlist_del_rcu(&cic
->cic_list
);
2833 spin_unlock_irqrestore(&ioc
->lock
, flags
);
2838 static struct cfq_io_context
*
2839 cfq_cic_lookup(struct cfq_data
*cfqd
, struct io_context
*ioc
)
2841 struct cfq_io_context
*cic
;
2842 unsigned long flags
;
2851 * we maintain a last-hit cache, to avoid browsing over the tree
2853 cic
= rcu_dereference(ioc
->ioc_data
);
2854 if (cic
&& cic
->key
== cfqd
) {
2860 cic
= radix_tree_lookup(&ioc
->radix_root
, (unsigned long) cfqd
);
2864 /* ->key must be copied to avoid race with cfq_exit_queue() */
2867 cfq_drop_dead_cic(cfqd
, ioc
, cic
);
2872 spin_lock_irqsave(&ioc
->lock
, flags
);
2873 rcu_assign_pointer(ioc
->ioc_data
, cic
);
2874 spin_unlock_irqrestore(&ioc
->lock
, flags
);
2882 * Add cic into ioc, using cfqd as the search key. This enables us to lookup
2883 * the process specific cfq io context when entered from the block layer.
2884 * Also adds the cic to a per-cfqd list, used when this queue is removed.
2886 static int cfq_cic_link(struct cfq_data
*cfqd
, struct io_context
*ioc
,
2887 struct cfq_io_context
*cic
, gfp_t gfp_mask
)
2889 unsigned long flags
;
2892 ret
= radix_tree_preload(gfp_mask
);
2897 spin_lock_irqsave(&ioc
->lock
, flags
);
2898 ret
= radix_tree_insert(&ioc
->radix_root
,
2899 (unsigned long) cfqd
, cic
);
2901 hlist_add_head_rcu(&cic
->cic_list
, &ioc
->cic_list
);
2902 spin_unlock_irqrestore(&ioc
->lock
, flags
);
2904 radix_tree_preload_end();
2907 spin_lock_irqsave(cfqd
->queue
->queue_lock
, flags
);
2908 list_add(&cic
->queue_list
, &cfqd
->cic_list
);
2909 spin_unlock_irqrestore(cfqd
->queue
->queue_lock
, flags
);
2914 printk(KERN_ERR
"cfq: cic link failed!\n");
2920 * Setup general io context and cfq io context. There can be several cfq
2921 * io contexts per general io context, if this process is doing io to more
2922 * than one device managed by cfq.
2924 static struct cfq_io_context
*
2925 cfq_get_io_context(struct cfq_data
*cfqd
, gfp_t gfp_mask
)
2927 struct io_context
*ioc
= NULL
;
2928 struct cfq_io_context
*cic
;
2930 might_sleep_if(gfp_mask
& __GFP_WAIT
);
2932 ioc
= get_io_context(gfp_mask
, cfqd
->queue
->node
);
2936 cic
= cfq_cic_lookup(cfqd
, ioc
);
2940 cic
= cfq_alloc_io_context(cfqd
, gfp_mask
);
2944 if (cfq_cic_link(cfqd
, ioc
, cic
, gfp_mask
))
2948 smp_read_barrier_depends();
2949 if (unlikely(ioc
->ioprio_changed
))
2950 cfq_ioc_set_ioprio(ioc
);
2952 #ifdef CONFIG_CFQ_GROUP_IOSCHED
2953 if (unlikely(ioc
->cgroup_changed
))
2954 cfq_ioc_set_cgroup(ioc
);
2960 put_io_context(ioc
);
2965 cfq_update_io_thinktime(struct cfq_data
*cfqd
, struct cfq_io_context
*cic
)
2967 unsigned long elapsed
= jiffies
- cic
->last_end_request
;
2968 unsigned long ttime
= min(elapsed
, 2UL * cfqd
->cfq_slice_idle
);
2970 cic
->ttime_samples
= (7*cic
->ttime_samples
+ 256) / 8;
2971 cic
->ttime_total
= (7*cic
->ttime_total
+ 256*ttime
) / 8;
2972 cic
->ttime_mean
= (cic
->ttime_total
+ 128) / cic
->ttime_samples
;
2976 cfq_update_io_seektime(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
,
2982 if (!cfqq
->last_request_pos
)
2984 else if (cfqq
->last_request_pos
< blk_rq_pos(rq
))
2985 sdist
= blk_rq_pos(rq
) - cfqq
->last_request_pos
;
2987 sdist
= cfqq
->last_request_pos
- blk_rq_pos(rq
);
2990 * Don't allow the seek distance to get too large from the
2991 * odd fragment, pagein, etc
2993 if (cfqq
->seek_samples
<= 60) /* second&third seek */
2994 sdist
= min(sdist
, (cfqq
->seek_mean
* 4) + 2*1024*1024);
2996 sdist
= min(sdist
, (cfqq
->seek_mean
* 4) + 2*1024*64);
2998 cfqq
->seek_samples
= (7*cfqq
->seek_samples
+ 256) / 8;
2999 cfqq
->seek_total
= (7*cfqq
->seek_total
+ (u64
)256*sdist
) / 8;
3000 total
= cfqq
->seek_total
+ (cfqq
->seek_samples
/2);
3001 do_div(total
, cfqq
->seek_samples
);
3002 cfqq
->seek_mean
= (sector_t
)total
;
3005 * If this cfqq is shared between multiple processes, check to
3006 * make sure that those processes are still issuing I/Os within
3007 * the mean seek distance. If not, it may be time to break the
3008 * queues apart again.
