* Copyright (C) 2003 Jens Axboe <axboe@kernel.dk>
*/
#include <linux/module.h>
+#include <linux/slab.h>
#include <linux/blkdev.h>
#include <linux/elevator.h>
#include <linux/jiffies.h>
#define RQ_CIC(rq) \
((struct cfq_io_context *) (rq)->elevator_private)
#define RQ_CFQQ(rq) (struct cfq_queue *) ((rq)->elevator_private2)
+#define RQ_CFQG(rq) (struct cfq_group *) ((rq)->elevator_private3)
static struct kmem_cache *cfq_pool;
static struct kmem_cache *cfq_ioc_pool;
CFQ_CFQQ_FNS(wait_busy);
#undef CFQ_CFQQ_FNS
-#ifdef CONFIG_DEBUG_CFQ_IOSCHED
+#ifdef CONFIG_CFQ_GROUP_IOSCHED
#define cfq_log_cfqq(cfqd, cfqq, fmt, args...) \
blk_add_trace_msg((cfqd)->queue, "cfq%d%c %s " fmt, (cfqq)->pid, \
cfq_cfqq_sync((cfqq)) ? 'S' : 'A', \
}
static void cfq_group_served(struct cfq_data *cfqd, struct cfq_group *cfqg,
- struct cfq_queue *cfqq, bool forced)
+ struct cfq_queue *cfqq)
{
struct cfq_rb_root *st = &cfqd->grp_service_tree;
unsigned int used_sl, charge_sl;
cfq_log_cfqg(cfqd, cfqg, "served: vt=%llu min_vt=%llu", cfqg->vdisktime,
st->min_vdisktime);
blkiocg_update_timeslice_used(&cfqg->blkg, used_sl);
- blkiocg_set_start_empty_time(&cfqg->blkg, forced);
+ blkiocg_set_start_empty_time(&cfqg->blkg);
}
#ifdef CONFIG_CFQ_GROUP_IOSCHED
unsigned int major, minor;
cfqg = cfqg_of_blkg(blkiocg_lookup_group(blkcg, key));
+ if (cfqg && !cfqg->blkg.dev && bdi->dev && dev_name(bdi->dev)) {
+ sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor);
+ cfqg->blkg.dev = MKDEV(major, minor);
+ goto done;
+ }
if (cfqg || !create)
goto done;
if (!cfqg)
goto done;
- cfqg->weight = blkcg->weight;
for_each_cfqg_st(cfqg, i, j, st)
*st = CFQ_RB_ROOT;
RB_CLEAR_NODE(&cfqg->rb_node);
- blkio_group_init(&cfqg->blkg);
/*
* Take the initial reference that will be released on destroy
sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor);
blkiocg_add_blkio_group(blkcg, &cfqg->blkg, (void *)cfqd,
MKDEV(major, minor));
+ cfqg->weight = blkcg_get_weight(blkcg, cfqg->blkg.dev);
/* Add group on cfqd list */
hlist_add_head(&cfqg->cfqd_node, &cfqd->cfqg_list);
return cfqg;
}
+static inline struct cfq_group *cfq_ref_get_cfqg(struct cfq_group *cfqg)
+{
+ atomic_inc(&cfqg->ref);
+ return cfqg;
+}
+
static void cfq_link_cfqq_cfqg(struct cfq_queue *cfqq, struct cfq_group *cfqg)
{
/* Currently, all async queues are mapped to root group */
{
return &cfqd->root_group;
}
+
+static inline struct cfq_group *cfq_ref_get_cfqg(struct cfq_group *cfqg)
+{
+ return cfqg;
+}
+
static inline void
cfq_link_cfqq_cfqg(struct cfq_queue *cfqq, struct cfq_group *cfqg) {
cfqq->cfqg = cfqg;
{
elv_rb_del(&cfqq->sort_list, rq);
cfqq->queued[rq_is_sync(rq)]--;
- blkiocg_update_request_remove_stats(&cfqq->cfqg->blkg, rq_data_dir(rq),
+ blkiocg_update_io_remove_stats(&(RQ_CFQG(rq))->blkg, rq_data_dir(rq),
rq_is_sync(rq));
cfq_add_rq_rb(rq);
- blkiocg_update_request_add_stats(
- &cfqq->cfqg->blkg, &cfqq->cfqd->serving_group->blkg,
- rq_data_dir(rq), rq_is_sync(rq));
+ blkiocg_update_io_add_stats(&(RQ_CFQG(rq))->blkg,
+ &cfqq->cfqd->serving_group->blkg, rq_data_dir(rq),
+ rq_is_sync(rq));
}
