diff options
Diffstat (limited to 'block/bfq-iosched.c')
| -rw-r--r-- | block/bfq-iosched.c | 1490 |
1 files changed, 898 insertions, 592 deletions
diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c index 727955918563..4a8d3d96bfe4 100644 --- a/block/bfq-iosched.c +++ b/block/bfq-iosched.c @@ -117,7 +117,6 @@ #include <linux/slab.h> #include <linux/blkdev.h> #include <linux/cgroup.h> -#include <linux/elevator.h> #include <linux/ktime.h> #include <linux/rbtree.h> #include <linux/ioprio.h> @@ -127,9 +126,9 @@ #include <trace/events/block.h> +#include "elevator.h" #include "blk.h" #include "blk-mq.h" -#include "blk-mq-tag.h" #include "blk-mq-sched.h" #include "bfq-iosched.h" #include "blk-wbt.h" @@ -374,18 +373,27 @@ static const unsigned long bfq_activation_stable_merging = 600; */ static const unsigned long bfq_late_stable_merging = 600; -#define RQ_BIC(rq) icq_to_bic((rq)->elv.priv[0]) +#define RQ_BIC(rq) ((struct bfq_io_cq *)((rq)->elv.priv[0])) #define RQ_BFQQ(rq) ((rq)->elv.priv[1]) -struct bfq_queue *bic_to_bfqq(struct bfq_io_cq *bic, bool is_sync) +struct bfq_queue *bic_to_bfqq(struct bfq_io_cq *bic, bool is_sync, + unsigned int actuator_idx) { - return bic->bfqq[is_sync]; + if (is_sync) + return bic->bfqq[1][actuator_idx]; + + return bic->bfqq[0][actuator_idx]; } static void bfq_put_stable_ref(struct bfq_queue *bfqq); -void bic_set_bfqq(struct bfq_io_cq *bic, struct bfq_queue *bfqq, bool is_sync) +void bic_set_bfqq(struct bfq_io_cq *bic, + struct bfq_queue *bfqq, + bool is_sync, + unsigned int actuator_idx) { + struct bfq_queue *old_bfqq = bic->bfqq[is_sync][actuator_idx]; + /* * If bfqq != NULL, then a non-stable queue merge between * bic->bfqq and bfqq is happening here. This causes troubles @@ -399,9 +407,18 @@ void bic_set_bfqq(struct bfq_io_cq *bic, struct bfq_queue *bfqq, bool is_sync) * we cancel the stable merge if * bic->stable_merge_bfqq == bfqq. */ - bic->bfqq[is_sync] = bfqq; + struct bfq_iocq_bfqq_data *bfqq_data = &bic->bfqq_data[actuator_idx]; - if (bfqq && bic->stable_merge_bfqq == bfqq) { + /* Clear bic pointer if bfqq is detached from this bic */ + if (old_bfqq && old_bfqq->bic == bic) + old_bfqq->bic = NULL; + + if (is_sync) + bic->bfqq[1][actuator_idx] = bfqq; + else + bic->bfqq[0][actuator_idx] = bfqq; + + if (bfqq && bfqq_data->stable_merge_bfqq == bfqq) { /* * Actually, these same instructions are executed also * in bfq_setup_cooperator, in case of abort or actual @@ -410,9 +427,9 @@ void bic_set_bfqq(struct bfq_io_cq *bic, struct bfq_queue *bfqq, bool is_sync) * did so, we would nest even more complexity in this * function. */ - bfq_put_stable_ref(bic->stable_merge_bfqq); + bfq_put_stable_ref(bfqq_data->stable_merge_bfqq); - bic->stable_merge_bfqq = NULL; + bfqq_data->stable_merge_bfqq = NULL; } } @@ -433,26 +450,14 @@ static struct bfq_io_cq *icq_to_bic(struct io_cq *icq) /** * bfq_bic_lookup - search into @ioc a bic associated to @bfqd. - * @bfqd: the lookup key. - * @ioc: the io_context of the process doing I/O. * @q: the request queue. */ -static struct bfq_io_cq *bfq_bic_lookup(struct bfq_data *bfqd, - struct io_context *ioc, - struct request_queue *q) +static struct bfq_io_cq *bfq_bic_lookup(struct request_queue *q) { - if (ioc) { - unsigned long flags; - struct bfq_io_cq *icq; - - spin_lock_irqsave(&q->queue_lock, flags); - icq = icq_to_bic(ioc_lookup_icq(ioc, q)); - spin_unlock_irqrestore(&q->queue_lock, flags); - - return icq; - } + if (!current->io_context) + return NULL; - return NULL; + return icq_to_bic(ioc_lookup_icq(q)); } /* @@ -461,6 +466,8 @@ static struct bfq_io_cq *bfq_bic_lookup(struct bfq_data *bfqd, */ void bfq_schedule_dispatch(struct bfq_data *bfqd) { + lockdep_assert_held(&bfqd->lock); + if (bfqd->queued != 0) { bfq_log(bfqd, "schedule dispatch"); blk_mq_run_hw_queues(bfqd->queue, true); @@ -565,26 +572,158 @@ static struct request *bfq_choose_req(struct bfq_data *bfqd, } } +#define BFQ_LIMIT_INLINE_DEPTH 16 + +#ifdef CONFIG_BFQ_GROUP_IOSCHED +static bool bfqq_request_over_limit(struct bfq_data *bfqd, + struct bfq_io_cq *bic, blk_opf_t opf, + unsigned int act_idx, int limit) +{ + struct bfq_entity *inline_entities[BFQ_LIMIT_INLINE_DEPTH]; + struct bfq_entity **entities = inline_entities; + int alloc_depth = BFQ_LIMIT_INLINE_DEPTH; + struct bfq_sched_data *sched_data; + struct bfq_entity *entity; + struct bfq_queue *bfqq; + unsigned long wsum; + bool ret = false; + int depth; + int level; + +retry: + spin_lock_irq(&bfqd->lock); + bfqq = bic_to_bfqq(bic, op_is_sync(opf), act_idx); + if (!bfqq) + goto out; + + entity = &bfqq->entity; + if (!entity->on_st_or_in_serv) + goto out; + + /* +1 for bfqq entity, root cgroup not included */ + depth = bfqg_to_blkg(bfqq_group(bfqq))->blkcg->css.cgroup->level + 1; + if (depth > alloc_depth) { + spin_unlock_irq(&bfqd->lock); + if (entities != inline_entities) + kfree(entities); + entities = kmalloc_array(depth, sizeof(*entities), GFP_NOIO); + if (!entities) + return false; + alloc_depth = depth; + goto retry; + } + + sched_data = entity->sched_data; + /* Gather our ancestors as we need to traverse them in reverse order */ + level = 0; + for_each_entity(entity) { + /* + * If at some level entity is not even active, allow request + * queueing so that BFQ knows there's work to do and activate + * entities. + */ + if (!entity->on_st_or_in_serv) + goto out; + /* Uh, more parents than cgroup subsystem thinks? */ + if (WARN_ON_ONCE(level >= depth)) + break; + entities[level++] = entity; + } + WARN_ON_ONCE(level != depth); + for (level--; level >= 0; level--) { + entity = entities[level]; + if (level > 0) { + wsum = bfq_entity_service_tree(entity)->wsum; + } else { + int i; + /* + * For bfqq itself we take into account service trees + * of all higher priority classes and multiply their + * weights so that low prio queue from higher class + * gets more requests than high prio queue from lower + * class. + */ + wsum = 0; + for (i = 0; i <= bfqq->ioprio_class - 1; i++) { + wsum = wsum * IOPRIO_BE_NR + + sched_data->service_tree[i].wsum; + } + } + if (!wsum) + continue; + limit = DIV_ROUND_CLOSEST(limit * entity->weight, wsum); + if (entity->allocated >= limit) { + bfq_log_bfqq(bfqq->bfqd, bfqq, + "too many requests: allocated %d limit %d level %d", + entity->allocated, limit, level); + ret = true; + break; + } + } +out: + spin_unlock_irq(&bfqd->lock); + if (entities != inline_entities) + kfree(entities); + return ret; +} +#else +static bool bfqq_request_over_limit(struct bfq_data *bfqd, + struct bfq_io_cq *bic, blk_opf_t opf, + unsigned int act_idx, int limit) +{ + return false; +} +#endif + /* * Async I/O can easily starve sync I/O (both sync reads and sync * writes), by consuming all tags. Similarly, storms of sync writes, * such as those that sync(2) may trigger, can starve sync reads. * Limit depths of async I/O and sync writes so as to counter both * problems. + * + * Also if a bfq queue or its parent cgroup consume more tags than would be + * appropriate for their weight, we trim the available tag depth to 1. This + * avoids a situation where one cgroup can starve another cgroup from tags and + * thus block service differentiation among cgroups. Note that because the + * queue / cgroup already has many requests allocated and queued, this does not + * significantly affect service guarantees coming from the BFQ scheduling + * algorithm. */ -static void bfq_limit_depth(unsigned int op, struct blk_mq_alloc_data *data) +static void bfq_limit_depth(blk_opf_t