3010 if (cfq_cfqq_coop(cfqq
)) {
3011 if (CFQQ_SEEKY(cfqq
) && !cfqq
->seeky_start
)
3012 cfqq
->seeky_start
= jiffies
;
3013 else if (!CFQQ_SEEKY(cfqq
))
3014 cfqq
->seeky_start
= 0;
3019 * Disable idle window if the process thinks too long or seeks so much that
3023 cfq_update_idle_window(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
,
3024 struct cfq_io_context
*cic
)
3026 int old_idle
, enable_idle
;
3029 * Don't idle for async or idle io prio class
3031 if (!cfq_cfqq_sync(cfqq
) || cfq_class_idle(cfqq
))
3034 enable_idle
= old_idle
= cfq_cfqq_idle_window(cfqq
);
3036 if (cfqq
->queued
[0] + cfqq
->queued
[1] >= 4)
3037 cfq_mark_cfqq_deep(cfqq
);
3039 if (!atomic_read(&cic
->ioc
->nr_tasks
) || !cfqd
->cfq_slice_idle
||
3040 (!cfq_cfqq_deep(cfqq
) && sample_valid(cfqq
->seek_samples
)
3041 && CFQQ_SEEKY(cfqq
)))
3043 else if (sample_valid(cic
->ttime_samples
)) {
3044 if (cic
->ttime_mean
> cfqd
->cfq_slice_idle
)
3050 if (old_idle
!= enable_idle
) {
3051 cfq_log_cfqq(cfqd
, cfqq
, "idle=%d", enable_idle
);
3053 cfq_mark_cfqq_idle_window(cfqq
);
3055 cfq_clear_cfqq_idle_window(cfqq
);
3060 * Check if new_cfqq should preempt the currently active queue. Return 0 for
3061 * no or if we aren't sure, a 1 will cause a preempt.
3064 cfq_should_preempt(struct cfq_data
*cfqd
, struct cfq_queue
*new_cfqq
,
3067 struct cfq_queue
*cfqq
;
3069 cfqq
= cfqd
->active_queue
;
3073 if (cfq_class_idle(new_cfqq
))
3076 if (cfq_class_idle(cfqq
))
3080 * if the new request is sync, but the currently running queue is
3081 * not, let the sync request have priority.
3083 if (rq_is_sync(rq
) && !cfq_cfqq_sync(cfqq
))
3086 if (new_cfqq
->cfqg
!= cfqq
->cfqg
)
3089 if (cfq_slice_used(cfqq
))
3092 /* Allow preemption only if we are idling on sync-noidle tree */
3093 if (cfqd
->serving_type
== SYNC_NOIDLE_WORKLOAD
&&
3094 cfqq_type(new_cfqq
) == SYNC_NOIDLE_WORKLOAD
&&
3095 new_cfqq
->service_tree
->count
== 2 &&
3096 RB_EMPTY_ROOT(&cfqq
->sort_list
))
3100 * So both queues are sync. Let the new request get disk time if
3101 * it's a metadata request and the current queue is doing regular IO.
3103 if (rq_is_meta(rq
) && !cfqq
->meta_pending
)
3107 * Allow an RT request to pre-empt an ongoing non-RT cfqq timeslice.
3109 if (cfq_class_rt(new_cfqq
) && !cfq_class_rt(cfqq
))
3112 if (!cfqd
->active_cic
|| !cfq_cfqq_wait_request(cfqq
))
3116 * if this request is as-good as one we would expect from the
3117 * current cfqq, let it preempt
3119 if (cfq_rq_close(cfqd
, cfqq
, rq
))
3126 * cfqq preempts the active queue. if we allowed preempt with no slice left,
3127 * let it have half of its nominal slice.
3129 static void cfq_preempt_queue(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
)
3131 cfq_log_cfqq(cfqd
, cfqq
, "preempt");
3132 cfq_slice_expired(cfqd
, 1);
3135 * Put the new queue at the front of the of the current list,
3136 * so we know that it will be selected next.