static struct request *
cfq_del_rq_rb(rq);
cfqq->cfqd->rq_queued--;
- blkiocg_update_request_remove_stats(&cfqq->cfqg->blkg, rq_data_dir(rq),
+ blkiocg_update_io_remove_stats(&(RQ_CFQG(rq))->blkg, rq_data_dir(rq),
rq_is_sync(rq));
if (rq_is_meta(rq)) {
WARN_ON(!cfqq->meta_pending);
static void cfq_bio_merged(struct request_queue *q, struct request *req,
struct bio *bio)
{
- struct cfq_queue *cfqq = RQ_CFQQ(req);
- blkiocg_update_io_merged_stats(&cfqq->cfqg->blkg, bio_data_dir(bio),
+ blkiocg_update_io_merged_stats(&(RQ_CFQG(req))->blkg, bio_data_dir(bio),
cfq_bio_sync(bio));
}
if (cfqq->next_rq == next)
cfqq->next_rq = rq;
cfq_remove_request(next);
- blkiocg_update_io_merged_stats(&cfqq->cfqg->blkg, rq_data_dir(next),
+ blkiocg_update_io_merged_stats(&(RQ_CFQG(rq))->blkg, rq_data_dir(next),
rq_is_sync(next));
}
if (cfqq) {
cfq_log_cfqq(cfqd, cfqq, "set_active wl_prio:%d wl_type:%d",
cfqd->serving_prio, cfqd->serving_type);
- blkiocg_update_set_active_queue_stats(&cfqq->cfqg->blkg);
+ blkiocg_update_avg_queue_size_stats(&cfqq->cfqg->blkg);
cfqq->slice_start = 0;
cfqq->dispatch_start = jiffies;
cfqq->allocated_slice = 0;
*/
static void
__cfq_slice_expired(struct cfq_data *cfqd, struct cfq_queue *cfqq,
- bool timed_out, bool forced)
+ bool timed_out)
{
cfq_log_cfqq(cfqd, cfqq, "slice expired t=%d", timed_out);
cfq_log_cfqq(cfqd, cfqq, "resid=%ld", cfqq->slice_resid);
}
- cfq_group_served(cfqd, cfqq->cfqg, cfqq, forced);
+ cfq_group_served(cfqd, cfqq->cfqg, cfqq);
if (cfq_cfqq_on_rr(cfqq) && RB_EMPTY_ROOT(&cfqq->sort_list))
cfq_del_cfqq_rr(cfqd, cfqq);
}
}
-static inline void cfq_slice_expired(struct cfq_data *cfqd, bool timed_out,
- bool forced)
+static inline void cfq_slice_expired(struct cfq_data *cfqd, bool timed_out)
{
struct cfq_queue *cfqq = cfqd->active_queue;
if (cfqq)
- __cfq_slice_expired(cfqd, cfqq, timed_out, forced);
+ __cfq_slice_expired(cfqd, cfqq, timed_out);
}
/*
}
expire:
- cfq_slice_expired(cfqd, 0, false);
+ cfq_slice_expired(cfqd, 0);
new_queue:
/*
* Current queue expired. Check if we have to switch to a new
BUG_ON(!list_empty(&cfqq->fifo));
/* By default cfqq is not expired if it is empty. Do it explicitly */
- __cfq_slice_expired(cfqq->cfqd, cfqq, 0, true);
+ __cfq_slice_expired(cfqq->cfqd, cfqq, 0);
return dispatched;
}
struct cfq_queue *cfqq;
int dispatched = 0;
- while ((cfqq = cfq_get_next_queue_forced(cfqd)) != NULL)
+ /* Expire the timeslice of the current active queue first */
+ cfq_slice_expired(cfqd, 0);
+ while ((cfqq = cfq_get_next_queue_forced(cfqd)) != NULL) {
+ __cfq_set_active_queue(cfqd, cfqq);
dispatched += __cfq_forced_dispatch_cfqq(cfqq);
+ }
- cfq_slice_expired(cfqd, 0, true);
BUG_ON(cfqd->busy_queues);
cfq_log(cfqd, "forced_dispatch=%d", dispatched);
cfqq->slice_dispatch >= cfq_prio_to_maxrq(cfqd, cfqq)) ||
cfq_class_idle(cfqq))) {
cfqq->slice_end = jiffies + 1;
- cfq_slice_expired(cfqd, 0, false);
+ cfq_slice_expired(cfqd, 0);
}
cfq_log_cfqq(cfqd, cfqq, "dispatched a request");
- /*
- * This is needed since we don't exactly match the mod_timer() and
- * del_timer() calls in CFQ.