opf, struct blk_mq_alloc_data *data) { struct bfq_data *bfqd = data->q->elevator->elevator_data; + struct bfq_io_cq *bic = bfq_bic_lookup(data->q); + unsigned int limit, act_idx; - if (op_is_sync(op) && !op_is_write(op)) - return; + /* Sync reads have full depth available */ + if (op_is_sync(opf) && !op_is_write(opf)) + limit = data->q->nr_requests; + else + limit = bfqd->async_depths[!!bfqd->wr_busy_queues][op_is_sync(opf)]; - data->shallow_depth = - bfqd->word_depths[!!bfqd->wr_busy_queues][op_is_sync(op)]; + for (act_idx = 0; bic && act_idx < bfqd->num_actuators; act_idx++) { + /* Fast path to check if bfqq is already allocated. */ + if (!bic_to_bfqq(bic, op_is_sync(opf), act_idx)) + continue; + + /* + * Does queue (or any parent entity) exceed number of + * requests that should be available to it? Heavily + * limit depth so that it cannot consume more + * available requests and thus starve other entities. + */ + if (bfqq_request_over_limit(bfqd, bic, opf, act_idx, limit)) { + limit = 1; + break; + } + } bfq_log(bfqd, "[%s] wr_busy %d sync %d depth %u", - __func__, bfqd->wr_busy_queues, op_is_sync(op), - data->shallow_depth); + __func__, bfqd->wr_busy_queues, op_is_sync(opf), limit); + + if (limit < data->q->nr_requests) + data->shallow_depth = limit; } static struct bfq_queue * @@ -671,7 +810,7 @@ bfq_pos_tree_add_move(struct bfq_data *bfqd, struct bfq_queue *bfqq) if (!bfqq->next_rq) return; - bfqq->pos_root = &bfq_bfqq_to_bfqg(bfqq)->rq_pos_tree; + bfqq->pos_root = &bfqq_group(bfqq)->rq_pos_tree; __bfqq = bfq_rq_pos_tree_lookup(bfqd, bfqq->pos_root, blk_rq_pos(bfqq->next_rq), &parent, &p); if (!__bfqq) { @@ -709,7 +848,7 @@ bfq_pos_tree_add_move(struct bfq_data *bfqd, struct bfq_queue *bfqq) * much easier to maintain the needed state: * 1) all active queues have the same weight, * 2) all active queues belong to the same I/O-priority class, - * 3) there are no active groups. + * 3) there is at most one active group. * In particular, the last condition is always true if hierarchical * support or the cgroups interface are not enabled, thus no state * needs to be maintained in this case. @@ -741,7 +880,7 @@ static bool bfq_asymmetric_scenario(struct bfq_data *bfqd, return varied_queue_weights || multiple_classes_busy #ifdef CONFIG_BFQ_GROUP_IOSCHED - || bfqd->num_groups_with_pending_reqs > 0 + || bfqd->num_groups_with_pending_reqs > 1 #endif ; } @@ -759,9 +898,9 @@ static bool bfq_asymmetric_scenario(struct bfq_data *bfqd, * In most scenarios, the rate at which nodes are created/destroyed * should be low too. */ -void bfq_weights_tree_add(struct bfq_data *bfqd, struct bfq_queue *bfqq, - struct rb_root_cached *root) +void bfq_weights_tree_add(struct bfq_queue *bfqq) { + struct rb_root_cached *root = &bfqq->bfqd->queue_weights_tree; struct bfq_entity *entity = &bfqq->entity; struct rb_node **new = &(root->rb_root.rb_node), *parent = NULL; bool leftmost = true; @@ -833,13 +972,14 @@ inc_counter: * See the comments to the function bfq_weights_tree_add() for considerations * about overhead. */ -void __bfq_weights_tree_remove(struct bfq_data *bfqd, - struct bfq_queue *bfqq, - struct rb_root_cached *root) +void bfq_weights_tree_remove(struct bfq_queue *bfqq) { + struct rb_root_cached *root; + if (!bfqq->weight_counter) return; + root = &bfqq->bfqd->queue_weights_tree; bfqq->weight_counter->num_active--; if (bfqq->weight_counter->num_active > 0) goto reset_entity_pointer; @@ -853,59 +993,6 @@ reset_entity_pointer: } /* - * Invoke __bfq_weights_tree_remove on bfqq and decrement the number - * of active groups for each queue's inactive parent entity. - */ -void bfq_weights_tree_remove(struct bfq_data *bfqd, - struct bfq_queue *bfqq) -{ - struct bfq_entity *entity = bfqq->entity.parent; - - for_each_entity(entity) { - struct bfq_sched_data *sd = entity->my_sched_data; - - if (sd->next_in_service || sd->in_service_entity) { - /* - * entity is still active, because either - * next_in_service or in_service_entity is not - * NULL (see the comments on the definition of - * next_in_service for details on why - * in_service_entity must be checked too). - * - * As a consequence, its parent entities are - * active as well, and thus this loop must - * stop here. - */ - break; - } - - /* - * The decrement of num_groups_with_pending_reqs is - * not performed immediately upon the deactivation of - * entity, but it is delayed to when it also happens - * that the first leaf descendant bfqq of entity gets - * all its pending requests completed. The following - * instructions perform this delayed decrement, if - * needed. See the comments on - * num_groups_with_pending_reqs for details. - */ - if (entity->in_groups_with_pending_reqs) { - entity->in_groups_with_pending_reqs = false; - bfqd->num_groups_with_pending_reqs--; - } - } - - /* - * Next function is invoked last, because it causes bfqq to be - * freed if the following holds: bfqq is not in service and - * has no dispatched request. DO NOT use bfqq after the next - * function invocation. - */ - __bfq_weights_tree_remove(bfqd, bfqq, - &bfqd->queue_weights_tree); -} - -/* * Return expired entry, or NULL to just start from scratch in rbtree. */ static struct request *bfq_check_fifo(struct bfq_queue *bfqq, @@ -920,7 +1007,7 @@ static struct request *bfq_check_fifo(struct bfq_queue *bfqq, rq = rq_entry_fifo(bfqq->fifo.next); - if (rq == last || ktime_get_ns() < rq->fifo_time) + if (rq == last || blk_time_get_ns() < rq->fifo_time) return NULL; bfq_log_bfqq(bfqq->bfqd, bfqq, "check_fifo: returned %p", rq); @@ -1009,9 +1096,6 @@ static unsigned int bfq_wr_duration(struct bfq_data *bfqd) { u64 dur; - if (bfqd->bfq_wr_max_time > 0) - return bfqd->bfq_wr_max_time; - dur = bfqd->rate_dur_prod; do_div(dur, bfqd->peak_rate); @@ -1053,36 +1137,39 @@ bfq_bfqq_resume_state(struct bfq_queue *bfqq, struct bfq_data *bfqd, { unsigned int old_wr_coeff = 1; bool busy = bfq_already_existing && bfq_bfqq_busy(bfqq); + unsigned int a_idx = bfqq->actuator_idx; + struct bfq_iocq_bfqq_data *bfqq_data = &bic->bfqq_data[a_idx]; - if (bic->saved_has_short_ttime) + if (bfqq_data->saved_has_short_ttime) bfq_mark_bfqq_has_short_ttime(bfqq); else bfq_clear_bfqq_has_short_ttime(bfqq); - if (bic->saved_IO_bound) + if (bfqq_data->saved_IO_bound) bfq_mark_bfqq_IO_bound(bfqq); else bfq_clear_bfqq_IO_bound(bfqq); - bfqq->last_serv_time_ns = bic->saved_last_serv_time_ns; - bfqq->inject_limit = bic->saved_inject_limit; - bfqq->decrease_time_jif = bic->saved_decrease_time_jif; + bfqq->last_serv_time_ns = bfqq_data->saved_last_serv_time_ns; + bfqq->inject_limit = bfqq_data->saved_inject_limit; + bfqq->decrease_time_jif = bfqq_data->saved_decrease_time_jif; - bfqq->entity.new_weight = bic->saved_weight; - bfqq->ttime = bic->saved_ttime; - bfqq->io_start_time = bic->saved_io_start_time; - bfqq->tot_idle_time = bic->saved_tot_idle_time; + bfqq->entity.new_weight = bfqq_data->saved_weight; + bfqq->ttime = bfqq_data->saved_ttime; + bfqq->io_start_time = bfqq_data->saved_io_start_time; + bfqq->tot_idle_time = bfqq_data->saved_tot_idle_time; /* * Restore weight coefficient only if low_latency is on */ if (bfqd->low_latency) { old_wr_coeff = bfqq->wr_coeff; - bfqq->wr_coeff = bic->saved_wr_coeff; + bfqq->wr_coeff = bfqq_data->saved_wr_coeff; } - bfqq->service_from_wr = bic->saved_service_from_wr; - bfqq->wr_start_at_switch_to_srt = bic->saved_wr_start_at_switch_to_srt; - bfqq->last_wr_start_finish = bic->saved_last_wr_start_finish; - bfqq->wr_cur_max_time = bic->saved_wr_cur_max_time; + bfqq->service_from_wr = bfqq_data->saved_service_from_wr; + bfqq->wr_start_at_switch_to_srt = + bfqq_data->saved_wr_start_at_switch_to_srt; + bfqq->last_wr_start_finish = bfqq_data->saved_last_wr_start_finish; + bfqq->wr_cur_max_time = bfqq_data->saved_wr_cur_max_time; if (bfqq->wr_coeff > 1 && (bfq_bfqq_in_large_burst(bfqq) || time_is_before_jiffies(bfqq->last_wr_start_finish + @@ -1113,7 +1200,8 @@ bfq_bfqq_resume_state(struct bfq_queue *bfqq, struct bfq_data *bfqd, static int bfqq_process_refs(struct bfq_queue *bfqq) { - return bfqq->ref - bfqq->allocated - bfqq->entity.on_st_or_in_serv - + return bfqq->ref - bfqq->entity.allocated - + bfqq->entity.on_st_or_in_serv - (bfqq->weight_counter != NULL) - bfqq->stable_ref; } @@ -1700,6 +1788,33 @@ static bool bfq_bfqq_higher_class_or_weight(struct bfq_queue *bfqq, return bfqq_weight > in_serv_weight; } +/* + * Get the index of the actuator that will serve bio. + */ +static unsigned int bfq_actuator_index(struct bfq_data *bfqd, struct bio *bio) +{ + unsigned int i; + sector_t end; + + /* no search needed if one or zero ranges present */ + if (bfqd->num_actuators == 1) + return 0; + + /* bio_end_sector(bio) gives the sector after the last one */ + end = bio_end_sector(bio) - 1; + + for (i = 0; i < bfqd->num_actuators; i++) { + if (end >= bfqd->sector[i] && + end < bfqd->sector[i] + bfqd->nr_sectors[i]) + return i; + } + + WARN_ONCE(true, + "bfq_actuator_index: bio sector out of ranges: end=%llu\n", + end); + return 0; +} + static bool bfq_better_to_idle(struct bfq_queue *bfqq); static void bfq_bfqq_handle_idle_busy_switch(struct bfq_data *bfqd, @@ -1716,10 +1831,12 @@ static void bfq_bfqq_handle_idle_busy_switch(struct bfq_data *bfqd, * bfq_bfqq_update_budg_for_activation for * details on the usage of the next variable. */ - arrived_in_time = ktime_get_ns() <= + arrived_in_time = blk_time_get_ns() <= bfqq->ttime.last_end_request + bfqd->bfq_slice_idle * 3; - + unsigned int act_idx = bfq_actuator_index(bfqd, rq->bio); + bool bfqq_non_merged_or_stably_merged = + bfqq->bic || RQ_BIC(rq)->bfqq_data[act_idx].stably_merged; /* * bfqq deserves to be weight-raised if: @@ -1753,9 +1870,8 @@ static void bfq_bfqq_handle_idle_busy_switch(struct bfq_data *bfqd, */ wr_or_deserves_wr = bfqd->low_latency && (bfqq->wr_coeff > 1 || - (bfq_bfqq_sync(bfqq) && - (bfqq->bic || RQ_BIC(rq)->stably_merged) && - (*interactive || soft_rt))); + (bfq_bfqq_sync(bfqq) && bfqq_non_merged_or_stably_merged && + (*interactive || soft_rt))); /* * Using the last flag, update budget and check whether bfqq @@ -1813,7 +1929,7 @@ static void bfq_bfqq_handle_idle_busy_switch(struct bfq_data *bfqd, bfqq->service_from_backlogged = 0; bfq_clear_bfqq_softrt_update(bfqq); - bfq_add_bfqq_busy(bfqd, bfqq); + bfq_add_bfqq_busy(bfqq); /* * Expire in-service queue if preemption may be needed for @@ -1982,20 +2098,19 @@ static void bfq_update_io_intensity(struct bfq_queue *bfqq, u64 now_ns) * aspect, see the comments on the choice of the queue for injection * in bfq_select_queue(). * - * Turning back to the detection of a waker queue, a queue Q is deemed - * as a waker queue for bfqq if, for three consecutive times, bfqq - * happens to become non empty right after a request of Q has been - * completed. In this respect, even if bfqq is empty, we do not check - * for a waker if it still has some in-flight I/O. In fact, in this - * case bfqq is actually still being served by the drive, and may - * receive new I/O on the completion of some of the in-flight - * requests. In particular, on the first time, Q is tentatively set as - * a candidate waker queue, while on the third consecutive time that Q - * is detected, the field waker_bfqq is set to Q, to confirm that Q is - * a waker queue for bfqq. These detection steps are performed only if - * bfqq has a long think time, so as to make it more likely that - * bfqq's I/O is actually being blocked by a synchronization. This - * last filter, plus the above three-times requirement, make false + * Turning back to the detection of a waker queue, a queue Q is deemed as a + * waker queue for bfqq if, for three consecutive times, bfqq happens to become + * non empty right after a request of Q has been completed within given + * timeout. In this respect, even if bfqq is empty, we do not check for a waker + * if it still has some in-flight I/O. In fact, in this case bfqq is actually + * still being served by the drive, and may receive new I/O on the completion + * of some of the in-flight requests. In particular, on the first time, Q is + * tentatively set as a candidate waker queue, while on the third consecutive + * time that Q is detected, the field waker_bfqq is set to Q, to confirm that Q + * is a waker queue for bfqq. These detection steps are performed only if bfqq + * has a long think time, so as to make it more likely that bfqq's I/O is + * actually being blocked by a synchronization. This last filter, plus the + * above three-times requirement and time limit for detection, make false * positives less likely. * * NOTE @@ -2019,16 +2134,26 @@ static void bfq_update_io_intensity(struct bfq_queue *bfqq, u64 now_ns) static void bfq_check_waker(struct bfq_data *bfqd, struct bfq_queue *bfqq, u64 now_ns) { + char waker_name[MAX_BFQQ_NAME_LENGTH]; + if (!bfqd->last_completed_rq_bfqq || bfqd->last_completed_rq_bfqq == bfqq || bfq_bfqq_has_short_ttime(bfqq) || - bfqq->dispatched > 0 || now_ns - bfqd->last_completion >= 4 * NSEC_PER_MSEC || - bfqd->last_completed_rq_bfqq == bfqq->waker_bfqq) + bfqd->last_completed_rq_bfqq == &bfqd->oom_bfqq || + bfqq == &bfqd->oom_bfqq) return; + /* + * We reset waker detection logic also if too much time has passed + * since the first detection. If wakeups are rare, pointless idling + * doesn't hurt throughput that much. The condition below makes sure + * we do not uselessly idle blocking waker in more than 1/64 cases. + */ if (bfqd->last_completed_rq_bfqq != - bfqq->tentative_waker_bfqq) { + bfqq->tentative_waker_bfqq || + now_ns > bfqq->waker_detection_started + + 128 * (u64)bfqd->bfq_slice_idle) { /* * First synchronization detected with a * candidate waker queue, or with a different @@ -2037,12 +2162,19 @@ static void bfq_check_waker(struct bfq_data *bfqd, struct bfq_queue *bfqq, bfqq->tentative_waker_bfqq = bfqd->last_completed_rq_bfqq; bfqq->num_waker_detections = 1; + bfqq->waker_detection_started = now_ns; + bfq_bfqq_name(bfqq->tentative_waker_bfqq, waker_name, + MAX_BFQQ_NAME_LENGTH); + bfq_log_bfqq(bfqd, bfqq, "set tentative waker %s", waker_name); } else /* Same tentative waker queue detected again */ bfqq->num_waker_detections++; if (bfqq->num_waker_detections == 3) { bfqq->waker_bfqq = bfqd->last_completed_rq_bfqq; bfqq->tentative_waker_bfqq = NULL; + bfq_bfqq_name(bfqq->waker_bfqq, waker_name, + MAX_BFQQ_NAME_LENGTH); + bfq_log_bfqq(bfqd, bfqq, "set waker %s", waker_name); /* * If the waker queue disappears, then @@ -2078,13 +2210,17 @@ static void bfq_add_request(struct request *rq) struct request *next_rq, *prev; unsigned int old_wr_coeff = bfqq->wr_coeff; bool interactive = false; - u64 now_ns = ktime_get_ns(); + u64 now_ns = blk_time_get_ns(); bfq_log_bfqq(bfqd, bfqq, "add_request %d", rq_is_sync(rq)); bfqq->queued[rq_is_sync(rq)]++; - bfqd->queued++; + /* + * Updating of 'bfqd->queued' is protected by 'bfqd->lock', however, it + * may be read without holding the lock in bfq_has_work(). + */ + WRITE_ONCE(bfqd->queued, bfqd->queued + 1); - if (RB_EMPTY_ROOT(&bfqq->sort_list) && bfq_bfqq_sync(bfqq)) { + if (bfq_bfqq_sync(bfqq) && RQ_BIC(rq)->requests <= 1) { bfq_check_waker(bfqd, bfqq, now_ns); /* @@ -2123,12 +2259,12 @@ static void bfq_add_request(struct request *rq) * elapsed. */ if (bfqq == bfqd->in_service_queue && - (bfqd->rq_in_driver == 0 || + (bfqd->tot_rq_in_driver == 0 || (bfqq->last_serv_time_ns > 0 && - bfqd->rqs_injected && bfqd->rq_in_driver > 0)) && + bfqd->rqs_injected && bfqd->tot_rq_in_driver > 0)) && time_is_before_eq_jiffies(bfqq->decrease_time_jif + msecs_to_jiffies(10))) { - bfqd->last_empty_occupied_ns = ktime_get_ns(); + bfqd->last_empty_occupied_ns = blk_time_get_ns(); /* * Start the state machine for measuring the * total service time of rq: setting @@ -2149,7 +2285,7 @@ static void bfq_add_request(struct request *rq) * will be set in case injection is performed * on bfqq before rq is completed). */ - if (bfqd->rq_in_driver == 0) + if (bfqd->tot_rq_in_driver == 0) bfqd->rqs_injected = false; } } @@ -2243,22 +2379,6 @@ static sector_t get_sdist(sector_t last_pos, struct request *rq) return 0; } -#if 0 /* Still not clear if we can do without next two functions */ -static void bfq_activate_request(struct request_queue *q, struct request *rq) -{ - struct bfq_data *bfqd = q->elevator->elevator_data; - - bfqd->rq_in_driver++; -} - -static void bfq_deactivate_request(struct request_queue *q, struct request *rq) -{ - struct bfq_data *bfqd = q->elevator->elevator_data; - - bfqd->rq_in_driver--; -} -#endif - static void bfq_remove_request(struct request_queue *q, struct request *rq) { @@ -2274,7 +2394,11 @@ static void bfq_remove_request(struct request_queue *q, if (rq->queuelist.prev != &rq->queuelist) list_del_init(&rq->queuelist); bfqq->queued[sync]--; - bfqd->queued--; + /* + * Updating of 'bfqd->queued' is protected by 'bfqd->lock', however, it + * may be read without holding the lock in bfq_has_work(). + */ + WRITE_ONCE(bfqd->queued, bfqd->queued - 1); elv_rb_del(&bfqq->sort_list, rq); elv_rqhash_del(q, rq); @@ -2285,7 +2409,7 @@ static void bfq_remove_request(struct request_queue *q, bfqq->next_rq = NULL; if (bfq_bfqq_busy(bfqq) && bfqq != bfqd->in_service_queue) { - bfq_del_bfqq_busy(bfqd, bfqq, false); + bfq_del_bfqq_busy(bfqq, false); /* * bfqq emptied. In normal operation, when * bfqq is empty, bfqq->entity.service and @@ -2324,23 +2448,24 @@ static bool bfq_bio_merge(struct request_queue *q, struct bio *bio, unsigned int nr_segs) { struct bfq_data *bfqd = q->elevator->elevator_data; + struct bfq_io_cq *bic = bfq_bic_lookup(q); struct request *free = NULL; - /* - * bfq_bic_lookup grabs the queue_lock: invoke it now and - * store its return value for later use, to avoid nesting - * queue_lock inside the bfqd->lock. We assume that the bic - * returned by bfq_bic_lookup does not go away before - * bfqd->lock is taken. - */ - struct bfq_io_cq *bic = bfq_bic_lookup(bfqd, current->io_context, q); bool ret; spin_lock_irq(&bfqd->lock); - if (bic) - bfqd->bio_bfqq = bic_to_bfqq(bic, op_is_sync(bio->bi_opf)); - else + if (bic) { + /* + * Make sure cgroup info is uptodate for current process before + * considering the merge. + */ + bfq_bic_update_cgroup(bic, bio); + + bfqd->bio_bfqq = bic_to_bfqq(bic, op_is_sync(bio->bi_opf), + bfq_actuator_index(bfqd, bio)); + } else { bfqd->bio_bfqq = NULL; + } bfqd->bio_bic = bic; ret = blk_mq_sched_try_merge(q, bio, nr_segs, &free); @@ -2361,14 +2486,15 @@ static int bfq_request_merge(struct request_queue *q, struct request **req, __rq = bfq_find_rq_fmerge(bfqd, bio, q); if (__rq && elv_bio_merge_ok(__rq, bio)) { *req = __rq; + + if (blk_discard_mergable(__rq)) + return ELEVATOR_DISCARD_MERGE; return ELEVATOR_FRONT_MERGE; } return ELEVATOR_NO_MERGE; } -static struct bfq_queue *bfq_init_rq(struct request *rq); - static void bfq_request_merged(struct request_queue *q, struct request *req, enum elv_merge type) { @@ -2377,7 +2503,7 @@ static void bfq_request_merged(struct request_queue *q, struct request *req, blk_rq_pos(req) < blk_rq_pos(container_of(rb_prev(&req->rb_node), struct request, rb_node))) { - struct bfq_queue *bfqq = bfq_init_rq(req); + struct bfq_queue *bfqq = RQ_BFQQ(req); struct bfq_data *bfqd; struct request *prev, *next_rq; @@ -2429,8 +2555,8 @@ static void bfq_request_merged(struct request_queue *q, struct request *req, static void bfq_requests_merged(struct request_queue *q, struct request *rq, struct request *next) { - struct bfq_queue *bfqq = bfq_init_rq(rq), - *next_bfqq = bfq_init_rq(next); + struct bfq_queue *bfqq = RQ_BFQQ(rq), + *next_bfqq = RQ_BFQQ(next); if (!bfqq) goto remove; @@ -2502,24 +2628,29 @@ static void bfq_bfqq_end_wr(struct bfq_queue *bfqq) void bfq_end_wr_async_queues(struct bfq_data *bfqd, struct bfq_group *bfqg) { - int i, j; + int i, j, k; - for (i = 0; i < 2; i++) - for (j = 0; j < IOPRIO_BE_NR; j++) - if (bfqg->async_bfqq[i][j]) - bfq_bfqq_end_wr(bfqg->async_bfqq[i][j]); - if (bfqg->async_idle_bfqq) - bfq_bfqq_end_wr(bfqg->async_idle_bfqq); + for (k = 0; k < bfqd->num_actuators; k++) { + for (i = 0; i < 2; i++) + for (j = 0; j < IOPRIO_NR_LEVELS; j++) + if (bfqg->async_bfqq[i][j][k]) + bfq_bfqq_end_wr(bfqg->async_bfqq[i][j][k]); + if (bfqg->async_idle_bfqq[k]) + bfq_bfqq_end_wr(bfqg->async_idle_bfqq[k]); + } } static void bfq_end_wr(struct bfq_data *bfqd) { struct bfq_queue *bfqq; + int i; spin_lock_irq(&bfqd->lock); - list_for_each_entry(bfqq, &bfqd->active_list, bfqq_list) - bfq_bfqq_end_wr(bfqq); + for (i = 0; i < bfqd->num_actuators; i++) { + list_for_each_entry(bfqq, &bfqd->active_list[i], bfqq_list) + bfq_bfqq_end_wr(bfqq); + } list_for_each_entry(bfqq, &bfqd->idle_list, bfqq_list) bfq_bfqq_end_wr(bfqq); bfq_end_wr_async(bfqd); @@ -2546,7 +2677,7 @@ static struct bfq_queue *bfqq_find_close(struct bfq_data *bfqd, struct bfq_queue *bfqq, sector_t sector) { - struct rb_root *root = &bfq_bfqq_to_bfqg(bfqq)->rq_pos_tree; + struct rb_root *root = &bfqq_group(bfqq)->rq_pos_tree; struct rb_node *parent, *node; struct bfq_queue *__bfqq; @@ -2635,6 +2766,14 @@ bfq_setup_merge(struct bfq_queue *bfqq, struct bfq_queue *new_bfqq) if (process_refs == 0 || new_process_refs == 0) return NULL; + /* + * Make sure merged queues belong to the same parent. Parents could + * have changed since the time we decided the two queues are suitable + * for merging. + */ + if (new_bfqq->entity.parent != bfqq->entity.parent) + return NULL; + bfq_log_bfqq(bfqq->bfqd, bfqq, "scheduling merge with queue %d", new_bfqq->pid); @@ -2659,6 +2798,15 @@ bfq_setup_merge(struct bfq_queue *bfqq, struct bfq_queue *new_bfqq) * are likely to increase the throughput. */ bfqq->new_bfqq = new_bfqq; + /* + * The above assignment schedules the following redirections: + * each time some I/O for bfqq arrives, the process that + * generated that I/O is disassociated from bfqq and + * associated with new_bfqq. Here we increases new_bfqq->ref + * in advance, adding the number of processes that are + * expected to be associated with new_bfqq as they happen to + * issue I/O. + */ new_bfqq->ref += process_refs; return new_bfqq; } @@ -2695,6 +2843,40 @@ static bool bfq_may_be_close_cooperator(struct bfq_queue *bfqq, static bool idling_boosts_thr_without_issues(struct bfq_data *bfqd, struct bfq_queue *bfqq); +static struct bfq_queue * +bfq_setup_stable_merge(struct bfq_data *bfqd, struct bfq_queue *bfqq, + struct bfq_queue *stable_merge_bfqq, + struct bfq_iocq_bfqq_data *bfqq_data) +{ + int proc_ref = min(bfqq_process_refs(bfqq), + bfqq_process_refs(stable_merge_bfqq)); + struct bfq_queue *new_bfqq = NULL; + + bfqq_data->stable_merge_bfqq = NULL; + if (idling_boosts_thr_without_issues(bfqd, bfqq) || proc_ref == 0) + goto out; + + /* next function will take at least one ref */ + new_bfqq = bfq_setup_merge(bfqq, stable_merge_bfqq); + + if (new_bfqq) { + bfqq_data->stably_merged = true; + if (new_bfqq->bic) { + unsigned int new_a_idx = new_bfqq->actuator_idx; + struct bfq_iocq_bfqq_data *new_bfqq_data = + &new_bfqq->bic->bfqq_data[new_a_idx]; + + new_bfqq_data->stably_merged = true; + } + } + +out: + /* deschedule stable merge, because done or aborted here */ + bfq_put_stable_ref(stable_merge_bfqq); + + return new_bfqq; +} + /* * Attempt to schedule a merge of bfqq with the currently in-service * queue or with a close queue among the scheduled queues. Return @@ -2720,6 +2902,16 @@ bfq_setup_cooperator(struct bfq_data *bfqd, struct bfq_queue *bfqq, void *io_struct, bool request, struct bfq_io_cq *bic) { struct bfq_queue *in_service_bfqq, *new_bfqq; + unsigned int a_idx = bfqq->actuator_idx; + struct bfq_iocq_bfqq_data *bfqq_data = &bic->bfqq_data[a_idx]; + + /* if a merge has already been setup, then proceed with that first */ + new_bfqq = bfqq->new_bfqq; + if (new_bfqq) { + while (new_bfqq->new_bfqq) + new_bfqq = new_bfqq->new_bfqq; + return new_bfqq; + } /* * Check delayed stable merge for rotational or non-queueing @@ -2737,34 +2929,18 @@ bfq_setup_cooperator(struct bfq_data *bfqd, struct bfq_queue *bfqq, * stable merging) also if bic is associated with a * sync queue, but this bfqq is async */ - if (bfq_bfqq_sync(bfqq) && bic->stable_merge_bfqq && + if (bfq_bfqq_sync(bfqq) && bfqq_data->stable_merge_bfqq && !bfq_bfqq_just_created(bfqq) && time_is_before_jiffies(bfqq->split_time + msecs_to_jiffies(bfq_late_stable_merging)) && time_is_before_jiffies(bfqq->creation_time + msecs_to_jiffies(bfq_late_stable_merging))) { struct bfq_queue *stable_merge_bfqq = - bic->stable_merge_bfqq; - int proc_ref = min(bfqq_process_refs(bfqq), - bfqq_process_refs(stable_merge_bfqq)); - - /* deschedule stable merge, because done or aborted here */ - bfq_put_stable_ref(stable_merge_bfqq); - - bic->stable_merge_bfqq = NULL; - - if (!idling_boosts_thr_without_issues(bfqd, bfqq) && - proc_ref > 0) { - /* next function will take at least one ref */ - struct bfq_queue *new_bfqq = - bfq_setup_merge(bfqq, stable_merge_bfqq); - - bic->stably_merged = true; - if (new_bfqq && new_bfqq->bic) - new_bfqq->bic->stably_merged = true; - return new_bfqq; - } else - return NULL; + bfqq_data->stable_merge_bfqq; + + return bfq_setup_stable_merge(bfqd, bfqq, + stable_merge_bfqq, + bfqq_data); } } @@ -2822,9 +2998,6 @@ bfq_setup_cooperator(struct bfq_data *bfqd, struct bfq_queue *bfqq, if (bfq_too_late_for_merging(bfqq)) return NULL; - if (bfqq->new_bfqq) - return bfqq->new_bfqq; - if (!io_struct || unlikely(bfqq == &bfqd->oom_bfqq)) return NULL; @@ -2862,6 +3035,8 @@ bfq_setup_cooperator(struct bfq_data *bfqd, struct bfq_queue *bfqq, static void bfq_bfqq_save_state(struct bfq_queue *bfqq) { struct bfq_io_cq *bic = bfqq->bic; + unsigned int a_idx = bfqq->actuator_idx; + struct bfq_iocq_bfqq_data *bfqq_data = &bic->bfqq_data[a_idx]; /* * If !bfqq->bic, the queue is already shared or its requests @@ -2871,18 +3046,21 @@ static void bfq_bfqq_save_state(struct bfq_queue *bfqq) if (!bic) return; - bic->saved_last_serv_time_ns = bfqq->last_serv_time_ns; - bic->saved_inject_limit = bfqq->inject_limit; - bic->saved_decrease_time_jif = bfqq->decrease_time_jif; - - bic->saved_weight = bfqq->entity.orig_weight; - bic->saved_ttime = bfqq->ttime; - bic->saved_has_short_ttime = bfq_bfqq_has_short_ttime(bfqq); - bic->saved_IO_bound = bfq_bfqq_IO_bound(bfqq); - bic->saved_io_start_time = bfqq->io_start_time; - bic->saved_tot_idle_time = bfqq->tot_idle_time; - bic->saved_in_large_burst = bfq_bfqq_in_large_burst(bfqq); - bic->was_in_burst_list = !hlist_unhashed(&bfqq->burst_list_node); + bfqq_data->saved_last_serv_time_ns = bfqq->last_serv_time_ns; + bfqq_data->saved_inject_limit = bfqq->inject_limit; + bfqq_data->saved_decrease_time_jif = bfqq->decrease_time_jif; + + bfqq_data->saved_weight = bfqq->entity.orig_weight; + bfqq_data->saved_ttime = bfqq->ttime; + bfqq_data->saved_has_short_ttime = + bfq_bfqq_has_short_ttime(bfqq); + bfqq_data->saved_IO_bound = bfq_bfqq_IO_bound(bfqq); + bfqq_data->saved_io_start_time = bfqq->io_start_time; + bfqq_data->saved_tot_idle_time = bfqq->tot_idle_time; + bfqq_data->saved_in_large_burst = bfq_bfqq_in_large_burst(bfqq); + bfqq_data->was_in_burst_list = + !hlist_unhashed(&bfqq->burst_list_node); + if (unlikely(bfq_bfqq_just_created(bfqq) && !bfq_bfqq_in_large_burst(bfqq) && bfqq->bfqd->low_latency)) { @@ -2895,23 +3073,27 @@ static void bfq_bfqq_save_state(struct bfq_queue *bfqq) * to bfqq, so that to avoid that bfqq unjustly fails * to enjoy weight raising if split soon. */ - bic->saved_wr_coeff = bfqq->bfqd->bfq_wr_coeff; - bic->saved_wr_start_at_switch_to_srt = bfq_smallest_from_now(); - bic->saved_wr_cur_max_time = bfq_wr_duration(bfqq->bfqd); - bic->saved_last_wr_start_finish = jiffies; + bfqq_data->saved_wr_coeff = bfqq->bfqd->bfq_wr_coeff; + bfqq_data->saved_wr_start_at_switch_to_srt = + bfq_smallest_from_now(); + bfqq_data->saved_wr_cur_max_time = + bfq_wr_duration(bfqq->bfqd); + bfqq_data->saved_last_wr_start_finish = jiffies; } else { - bic->saved_wr_coeff = bfqq->wr_coeff; - bic->saved_wr_start_at_switch_to_srt = + bfqq_data->saved_wr_coeff = bfqq->wr_coeff; + bfqq_data->saved_wr_start_at_switch_to_srt = bfqq->wr_start_at_switch_to_srt; - bic->saved_service_from_wr = bfqq->service_from_wr; - bic->saved_last_wr_start_finish = bfqq->last_wr_start_finish; - bic->saved_wr_cur_max_time = bfqq->wr_cur_max_time; + bfqq_data->saved_service_from_wr = + bfqq->service_from_wr; + bfqq_data->saved_last_wr_start_finish = + bfqq->last_wr_start_finish; + bfqq_data->saved_wr_cur_max_time = bfqq->wr_cur_max_time; } } -static void -bfq_reassign_last_bfqq(struct bfq_queue *cur_bfqq, struct bfq_queue *new_bfqq) +void bfq_reassign_last_bfqq(struct bfq_queue *cur_bfqq, + struct bfq_queue *new_bfqq) { if (cur_bfqq->entity.parent && cur_bfqq->entity.parent->last_bfqq_created == cur_bfqq) @@ -2935,17 +3117,19 @@ void bfq_release_process_ref(struct bfq_data *bfqd, struct bfq_queue *bfqq) */ if (bfq_bfqq_busy(bfqq) && RB_EMPTY_ROOT(&bfqq->sort_list) && bfqq != bfqd->in_service_queue) - bfq_del_bfqq_busy(bfqd, bfqq, false); + bfq_del_bfqq_busy(bfqq, false); bfq_reassign_last_bfqq(bfqq, NULL); bfq_put_queue(bfqq); } -static void -bfq_merge_bfqqs(struct bfq_data *bfqd, struct bfq_io_cq *bic, - struct bfq_queue *bfqq, struct bfq_queue *new_bfqq) +static struct bfq_queue *bfq_merge_bfqqs(struct bfq_data *bfqd, + struct bfq_io_cq *bic, + struct bfq_queue *bfqq) { + struct bfq_queue *new_bfqq = bfqq->new_bfqq; + bfq_log_bfqq(bfqd, bfqq, "merging with queue %lu", (unsigned long)new_bfqq->pid); /* Save weight raising and idle window of the merged queues */ @@ -3011,7 +3195,7 @@ bfq_merge_bfqqs(struct bfq_data *bfqd, struct bfq_io_cq *bic, /* * Merge queues (that is, let bic redirect its requests to new_bfqq) */ - bic_set_bfqq(bic, new_bfqq, 1); + bic_set_bfqq(bic, new_bfqq, true, bfqq->actuator_idx); bfq_mark_bfqq_coop(new_bfqq); /* * new_bfqq now belongs to at least two bics (it is a shared queue): @@ -3039,6 +3223,8 @@ bfq_merge_bfqqs(struct bfq_data *bfqd, struct bfq_io_cq *bic, bfq_reassign_last_bfqq(bfqq, new_bfqq); bfq_release_process_ref(bfqd, bfqq); + + return new_bfqq; } static bool bfq_allow_bio_merge(struct request_queue *q, struct request *rq, @@ -3074,14 +3260,8 @@ static bool bfq_allow_bio_merge(struct request_queue *q, struct request *rq, * fulfilled, i.e., bic can be redirected to new_bfqq * and bfqq can be put. */ - bfq_merge_bfqqs(bfqd, bfqd->bio_bic, bfqq, - new_bfqq); - /* - * If we get here, bio will be queued into new_queue, - * so use new_bfqq to decide whether bio and rq can be - * merged. - */ - bfqq = new_bfqq; + while (bfqq != new_bfqq) + bfqq = bfq_merge_bfqqs(bfqd, bfqd->bio_bic, bfqq); /* * Change also bqfd->bio_bfqq, as @@ -3111,7 +3291,7 @@ static void bfq_set_budget_timeout(struct bfq_data *bfqd, else timeout_coeff = bfqq->entity.weight / bfqq->entity.orig_weight; - bfqd->last_budget_start = ktime_get(); + bfqd->last_budget_start = blk_time_get(); bfqq->budget_timeout = jiffies + bfqd->bfq_timeout * timeout_coeff; @@ -3211,7 +3391,7 @@ static void bfq_arm_slice_timer(struct bfq_data *bfqd) else if (bfqq->wr_coeff > 1) sl = max_t(u32, sl, 20ULL * NSEC_PER_MSEC); - bfqd->last_idling_start = ktime_get(); + bfqd->last_idling_start = blk_time_get(); bfqd->last_idling_start_jiffies = jiffies; hrtimer_start(&bfqd->idle_slice_timer, ns_to_ktime(sl), @@ -3250,7 +3430,7 @@ static void bfq_reset_rate_computation(struct bfq_data *bfqd, struct request *rq) { if (rq != NULL) { /* new rq dispatch now, reset accordingly */ - bfqd->last_dispatch = bfqd->first_dispatch = ktime_get_ns(); + bfqd->last_dispatch = bfqd->first_dispatch = blk_time_get_ns(); bfqd->peak_rate_samples = 1; bfqd->sequential_samples = 0; bfqd->tot_sectors_dispatched = bfqd->last_rq_max_size = @@ -3407,7 +3587,7 @@ reset_computation: */ static void bfq_update_peak_rate(struct bfq_data *bfqd, struct request *rq) { - u64 now_ns = ktime_get_ns(); + u64 now_ns = blk_time_get_ns(); if (bfqd->peak_rate_samples == 0) { /* first dispatch */ bfq_log(bfqd, "update_peak_rate: goto reset, samples %d", @@ -3429,13 +3609,13 @@ static void bfq_update_peak_rate(struct bfq_data *bfqd, struct request *rq) * - start a new observation interval with this dispatch */ if (now_ns - bfqd->last_dispatch > 100*NSEC_PER_MSEC && - bfqd->rq_in_driver == 0) + bfqd->tot_rq_in_driver == 0) goto update_rate_and_reset; /* Update sampling information */ bfqd->peak_rate_samples++; - if ((bfqd->rq_in_driver > 0 || + if ((bfqd->tot_rq_in_driver > 0 || now_ns - bfqd->last_completion < BFQ_MIN_TT) && !BFQ_RQ_SEEKY(bfqd, bfqd->last_position, rq)) bfqd->sequential_samples++; @@ -3700,10 +3880,8 @@ static bool idling_needed_for_service_guarantees(struct bfq_data *bfqd, return false; return (bfqq->wr_coeff > 1 && - (bfqd->wr_busy_queues < - tot_busy_queues || - bfqd->rq_in_driver >= - bfqq->dispatched + 4)) || + (bfqd->wr_busy_queues < tot_busy_queues || + bfqd->tot_rq_in_driver >= bfqq->dispatched + 4)) || bfq_asymmetric_scenario(bfqd, bfqq) || tot_busy_queues == 1; } @@ -3745,7 +3923,7 @@ static bool __bfq_bfqq_expire(struct bfq_data *bfqd, struct bfq_queue *bfqq, */ bfqq->budget_timeout = jiffies; - bfq_del_bfqq_busy(bfqd, bfqq, true); + bfq_del_bfqq_busy(bfqq, true); } else { bfq_requeue_bfqq(bfqd, bfqq, true); /* @@ -3969,8 +4147,7 @@ static void __bfq_bfqq_recalc_budget(struct bfq_data *bfqd, * function to evaluate the I/O speed of a process. */ static bool bfq_bfqq_is_slow(struct bfq_data *bfqd, struct bfq_queue *bfqq, - bool compensate, enum bfqq_expiration reason, - unsigned long *delta_ms) + bool compensate, unsigned long *delta_ms) { ktime_t delta_ktime; u32 delta_usecs; @@ -3982,7 +4159,7 @@ static bool bfq_bfqq_is_slow(struct bfq_data *bfqd, struct bfq_queue *bfqq, if (compensate) delta_ktime = bfqd->last_idling_start; else - delta_ktime = ktime_get(); + delta_ktime = blk_time_get(); delta_ktime = ktime_sub(delta_ktime, bfqd->last_budget_start); delta_usecs = ktime_to_us(delta_ktime); @@ -4166,7 +4343,7 @@ void bfq_bfqq_expire(struct bfq_data *bfqd, /* * Check whether the process is slow (see bfq_bfqq_is_slow). */ - slow = bfq_bfqq_is_slow(bfqd, bfqq, compensate, reason, &delta); + slow = bfq_bfqq_is_slow(bfqd, bfqq, compensate, &delta); /* * As above explained, charge slow (typically seeky) and @@ -4474,6 +4651,8 @@ bfq_choose_bfqq_for_injection(struct bfq_data *bfqd) { struct bfq_queue *bfqq, *in_serv_bfqq = bfqd->in_service_queue; unsigned int limit = in_serv_bfqq->inject_limit; + int i; + /* * If * - bfqq is not weight-raised and therefore does not carry @@ -4505,7 +4684,7 @@ bfq_choose_bfqq_for_injection(struct bfq_data *bfqd) ) limit = 1; - if (bfqd->rq_in_driver >= limit) + if (bfqd->tot_rq_in_driver >= limit) return NULL; /* @@ -4520,11 +4699,12 @@ bfq_choose_bfqq_for_injection(struct bfq_data *bfqd) * (and re-added only if it gets new requests, but then it * is assigned again enough budget for its new backlog). */ - list_for_each_entry(bfqq, &bfqd->active_list, bfqq_list) - if (!RB_EMPTY_ROOT(&bfqq->sort_list) && - (in_serv_always_inject || bfqq->wr_coeff > 1) && - bfq_serv_to_charge(bfqq->next_rq, bfqq) <= - bfq_bfqq_budget_left(bfqq)) { + for (i = 0; i < bfqd->num_actuators; i++) { + list_for_each_entry(bfqq, &bfqd->active_list[i], bfqq_list) + if (!RB_EMPTY_ROOT(&bfqq->sort_list) && + (in_serv_always_inject || bfqq->wr_coeff > 1) && + bfq_serv_to_charge(bfqq->next_rq, bfqq) <= + bfq_bfqq_budget_left(bfqq)) { /* * Allow for only one large in-flight request * on non-rotational devices, for the @@ -4544,27 +4724,80 @@ bfq_choose_bfqq_for_injection(struct bfq_data *bfqd) */ if (blk_queue_nonrot(bfqd->queue) && blk_rq_sectors(bfqq->next_rq) >= - BFQQ_SECT_THR_NONROT) - limit = min_t(unsigned int, 1, limit); - else - limit = in_serv_bfqq->inject_limit; - - if (bfqd->rq_in_driver < limit) { + BFQQ_SECT_THR_NONROT && + bfqd->tot_rq_in_driver >= 1) + continue; + else { bfqd->rqs_injected = true; return bfqq; } } + } + + return NULL; +} + +static struct bfq_queue * +bfq_find_active_bfqq_for_actuator(struct bfq_data *bfqd, int idx) +{ + struct bfq_queue *bfqq; + + if (bfqd->in_service_queue && + bfqd->in_service_queue->actuator_idx == idx) + return bfqd->in_service_queue; + + list_for_each_entry(bfqq, &bfqd->active_list[idx], bfqq_list) { + if (!RB_EMPTY_ROOT(&bfqq->sort_list) && + bfq_serv_to_charge(bfqq->next_rq, bfqq) <= + bfq_bfqq_budget_left(bfqq)) { + return bfqq; + } + } return NULL; } /* + * Perform a linear scan of each actuator, until an actuator is found + * for which the following three conditions hold: the load of the + * actuator is below the threshold (see comments on + * actuator_load_threshold for details) and lower than that of the + * next actuator (comments on this extra condition below), and there + * is a queue that contains I/O for that actuator. On success, return + * that queue. + * + * Performing a plain linear scan entails a prioritization among + * actuators. The extra condition above breaks this prioritization and + * tends to distribute injection uniformly across actuators. + */ +static struct bfq_queue * +bfq_find_bfqq_for_underused_actuator(struct bfq_data *bfqd) +{ + int i; + + for (i = 0 ; i < bfqd->num_actuators; i++) { + if (bfqd->rq_in_driver[i] < bfqd->actuator_load_threshold && + (i == bfqd->num_actuators - 1 || + bfqd->rq_in_driver[i] < bfqd->rq_in_driver[i+1])) { + struct bfq_queue *bfqq = + bfq_find_active_bfqq_for_actuator(bfqd, i); + + if (bfqq) + return bfqq; + } + } + + return NULL; +} + + +/* * Select a queue for service. If we have a current