3138 BUG_ON(!cfq_cfqq_on_rr(cfqq
));
3140 cfq_service_tree_add(cfqd
, cfqq
, 1);
3142 cfqq
->slice_end
= 0;
3143 cfq_mark_cfqq_slice_new(cfqq
);
3147 * Called when a new fs request (rq) is added (to cfqq). Check if there's
3148 * something we should do about it
3151 cfq_rq_enqueued(struct cfq_data
*cfqd
, struct cfq_queue
*cfqq
,
3154 struct cfq_io_context
*cic
= RQ_CIC(rq
);
3158 cfqq
->meta_pending
++;
3160 cfq_update_io_thinktime(cfqd
, cic
);
3161 cfq_update_io_seektime(cfqd
, cfqq
, rq
);
3162 cfq_update_idle_window(cfqd
, cfqq
, cic
);
3164 cfqq
->last_request_pos
= blk_rq_pos(rq
) + blk_rq_sectors(rq
);
3166 if (cfqq
== cfqd
->active_queue
) {
3167 if (cfq_cfqq_wait_busy(cfqq
)) {
3168 cfq_clear_cfqq_wait_busy(cfqq
);
3169 cfq_mark_cfqq_wait_busy_done(cfqq
);
3172 * Remember that we saw a request from this process, but
3173 * don't start queuing just yet. Otherwise we risk seeing lots
3174 * of tiny requests, because we disrupt the normal plugging
3175 * and merging. If the request is already larger than a single
3176 * page, let it rip immediately. For that case we assume that
3177 * merging is already done. Ditto for a busy system that
3178 * has other work pending, don't risk delaying until the
3179 * idle timer unplug to continue working.
3181 if (cfq_cfqq_wait_request(cfqq
)) {
3182 if (blk_rq_bytes(rq
) > PAGE_CACHE_SIZE
||
3183 cfqd
->busy_queues
> 1) {
3184 del_timer(&cfqd
->idle_slice_timer
);
3185 __blk_run_queue(cfqd
->queue
);
3187 cfq_mark_cfqq_must_dispatch(cfqq
);
3189 } else if (cfq_should_preempt(cfqd
, cfqq
, rq
)) {
3191 * not the active queue - expire current slice if it is
3192 * idle and has expired it's mean thinktime or this new queue
3193 * has some old slice time left and is of higher priority or
3194 * this new queue is RT and the current one is BE
3196 cfq_preempt_queue(cfqd
, cfqq
);
3197 __blk_run_queue(cfqd
->queue
);
3201 static void cfq_insert_request(struct request_queue
*q
, struct request
*rq
)
3203 struct cfq_data
*cfqd
= q
->elevator
->elevator_data
;
3204 struct cfq_queue
*cfqq
= RQ_CFQQ(rq
);
3206 cfq_log_cfqq(cfqd
, cfqq
, "insert_request");
3207 cfq_init_prio_data(cfqq
, RQ_CIC(rq
)->ioc
);
3209 rq_set_fifo_time(rq
, jiffies
+ cfqd
->cfq_fifo_expire
[rq_is_sync(rq
)]);
3210 list_add_tail(&rq
->queuelist
, &cfqq
->fifo
);
3213 cfq_rq_enqueued(cfqd
, cfqq
, rq
);
3217 * Update hw_tag based on peak queue depth over 50 samples under
3220 static void cfq_update_hw_tag(struct cfq_data
*cfqd
)
3222 struct cfq_queue
*cfqq
= cfqd
->active_queue
;
3224 if (rq_in_driver(cfqd
) > cfqd
->hw_tag_est_depth
)
3225 cfqd
->hw_tag_est_depth
= rq_in_driver(cfqd
);
3227 if (cfqd
->hw_tag
== 1)
3230 if (cfqd
->rq_queued
<= CFQ_HW_QUEUE_MIN
&&
3231 rq_in_driver(cfqd
) <= CFQ_HW_QUEUE_MIN
)
3235 * If active queue hasn't enough requests and can idle, cfq might not
3236 * dispatch sufficient requests to hardware. Don't zero hw_tag in this
3239 if (cfqq
&& cfq_cfqq_idle_window(cfqq
) &&
3240 cfqq
->dispatched
+ cfqq
->queued
[0] + cfqq
->queued
[1] <
3241 CFQ_HW_QUEUE_MIN
&& rq_in_driver(cfqd
) < CFQ_HW_QUEUE_MIN
)
3244 if (cfqd
->hw_tag_samples
++ < 50)
3247 if (cfqd
->hw_tag_est_depth
>= CFQ_HW_QUEUE_MIN
)
3253 static void cfq_completed_request(struct request_queue
*q
, struct request
*rq
)
3255 struct cfq_queue
*cfqq
= RQ_CFQQ(rq
);
3256 struct cfq_data
*cfqd
= cfqq
->cfqd
;
3257 const int sync
= rq_is_sync(rq
);
3261 cfq_log_cfqq(cfqd
, cfqq
, "complete rqnoidle %d", !!rq_noidle(rq
));
3263 cfq_update_hw_tag(cfqd
);
3265 WARN_ON(!cfqd
->rq_in_driver
[sync
]);
3266 WARN_ON(!cfqq
->dispatched
);
3267 cfqd
->rq_in_driver
[sync
]--;
3270 if (cfq_cfqq_sync(cfqq
))
3271 cfqd
->sync_flight
--;
3274 RQ_CIC(rq
)->last_end_request
= now
;
3275 cfqd
->last_end_sync_rq
= now
;
3279 * If this is the active queue, check if it needs to be expired,
3280 * or if we want to idle in case it has no pending requests.