- */
- blkiocg_update_idle_time_stats(&cfqq->cfqg->blkg);
return 1;
}
orig_cfqg = cfqq->orig_cfqg;
if (unlikely(cfqd->active_queue == cfqq)) {
- __cfq_slice_expired(cfqd, cfqq, 0, false);
+ __cfq_slice_expired(cfqd, cfqq, 0);
cfq_schedule_dispatch(cfqd);
}
struct cfq_queue *__cfqq, *next;
if (unlikely(cfqq == cfqd->active_queue)) {
- __cfq_slice_expired(cfqd, cfqq, 0, false);
+ __cfq_slice_expired(cfqd, cfqq, 0);
cfq_schedule_dispatch(cfqd);
}
static void cfq_preempt_queue(struct cfq_data *cfqd, struct cfq_queue *cfqq)
{
cfq_log_cfqq(cfqd, cfqq, "preempt");
- cfq_slice_expired(cfqd, 1, false);
+ cfq_slice_expired(cfqd, 1);
/*
* Put the new queue at the front of the of the current list,
cfq_del_timer(cfqd, cfqq);
cfq_clear_cfqq_wait_request(cfqq);
__blk_run_queue(cfqd->queue);
- } else
+ } else {
+ blkiocg_update_idle_time_stats(
+ &cfqq->cfqg->blkg);
cfq_mark_cfqq_must_dispatch(cfqq);
+ }
}
} else if (cfq_should_preempt(cfqd, cfqq, rq)) {
/*
rq_set_fifo_time(rq, jiffies + cfqd->cfq_fifo_expire[rq_is_sync(rq)]);
list_add_tail(&rq->queuelist, &cfqq->fifo);
cfq_add_rq_rb(rq);
-
- blkiocg_update_request_add_stats(&cfqq->cfqg->blkg,
+ blkiocg_update_io_add_stats(&(RQ_CFQG(rq))->blkg,
&cfqd->serving_group->blkg, rq_data_dir(rq),
rq_is_sync(rq));
cfq_rq_enqueued(cfqd, cfqq, rq);
* - when there is a close cooperator
*/
if (cfq_slice_used(cfqq) || cfq_class_idle(cfqq))
- cfq_slice_expired(cfqd, 1, false);
+ cfq_slice_expired(cfqd, 1);
else if (sync && cfqq_empty &&
!cfq_close_cooperator(cfqd, cfqq)) {
cfqd->noidle_tree_requires_idle |= !rq_noidle(rq);
rq->elevator_private = NULL;
rq->elevator_private2 = NULL;
+ /* Put down rq reference on cfqg */
+ cfq_put_cfqg(RQ_CFQG(rq));
+ rq->elevator_private3 = NULL;
+
cfq_put_queue(cfqq);
}
}
rq->elevator_private = cic;
rq->elevator_private2 = cfqq;
+ rq->elevator_private3 = cfq_ref_get_cfqg(cfqq->cfqg);
return 0;
queue_fail:
cfq_clear_cfqq_deep(cfqq);
}
expire:
- cfq_slice_expired(cfqd, timed_out, false);
+ cfq_slice_expired(cfqd, timed_out);
out_kick:
cfq_schedule_dispatch(cfqd);
out_cont:
spin_lock_irq(q->queue_lock);
if (cfqd->active_queue)
- __cfq_slice_expired(cfqd, cfqd->active_queue, 0, false);
+ __cfq_slice_expired(cfqd, cfqd->active_queue, 0);
while (!list_empty(&cfqd->cic_list)) {
struct cfq_io_context *cic = list_entry(cfqd->cic_list.next,
* to make sure that cfq_put_cfqg() does not try to kfree root group
*/
atomic_set(&cfqg->ref, 1);
+ rcu_read_lock();
blkiocg_add_blkio_group(&blkio_root_cgroup, &cfqg->blkg, (void *)cfqd,
0);
+ rcu_read_unlock();
#endif
/*
* Not strictly needed (since RB_ROOT just clears the node and we
* second, in order to have larger depth for async operations.
*/
cfqd->last_delayed_sync = jiffies - HZ;
- INIT_RCU_HEAD(&cfqd->rcu);
return cfqd;
}