queue in service, * check whether to continue servicing it, or retrieve and set a new one. */ static struct bfq_queue *bfq_select_queue(struct bfq_data *bfqd) { - struct bfq_queue *bfqq; + struct bfq_queue *bfqq, *inject_bfqq; struct request *next_rq; enum bfqq_expiration reason = BFQQE_BUDGET_TIMEOUT; @@ -4587,6 +4820,15 @@ static struct bfq_queue *bfq_select_queue(struct bfq_data *bfqd) check_queue: /* + * If some actuator is underutilized, but the in-service + * queue does not contain I/O for that actuator, then try to + * inject I/O for that actuator. + */ + inject_bfqq = bfq_find_bfqq_for_underused_actuator(bfqd); + if (inject_bfqq && inject_bfqq != bfqq) + return inject_bfqq; + + /* * This loop is rarely executed more than once. Even when it * happens, it is much more convenient to re-execute this loop * than to return NULL and trigger a new dispatch to get a @@ -4645,11 +4887,8 @@ check_queue: */ if (bfq_bfqq_wait_request(bfqq) || (bfqq->dispatched != 0 && bfq_better_to_idle(bfqq))) { - struct bfq_queue *async_bfqq = - bfqq->bic && bfqq->bic->bfqq[0] && - bfq_bfqq_busy(bfqq->bic->bfqq[0]) && - bfqq->bic->bfqq[0]->next_rq ? - bfqq->bic->bfqq[0] : NULL; + unsigned int act_idx = bfqq->actuator_idx; + struct bfq_queue *async_bfqq = NULL; struct bfq_queue *blocked_bfqq = !hlist_empty(&bfqq->woken_list) ? container_of(bfqq->woken_list.first, @@ -4657,6 +4896,10 @@ check_queue: woken_list_node) : NULL; + if (bfqq->bic && bfqq->bic->bfqq[0][act_idx] && + bfq_bfqq_busy(bfqq->bic->bfqq[0][act_idx]) && + bfqq->bic->bfqq[0][act_idx]->next_rq) + async_bfqq = bfqq->bic->bfqq[0][act_idx]; /* * The next four mutually-exclusive ifs decide * whether to try injection, and choose the queue to @@ -4741,7 +4984,7 @@ check_queue: icq_to_bic(async_bfqq->next_rq->elv.icq) == bfqq->bic && bfq_serv_to_charge(async_bfqq->next_rq, async_bfqq) <= bfq_bfqq_budget_left(async_bfqq)) - bfqq = bfqq->bic->bfqq[0]; + bfqq = async_bfqq; else if (bfqq->waker_bfqq && bfq_bfqq_busy(bfqq->waker_bfqq) && bfqq->waker_bfqq->next_rq && @@ -4872,7 +5115,7 @@ static struct request *bfq_dispatch_rq_from_bfqq(struct bfq_data *bfqd, bfq_dispatch_remove(bfqd->queue, rq); if (bfqq != bfqd->in_service_queue) - goto return_rq; + return rq; /* * If weight raising has to terminate for bfqq, then next @@ -4892,12 +5135,9 @@ static struct request *bfq_dispatch_rq_from_bfqq(struct bfq_data *bfqd, * belongs to CLASS_IDLE and other queues are waiting for * service. */ - if (!(bfq_tot_busy_queues(bfqd) > 1 && bfq_class_idle(bfqq))) - goto return_rq; - - bfq_bfqq_expire(bfqd, bfqq, false, BFQQE_BUDGET_EXHAUSTED); + if (bfq_tot_busy_queues(bfqd) > 1 && bfq_class_idle(bfqq)) + bfq_bfqq_expire(bfqd, bfqq, false, BFQQE_BUDGET_EXHAUSTED); -return_rq: return rq; } @@ -4906,11 +5146,11 @@ static bool bfq_has_work(struct blk_mq_hw_ctx *hctx) struct bfq_data *bfqd = hctx->queue->elevator->elevator_data; /* - * Avoiding lock: a race on bfqd->busy_queues should cause at + * Avoiding lock: a race on bfqd->queued should cause at * most a call to dispatch for nothing */ return !list_empty_careful(&bfqd->dispatch) || - bfq_tot_busy_queues(bfqd) > 0; + READ_ONCE(bfqd->queued); } static struct request *__bfq_dispatch_request(struct blk_mq_hw_ctx *hctx) @@ -4940,11 +5180,11 @@ static struct request *__bfq_dispatch_request(struct blk_mq_hw_ctx *hctx) /* * We exploit the bfq_finish_requeue_request hook to - * decrement rq_in_driver, but + * decrement tot_rq_in_driver, but * bfq_finish_requeue_request will not be invoked on * this request. So, to avoid unbalance, just start - * this request, without incrementing rq_in_driver. As - * a negative consequence, rq_in_driver is deceptively + * this request, without incrementing tot_rq_in_driver. As + * a negative consequence, tot_rq_in_driver is deceptively * lower than it should be while this request is in * service. This may cause bfq_schedule_dispatch to be * invoked uselessly. @@ -4953,7 +5193,7 @@ static struct request *__bfq_dispatch_request(struct blk_mq_hw_ctx *hctx) * bfq_finish_requeue_request hook, if defined, is * probably invoked also on this request. So, by * exploiting this hook, we could 1) increment - * rq_in_driver here, and 2) decrement it in + * tot_rq_in_driver here, and 2) decrement it in * bfq_finish_requeue_request. Such a solution would * let the value of the counter be always accurate, * but it would entail using an extra interface @@ -4982,7 +5222,7 @@ static struct request *__bfq_dispatch_request(struct blk_mq_hw_ctx *hctx) * Of course, serving one request at a time may cause loss of * throughput. */ - if (bfqd->strict_guarantees && bfqd->rq_in_driver > 0) + if (bfqd->strict_guarantees && bfqd->tot_rq_in_driver > 0) goto exit; bfqq = bfq_select_queue(bfqd); @@ -4993,7 +5233,8 @@ static struct request *__bfq_dispatch_request(struct blk_mq_hw_ctx *hctx) if (rq) { inc_in_driver_start_rq: - bfqd->rq_in_driver++; + bfqd->rq_in_driver[bfqq->actuator_idx]++; + bfqd->tot_rq_in_driver++; start_rq: rq->rq_flags |= RQF_STARTED; } @@ -5058,7 +5299,7 @@ static struct request *bfq_dispatch_request(struct blk_mq_hw_ctx *hctx) struct bfq_data *bfqd = hctx->queue->elevator->elevator_data; struct request *rq; struct bfq_queue *in_serv_queue; - bool waiting_rq, idle_timer_disabled; + bool waiting_rq, idle_timer_disabled = false; spin_lock_irq(&bfqd->lock); @@ -5066,14 +5307,15 @@ static struct request *bfq_dispatch_request(struct blk_mq_hw_ctx *hctx) waiting_rq = in_serv_queue && bfq_bfqq_wait_request(in_serv_queue); rq = __bfq_dispatch_request(hctx); - - idle_timer_disabled = - waiting_rq && !bfq_bfqq_wait_request(in_serv_queue); + if (in_serv_queue == bfqd->in_service_queue) { + idle_timer_disabled = + waiting_rq && !bfq_bfqq_wait_request(in_serv_queue); + } spin_unlock_irq(&bfqd->lock); - - bfq_update_dispatch_stats(hctx->queue, rq, in_serv_queue, - idle_timer_disabled); + bfq_update_dispatch_stats(hctx->queue, rq, + idle_timer_disabled ? in_serv_queue : NULL, + idle_timer_disabled); return rq; } @@ -5091,9 +5333,7 @@ void bfq_put_queue(struct bfq_queue *bfqq) struct hlist_node *n; struct bfq_group *bfqg = bfqq_group(bfqq); - if (bfqq->bfqd) - bfq_log_bfqq(bfqq->bfqd, bfqq, "put_queue: %p %d", - bfqq, bfqq->ref); + bfq_log_bfqq(bfqq->bfqd, bfqq, "put_queue: %p %d", bfqq, bfqq->ref); bfqq->ref--; if (bfqq->ref) @@ -5157,9 +5397,13 @@ void bfq_put_queue(struct bfq_queue *bfqq) hlist_del_init(&item->woken_list_node); } - if (bfqq->bfqd && bfqq->bfqd->last_completed_rq_bfqq == bfqq) + if (bfqq->bfqd->last_completed_rq_bfqq == bfqq) bfqq->bfqd->last_completed_rq_bfqq = NULL; + WARN_ON_ONCE(!list_empty(&bfqq->fifo)); + WARN_ON_ONCE(!RB_EMPTY_ROOT(&bfqq->sort_list)); + WARN_ON_ONCE(bfqq->dispatched); + kmem_cache_free(bfq_pool, bfqq); bfqg_and_blkg_put(bfqg); } @@ -5170,7 +5414,7 @@ static void bfq_put_stable_ref(struct bfq_queue *bfqq) bfq_put_queue(bfqq); } -static void bfq_put_cooperator(struct bfq_queue *bfqq) +void bfq_put_cooperator(struct bfq_queue *bfqq) { struct bfq_queue *__bfqq, *next; @@ -5181,8 +5425,6 @@ static void bfq_put_cooperator(struct bfq_queue *bfqq) */ __bfqq = bfqq->new_bfqq; while (__bfqq) { - if (__bfqq == bfqq) - break; next = __bfqq->new_bfqq; bfq_put_queue(__bfqq); __bfqq = next; @@ -5203,49 +5445,57 @@ static void bfq_exit_bfqq(struct bfq_data *bfqd, struct bfq_queue *bfqq) bfq_release_process_ref(bfqd, bfqq); } -static void bfq_exit_icq_bfqq(struct bfq_io_cq *bic, bool is_sync) +static void bfq_exit_icq_bfqq(struct bfq_io_cq *bic, bool