3282 if (cfqd
->active_queue
== cfqq
) {
3283 const bool cfqq_empty
= RB_EMPTY_ROOT(&cfqq
->sort_list
);
3285 if (cfq_cfqq_slice_new(cfqq
)) {
3286 cfq_set_prio_slice(cfqd
, cfqq
);
3287 cfq_clear_cfqq_slice_new(cfqq
);
3291 * If this queue consumed its slice and this is last queue
3292 * in the group, wait for next request before we expire
3295 if (cfq_slice_used(cfqq
) && cfqq
->cfqg
->nr_cfqq
== 1) {
3296 cfqq
->slice_end
= jiffies
+ cfqd
->cfq_slice_idle
;
3297 cfq_mark_cfqq_wait_busy(cfqq
);
3301 * Idling is not enabled on:
3303 * - idle-priority queues
3305 * - queues with still some requests queued
3306 * - when there is a close cooperator
3308 if (cfq_slice_used(cfqq
) || cfq_class_idle(cfqq
))
3309 cfq_slice_expired(cfqd
, 1);
3310 else if (sync
&& cfqq_empty
&&
3311 !cfq_close_cooperator(cfqd
, cfqq
)) {
3312 cfqd
->noidle_tree_requires_idle
|= !rq_noidle(rq
);
3314 * Idling is enabled for SYNC_WORKLOAD.
3315 * SYNC_NOIDLE_WORKLOAD idles at the end of the tree
3316 * only if we processed at least one !rq_noidle request
3318 if (cfqd
->serving_type
== SYNC_WORKLOAD
3319 || cfqd
->noidle_tree_requires_idle
3320 || cfqq
->cfqg
->nr_cfqq
== 1)
3321 cfq_arm_slice_timer(cfqd
);
3325 if (!rq_in_driver(cfqd
))
3326 cfq_schedule_dispatch(cfqd
);
3330 * we temporarily boost lower priority queues if they are holding fs exclusive
3331 * resources. they are boosted to normal prio (CLASS_BE/4)
3333 static void cfq_prio_boost(struct cfq_queue
*cfqq
)
3335 if (has_fs_excl()) {
3337 * boost idle prio on transactions that would lock out other
3338 * users of the filesystem
3340 if (cfq_class_idle(cfqq
))
3341 cfqq
->ioprio_class
= IOPRIO_CLASS_BE
;
3342 if (cfqq
->ioprio
> IOPRIO_NORM
)
3343 cfqq
->ioprio
= IOPRIO_NORM
;
3346 * unboost the queue (if needed)
3348 cfqq
->ioprio_class
= cfqq
->org_ioprio_class
;
3349 cfqq
->ioprio
= cfqq
->org_ioprio
;
3353 static inline int __cfq_may_queue(struct cfq_queue
*cfqq
)
3355 if (cfq_cfqq_wait_request(cfqq
) && !cfq_cfqq_must_alloc_slice(cfqq
)) {
3356 cfq_mark_cfqq_must_alloc_slice(cfqq
);
3357 return ELV_MQUEUE_MUST
;
3360 return ELV_MQUEUE_MAY
;
3363 static int cfq_may_queue(struct request_queue
*q
, int rw
)
3365 struct cfq_data
*cfqd
= q
->elevator
->elevator_data
;
3366 struct task_struct
*tsk
= current
;
3367 struct cfq_io_context
*cic
;
3368 struct cfq_queue
*cfqq
;
3371 * don't force setup of a queue from here, as a call to may_queue
3372 * does not necessarily imply that a request actually will be queued.
3373 * so just lookup a possibly existing queue, or return 'may queue'
3376 cic
= cfq_cic_lookup(cfqd
, tsk
->io_context
);
3378 return ELV_MQUEUE_MAY
;
3380 cfqq
= cic_to_cfqq(cic
, rw_is_sync(rw
));
3382 cfq_init_prio_data(cfqq
, cic
->ioc
);
3383 cfq_prio_boost(cfqq
);
3385 return __cfq_may_queue(cfqq
);
3388 return ELV_MQUEUE_MAY
;
3392 * queue lock held here
3394 static void cfq_put_request(struct request
*rq
)
3396 struct cfq_queue
*cfqq
= RQ_CFQQ(rq
);
3399 const int rw
= rq_data_dir(rq
);
3401 BUG_ON(!cfqq
->allocated
[rw
]);
3402 cfqq
->allocated
[rw
]--;
3404 put_io_context(RQ_CIC(rq
)->ioc
);
3406 rq
->elevator_private
= NULL
;
3407 rq
->elevator_private2
= NULL
;
3409 cfq_put_queue(cfqq
);
3413 static struct cfq_queue
*
3414 cfq_merge_cfqqs(struct cfq_data
*cfqd
, struct cfq_io_context
*cic
,
3415 struct cfq_queue
*cfqq
)
3417 cfq_log_cfqq(cfqd
, cfqq
, "merging with queue %p", cfqq
->new_cfqq
);
3418 cic_set_cfqq(cic
, cfqq
->new_cfqq
, 1);
3419 cfq_mark_cfqq_coop(cfqq
->new_cfqq
);
3420 cfq_put_queue(cfqq
);
3421 return cic_to_cfqq(cic
, 1);
3424 static int should_split_cfqq(struct cfq_queue
*cfqq
)
3426 if (cfqq
->seeky_start
&&
3427 time_after(jiffies
, cfqq
->seeky_start
+ CFQQ_COOP_TOUT
))
3433 * Returns NULL if a new cfqq should be allocated, or the old cfqq if this
3434 * was the last process referring to said cfqq.
3436 static struct cfq_queue
*
3437 split_cfqq(struct cfq_io_context
*cic
, struct cfq_queue
*cfqq
)
3439 if (cfqq_process_refs(cfqq
) == 1) {
3440 cfqq
->seeky_start
= 0;
3441 cfqq
->pid
= current
->pid
;
3442 cfq_clear_cfqq_coop(cfqq
);
3446 cic_set_cfqq(cic
, NULL
, 1);
3447 cfq_put_queue(cfqq
);
3451 * Allocate cfq data structures associated with this request.
3454 cfq_set_request(struct request_queue
*q
, struct request
*rq
, gfp_t gfp_mask
)
3456 struct cfq_data
*cfqd
= q
->elevator
->elevator_data
;
3457 struct cfq_io_context
*cic
;
3458 const int rw
= rq_data_dir(rq
);
3459 const bool is_sync
= rq_is_sync(rq
);
3460 struct cfq_queue
*cfqq
;
3461 unsigned long flags
;
3463 might_sleep_if(gfp_mask
& __GFP_WAIT
);
3465 cic
= cfq_get_io_context(cfqd
, gfp_mask
);
3467 spin_lock_irqsave(q
->queue_lock
, flags
);
3473 cfqq
= cic_to_cfqq(cic
, is_sync
);
3474 if (!cfqq
|| cfqq
== &cfqd
->oom_cfqq
) {
3475 cfqq
= cfq_get_queue(cfqd
, is_sync
, cic
->ioc
, gfp_mask
);
3476 cic_set_cfqq(cic
, cfqq
, is_sync
);
3479 * If the queue was seeky for too long, break it apart.
3481 if (cfq_cfqq_coop(cfqq
) && should_split_cfqq(cfqq
)) {
3482 cfq_log_cfqq(cfqd
, cfqq
, "breaking apart cfqq");
3483 cfqq
= split_cfqq(cic
, cfqq
);
3489 * Check to see if this queue is scheduled to merge with
3490 * another, closely cooperating queue. The merging of
3491 * queues happens here as it must be done in process context.
3492 * The reference on new_cfqq was taken in merge_cfqqs.
3495 cfqq
= cfq_merge_cfqqs(cfqd
, cic
, cfqq
);
3498 cfqq
->allocated
[rw
]++;
3499 atomic_inc(&cfqq
->ref
);
3501 spin_unlock_irqrestore(q
->queue_lock
, flags
);
3503 rq
->elevator_private
= cic
;
3504 rq
->elevator_private2
= cfqq
;
3509 put_io_context(cic
->ioc
);
3511 cfq_schedule_dispatch(cfqd
);
3512 spin_unlock_irqrestore(q
->queue_lock
, flags
);
3513 cfq_log(cfqd
, "set_request fail");
3517 static void cfq_kick_queue(struct work_struct
*work
)
3519 struct cfq_data
*cfqd
=
3520 container_of(work
, struct cfq_data
, unplug_work
);
3521 struct request_queue
*q
= cfqd
->queue
;
3523 spin_lock_irq(q
->queue_lock
);
3524 __blk_run_queue(cfqd
->queue
);
3525 spin_unlock_irq(q
->queue_lock
);
3529 * Timer running if the active_queue is currently idling inside its time slice
3531 static void cfq_idle_slice_timer(unsigned long data
)
3533 struct cfq_data
*cfqd
= (struct cfq_data
*) data
;
3534 struct cfq_queue
*cfqq
;
3535 unsigned long flags
;
3538 cfq_log(cfqd
, "idle timer fired");
3540 spin_lock_irqsave(cfqd
->queue
->queue_lock
, flags
);
3542 cfqq
= cfqd
->active_queue
;
3547 * We saw a request before the queue expired, let it through
3549 if (cfq_cfqq_must_dispatch(cfqq
))
3555 if (cfq_slice_used(cfqq
))
3559 * only expire and reinvoke request handler, if there are
3560 * other queues with pending requests
3562 if (!cfqd
->busy_queues
)
3566 * not expired and it has a request pending, let it dispatch
3568 if (!RB_EMPTY_ROOT(&cfqq
->sort_list
))
3572 * Queue depth flag is reset only when the idle didn't succeed
3574 cfq_clear_cfqq_deep(cfqq
);
3577 cfq_slice_expired(cfqd
, timed_out
);
3579 cfq_schedule_dispatch(cfqd
);
3581 spin_unlock_irqrestore(cfqd
->queue
->queue_lock
, flags
);
3584 static void cfq_shutdown_timer_wq(struct cfq_data
*cfqd
)
3586 del_timer_sync(&cfqd
->idle_slice_timer
);
3587 cancel_work_sync(&cfqd
->unplug_work
);
3590 static void cfq_put_async_queues(struct cfq_data
*cfqd
)
3594 for (i
= 0; i
< IOPRIO_BE_NR
; i
++) {
3595 if (cfqd
->async_cfqq
[0][i
])
3596 cfq_put_queue(cfqd
->async_cfqq
[0][i
]);
3597 if (cfqd
->async_cfqq
[1][i
])
3598 cfq_put_queue(cfqd
->async_cfqq
[1][i
]);
3601 if (cfqd
->async_idle_cfqq
)
3602 cfq_put_queue(cfqd
->async_idle_cfqq
);
3605 static void cfq_exit_queue(struct elevator_queue
*e
)
3607 struct cfq_data
*cfqd
= e
->elevator_data
;
3608 struct request_queue
*q
= cfqd
->queue
;
3610 cfq_shutdown_timer_wq(cfqd
);
3612 spin_lock_irq(q
->queue_lock
);
3614 if (cfqd
->active_queue
)
3615 __cfq_slice_expired(cfqd
, cfqd
->active_queue
, 0);
3617 while (!list_empty(&cfqd
->cic_list
)) {
3618 struct cfq_io_context
*cic
= list_entry(cfqd
->cic_list
.next
,
3619 struct cfq_io_context
,
3622 __cfq_exit_single_io_context(cfqd
, cic
);
3625 cfq_put_async_queues(cfqd
);
3626 cfq_release_cfq_groups(cfqd
);
3627 blkiocg_del_blkio_group(&cfqd
->root_group
.blkg
);
3629 spin_unlock_irq(q
->queue_lock
);
3631 cfq_shutdown_timer_wq(cfqd
);
3633 /* Wait for cfqg->blkg->key accessors to exit their grace periods. */
3638 static void *cfq_init_queue(struct request_queue
*q
)
3640 struct cfq_data
*cfqd
;
3642 struct cfq_group
*cfqg
;
3643 struct cfq_rb_root
*st
;
3645 cfqd
= kmalloc_node(sizeof(*cfqd
), GFP_KERNEL
| __GFP_ZERO
, q
->node
);
3649 /* Init root service tree */
3650 cfqd
->grp_service_tree
= CFQ_RB_ROOT
;
3652 /* Init root group */
3653 cfqg
= &cfqd
->root_group
;
3654 for_each_cfqg_st(cfqg
, i
, j
, st
)
3656 RB_CLEAR_NODE(&cfqg
->rb_node
);
3658 /* Give preference to root group over other groups */
3659 cfqg
->weight
= 2*BLKIO_WEIGHT_DEFAULT
;
3661 #ifdef CONFIG_CFQ_GROUP_IOSCHED
3663 * Take a reference to root group which we never drop. This is just
3664 * to make sure that cfq_put_cfqg() does not try to kfree root group
3666 atomic_set(&cfqg
->ref
, 1);
3667 blkiocg_add_blkio_group(&blkio_root_cgroup
, &cfqg
->blkg
, (void *)cfqd
,
3671 * Not strictly needed (since RB_ROOT just clears the node and we
3672 * zeroed cfqd on alloc), but better be safe in case someone decides
3673 * to add magic to the rb code
3675 for (i
= 0; i
< CFQ_PRIO_LISTS
; i
++)
3676 cfqd
->prio_trees
[i
] = RB_ROOT
;
3679 * Our fallback cfqq if cfq_find_alloc_queue() runs into OOM issues.
3680 * Grab a permanent reference to it, so that the normal code flow
3681 * will not attempt to free it.
3683 cfq_init_cfqq(cfqd
, &cfqd
->oom_cfqq
, 1, 0);
3684 atomic_inc(&cfqd
->oom_cfqq
.ref
);
3685 cfq_link_cfqq_cfqg(&cfqd
->oom_cfqq
, &cfqd
->root_group
);
3687 INIT_LIST_HEAD(&cfqd
->cic_list
);
3691 init_timer(&cfqd
->idle_slice_timer
);
3692 cfqd
->idle_slice_timer
.function
= cfq_idle_slice_timer
;
3693 cfqd
->idle_slice_timer
.data
= (unsigned long) cfqd
;
3695 INIT_WORK(&cfqd
->unplug_work
, cfq_kick_queue
);
3697 cfqd
->cfq_quantum
= cfq_quantum
;
3698 cfqd
->cfq_fifo_expire
[0] = cfq_fifo_expire
[0];
3699 cfqd
->cfq_fifo_expire
[1] = cfq_fifo_expire
[1];
3700 cfqd
->cfq_back_max
= cfq_back_max
;
3701 cfqd
->cfq_back_penalty
= cfq_back_penalty
;
3702 cfqd
->cfq_slice
[0] = cfq_slice_async
;
3703 cfqd
->cfq_slice
[1] = cfq_slice_sync
;
3704 cfqd
->cfq_slice_async_rq
= cfq_slice_async_rq
;
3705 cfqd
->cfq_slice_idle
= cfq_slice_idle
;
3706 cfqd
->cfq_latency
= 1;
3707 cfqd
->cfq_group_isolation
= 0;
3709 cfqd
->last_end_sync_rq
= jiffies
;
3713 static void cfq_slab_kill(void)
3716 * Caller already ensured that pending RCU callbacks are completed,
3717 * so we should have no busy allocations at this point.
3720 kmem_cache_destroy(cfq_pool
);
3722 kmem_cache_destroy(cfq_ioc_pool
);
3725 static int __init
cfq_slab_setup(void)
3727 cfq_pool
= KMEM_CACHE(cfq_queue
, 0);
3731 cfq_ioc_pool
= KMEM_CACHE(cfq_io_context
, 0);
3742 * sysfs parts below -->
3745 cfq_var_show(unsigned int var
, char *page
)
3747 return sprintf(page
, "%d\n", var
);
3751 cfq_var_store(unsigned int *var
, const char *page
, size_t count
)
3753 char *p
= (char *) page
;
3755 *var
= simple_strtoul(p
, &p
, 10);
3759 #define SHOW_FUNCTION(__FUNC, __VAR, __CONV) \
3760 static ssize_t __FUNC(struct elevator_queue *e, char *page) \
3762 struct cfq_data *cfqd = e->elevator_data; \
3763 unsigned int __data = __VAR; \
3765 __data = jiffies_to_msecs(__data); \
3766 return cfq_var_show(__data, (page)); \
3768 SHOW_FUNCTION(cfq_quantum_show
, cfqd
->cfq_quantum
, 0);
3769 SHOW_FUNCTION(cfq_fifo_expire_sync_show
, cfqd
->cfq_fifo_expire
[1], 1);
3770 SHOW_FUNCTION(cfq_fifo_expire_async_show
, cfqd
->cfq_fifo_expire
[0], 1);
3771 SHOW_FUNCTION(cfq_back_seek_max_show
, cfqd
->cfq_back_max
, 0);
3772 SHOW_FUNCTION(cfq_back_seek_penalty_show
, cfqd
->cfq_back_penalty
, 0);
3773 SHOW_FUNCTION(cfq_slice_idle_show
, cfqd
->cfq_slice_idle
, 1);
3774 SHOW_FUNCTION(cfq_slice_sync_show
, cfqd
->cfq_slice
[1], 1);
3775 SHOW_FUNCTION(cfq_slice_async_show
, cfqd
->cfq_slice
[0], 1);
3776 SHOW_FUNCTION(cfq_slice_async_rq_show
, cfqd
->cfq_slice_async_rq
, 0);
3777 SHOW_FUNCTION(cfq_low_latency_show
, cfqd
->cfq_latency
, 0);
3778 SHOW_FUNCTION(cfq_group_isolation_show
, cfqd
->cfq_group_isolation
, 0);
3779 #undef SHOW_FUNCTION
3781 #define STORE_FUNCTION(__FUNC, __PTR, MIN, MAX, __CONV) \
3782 static ssize_t __FUNC(struct elevator_queue *e, const char *page, size_t count) \
3784 struct cfq_data *cfqd = e->elevator_data; \
3785 unsigned int __data; \
3786 int ret = cfq_var_store(&__data, (page), count); \
3787 if (__data < (MIN)) \
3789 else if (__data > (MAX)) \
3792 *(__PTR) = msecs_to_jiffies(__data); \
3794 *(__PTR) = __data; \
3797 STORE_FUNCTION(cfq_quantum_store
, &cfqd
->cfq_quantum
, 1, UINT_MAX
, 0);
3798 STORE_FUNCTION(cfq_fifo_expire_sync_store
, &cfqd
->cfq_fifo_expire
[1], 1,
3800 STORE_FUNCTION(cfq_fifo_expire_async_store
, &cfqd
->cfq_fifo_expire
[0], 1,
3802 STORE_FUNCTION(cfq_back_seek_max_store
, &cfqd
->cfq_back_max
, 0, UINT_MAX
, 0);
3803 STORE_FUNCTION(cfq_back_seek_penalty_store
, &cfqd
->cfq_back_penalty
, 1,
3805 STORE_FUNCTION(cfq_slice_idle_store
, &cfqd
->cfq_slice_idle
, 0, UINT_MAX
, 1);
3806 STORE_FUNCTION(cfq_slice_sync_store
, &cfqd
->cfq_slice
[1], 1, UINT_MAX
, 1);
3807 STORE_FUNCTION(cfq_slice_async_store
, &cfqd
->cfq_slice
[0], 1, UINT_MAX
, 1);
3808 STORE_FUNCTION(cfq_slice_async_rq_store
, &cfqd
->cfq_slice_async_rq
, 1,
3810 STORE_FUNCTION(cfq_low_latency_store
, &cfqd
->cfq_latency
, 0, 1, 0);
3811 STORE_FUNCTION(cfq_group_isolation_store
, &cfqd
->cfq_group_isolation
, 0, 1, 0);
3812 #undef STORE_FUNCTION
3814 #define CFQ_ATTR(name) \
3815 __ATTR(name, S_IRUGO|S_IWUSR, cfq_##name##_show, cfq_##name##_store)
3817 static struct elv_fs_entry cfq_attrs
[] = {
3819 CFQ_ATTR(fifo_expire_sync
),
3820 CFQ_ATTR(fifo_expire_async
),
3821 CFQ_ATTR(back_seek_max
),
3822 CFQ_ATTR(back_seek_penalty
),
3823 CFQ_ATTR(slice_sync
),
3824 CFQ_ATTR(slice_async
),
3825 CFQ_ATTR(slice_async_rq
),
3826 CFQ_ATTR(slice_idle
),
3827 CFQ_ATTR(low_latency
),
3828 CFQ_ATTR(group_isolation
),
3832 static struct elevator_type iosched_cfq
= {
3834 .elevator_merge_fn
= cfq_merge
,
3835 .elevator_merged_fn
= cfq_merged_request
,
3836 .elevator_merge_req_fn
= cfq_merged_requests
,
3837 .elevator_allow_merge_fn
= cfq_allow_merge
,
3838 .elevator_dispatch_fn
= cfq_dispatch_requests
,
3839 .elevator_add_req_fn
= cfq_insert_request
,
3840 .elevator_activate_req_fn
= cfq_activate_request
,
3841 .elevator_deactivate_req_fn
= cfq_deactivate_request
,
3842 .elevator_queue_empty_fn
= cfq_queue_empty
,
3843 .elevator_completed_req_fn
= cfq_completed_request
,
3844 .elevator_former_req_fn
= elv_rb_former_request
,
3845 .elevator_latter_req_fn
= elv_rb_latter_request
,
3846 .elevator_set_req_fn
= cfq_set_request
,
3847 .elevator_put_req_fn
= cfq_put_request
,
3848 .elevator_may_queue_fn
= cfq_may_queue
,
3849 .elevator_init_fn
= cfq_init_queue
,
3850 .elevator_exit_fn
= cfq_exit_queue
,
3851 .trim
= cfq_free_io_context
,
3853 .elevator_attrs
= cfq_attrs
,
3854 .elevator_name
= "cfq",
3855 .elevator_owner
= THIS_MODULE
,
3858 static int __init
cfq_init(void)
3861 * could be 0 on HZ < 1000 setups
3863 if (!cfq_slice_async
)
3864 cfq_slice_async
= 1;
3865 if (!cfq_slice_idle
)
3868 if (cfq_slab_setup())
3871 elv_register(&iosched_cfq
);
3876 static void __exit
cfq_exit(void)
3878 DECLARE_COMPLETION_ONSTACK(all_gone
);
3879 elv_unregister(&iosched_cfq
);
3880 ioc_gone
= &all_gone
;
3881 /* ioc_gone's update must be visible before reading ioc_count */
3885 * this also protects us from entering cfq_slab_kill() with
3886 * pending RCU callbacks
3888 if (elv_ioc_count_read(cfq_ioc_count
))
3889 wait_for_completion(&all_gone
);
3893 module_init(cfq_init
);
3894 module_exit(cfq_exit
);
3896 MODULE_AUTHOR("Jens Axboe");
3897 MODULE_LICENSE("GPL");
3898 MODULE_DESCRIPTION("Completely Fair Queueing IO scheduler");