is_sync, + unsigned int actuator_idx) { - struct bfq_queue *bfqq = bic_to_bfqq(bic, is_sync); + struct bfq_queue *bfqq = bic_to_bfqq(bic, is_sync, actuator_idx); struct bfq_data *bfqd; if (bfqq) bfqd = bfqq->bfqd; /* NULL if scheduler already exited */ if (bfqq && bfqd) { - unsigned long flags; - - spin_lock_irqsave(&bfqd->lock, flags); - bfqq->bic = NULL; + bic_set_bfqq(bic, NULL, is_sync, actuator_idx); bfq_exit_bfqq(bfqd, bfqq); - bic_set_bfqq(bic, NULL, is_sync); - spin_unlock_irqrestore(&bfqd->lock, flags); } } -static void bfq_exit_icq(struct io_cq *icq) +static void _bfq_exit_icq(struct bfq_io_cq *bic, unsigned int num_actuators) { - struct bfq_io_cq *bic = icq_to_bic(icq); + struct bfq_iocq_bfqq_data *bfqq_data = bic->bfqq_data; + unsigned int act_idx; - if (bic->stable_merge_bfqq) { - struct bfq_data *bfqd = bic->stable_merge_bfqq->bfqd; - - /* - * bfqd is NULL if scheduler already exited, and in - * that case this is the last time bfqq is accessed. - */ - if (bfqd) { - unsigned long flags; + for (act_idx = 0; act_idx < num_actuators; act_idx++) { + if (bfqq_data[act_idx].stable_merge_bfqq) + bfq_put_stable_ref(bfqq_data[act_idx].stable_merge_bfqq); - spin_lock_irqsave(&bfqd->lock, flags); - bfq_put_stable_ref(bic->stable_merge_bfqq); - spin_unlock_irqrestore(&bfqd->lock, flags); - } else { - bfq_put_stable_ref(bic->stable_merge_bfqq); - } + bfq_exit_icq_bfqq(bic, true, act_idx); + bfq_exit_icq_bfqq(bic, false, act_idx); } +} + +static void bfq_exit_icq(struct io_cq *icq) +{ + struct bfq_io_cq *bic = icq_to_bic(icq); + struct bfq_data *bfqd = bic_to_bfqd(bic); + unsigned long flags; - bfq_exit_icq_bfqq(bic, true); - bfq_exit_icq_bfqq(bic, false); + /* + * If bfqd and thus bfqd->num_actuators is not available any + * longer, then cycle over all possible per-actuator bfqqs in + * next loop. We rely on bic being zeroed on creation, and + * therefore on its unused per-actuator fields being NULL. + * + * bfqd is NULL if scheduler already exited, and in that case + * this is the last time these queues are accessed. + */ + if (bfqd) { + spin_lock_irqsave(&bfqd->lock, flags); + _bfq_exit_icq(bic, bfqd->num_actuators); + spin_unlock_irqrestore(&bfqd->lock, flags); + } else { + _bfq_exit_icq(bic, BFQ_MAX_ACTUATORS); + } } /* @@ -5266,8 +5516,8 @@ bfq_set_next_ioprio_data(struct bfq_queue *bfqq, struct bfq_io_cq *bic) switch (ioprio_class) { default: pr_err("bdi %s: bfq: bad prio class %d\n", - bdi_dev_name(bfqq->bfqd->queue->backing_dev_info), - ioprio_class); + bdi_dev_name(bfqq->bfqd->queue->disk->bdi), + ioprio_class); fallthrough; case IOPRIO_CLASS_NONE: /* @@ -5277,23 +5527,23 @@ bfq_set_next_ioprio_data(struct bfq_queue *bfqq, struct bfq_io_cq *bic) bfqq->new_ioprio_class = task_nice_ioclass(tsk); break; case IOPRIO_CLASS_RT: - bfqq->new_ioprio = IOPRIO_PRIO_DATA(bic->ioprio); + bfqq->new_ioprio = IOPRIO_PRIO_LEVEL(bic->ioprio); bfqq->new_ioprio_class = IOPRIO_CLASS_RT; break; case IOPRIO_CLASS_BE: - bfqq->new_ioprio = IOPRIO_PRIO_DATA(bic->ioprio); + bfqq->new_ioprio = IOPRIO_PRIO_LEVEL(bic->ioprio); bfqq->new_ioprio_class = IOPRIO_CLASS_BE; break; case IOPRIO_CLASS_IDLE: bfqq->new_ioprio_class = IOPRIO_CLASS_IDLE; - bfqq->new_ioprio = 7; + bfqq->new_ioprio = IOPRIO_NR_LEVELS - 1; break; } - if (bfqq->new_ioprio >= IOPRIO_BE_NR) { + if (bfqq->new_ioprio >= IOPRIO_NR_LEVELS) { pr_crit("bfq_set_next_ioprio_data: new_ioprio %d\n", bfqq->new_ioprio); - bfqq->new_ioprio = IOPRIO_BE_NR; + bfqq->new_ioprio = IOPRIO_NR_LEVELS - 1; } bfqq->entity.new_weight = bfq_ioprio_to_weight(bfqq->new_ioprio); @@ -5322,23 +5572,27 @@ static void bfq_check_ioprio_change(struct bfq_io_cq *bic, struct bio *bio) bic->ioprio = ioprio; - bfqq = bic_to_bfqq(bic, false); + bfqq = bic_to_bfqq(bic, false, bfq_actuator_index(bfqd, bio)); if (bfqq) { - bfq_release_process_ref(bfqd, bfqq); - bfqq = bfq_get_queue(bfqd, bio, BLK_RW_ASYNC, bic, true); - bic_set_bfqq(bic, bfqq, false); + struct bfq_queue *old_bfqq = bfqq; + + bfqq = bfq_get_queue(bfqd, bio, false, bic, true); + bic_set_bfqq(bic, bfqq, false, bfq_actuator_index(bfqd, bio)); + bfq_release_process_ref(bfqd, old_bfqq); } - bfqq = bic_to_bfqq(bic, true); + bfqq = bic_to_bfqq(bic, true, bfq_actuator_index(bfqd, bio)); if (bfqq) bfq_set_next_ioprio_data(bfqq, bic); } static void bfq_init_bfqq(struct bfq_data *bfqd, struct bfq_queue *bfqq, - struct bfq_io_cq *bic, pid_t pid, int is_sync) + struct bfq_io_cq *bic, pid_t pid, int is_sync, + unsigned int act_idx) { - u64 now_ns = ktime_get_ns(); + u64 now_ns = blk_time_get_ns(); + bfqq->actuator_idx = act_idx; RB_CLEAR_NODE(&bfqq->entity.rb_node); INIT_LIST_HEAD(&bfqq->fifo); INIT_HLIST_NODE(&bfqq->burst_list_node); @@ -5398,22 +5652,24 @@ static void bfq_init_bfqq(struct bfq_data *bfqd, struct bfq_queue *bfqq, /* first request is almost certainly seeky */ bfqq->seek_history = 1; + + bfqq->decrease_time_jif = jiffies; } static struct bfq_queue **bfq_async_queue_prio(struct bfq_data *bfqd, struct bfq_group *bfqg, - int ioprio_class, int ioprio) + int ioprio_class, int ioprio, int act_idx) { switch (ioprio_class) { case IOPRIO_CLASS_RT: - return &bfqg->async_bfqq[0][ioprio]; + return &bfqg->async_bfqq[0][ioprio][act_idx]; case IOPRIO_CLASS_NONE: - ioprio = IOPRIO_NORM; + ioprio = IOPRIO_BE_NORM; fallthrough; case IOPRIO_CLASS_BE: - return &bfqg->async_bfqq[1][ioprio]; + return &bfqg->async_bfqq[1][ioprio][act_idx]; case IOPRIO_CLASS_IDLE: - return &bfqg->async_idle_bfqq; + return &bfqg->async_idle_bfqq[act_idx]; default: return NULL; } @@ -5424,6 +5680,7 @@ bfq_do_early_stable_merge(struct bfq_data *bfqd, struct bfq_queue *bfqq, struct bfq_io_cq *bic, struct bfq_queue *last_bfqq_created) { + unsigned int a_idx = last_bfqq_created->actuator_idx; struct bfq_queue *new_bfqq = bfq_setup_merge(bfqq, last_bfqq_created); @@ -5431,8 +5688,8 @@ bfq_do_early_stable_merge(struct bfq_data *bfqd, struct bfq_queue *bfqq, return bfqq; if (new_bfqq->bic) - new_bfqq->bic->stably_merged = true; - bic->stably_merged = true; + new_bfqq->bic->bfqq_data[a_idx].stably_merged = true; + bic->bfqq_data[a_idx].stably_merged = true; /* * Reusing merge functions. This implies that @@ -5441,9 +5698,7 @@ bfq_do_early_stable_merge(struct bfq_data *bfqd, struct bfq_queue *bfqq, * state before killing it. */ bfqq->bic = bic; - bfq_merge_bfqqs(bfqd, bic, bfqq, new_bfqq); - - return new_bfqq; + return bfq_merge_bfqqs(bfqd, bic, bfqq); } /* @@ -5507,9 +5762,13 @@ static struct bfq_queue *bfq_do_or_sched_stable_merge(struct bfq_data *bfqd, * it has been set already, but too long ago, then move it * forward to bfqq. Finally, move also if bfqq belongs to a * different group than last_bfqq_created, or if bfqq has a - * different ioprio or ioprio_class. If none of these - * conditions holds true, then try an early stable merge or - * schedule a delayed stable merge. + * different ioprio, ioprio_class or actuator_idx. If none of + * these conditions holds true, then try an early stable merge + * or schedule a delayed stable merge. As for the condition on + * actuator_idx, the reason is that, if queues associated with + * different actuators are merged, then control is lost on + * each actuator. Therefore some actuator may be + * underutilized, and throughput may decrease. * * A delayed merge is scheduled (instead of performing an * early merge), in case bfqq might soon prove to be more @@ -5527,7 +5786,8 @@ static struct bfq_queue *bfq_do_or_sched_stable_merge(struct bfq_data *bfqd, bfqq->creation_time) || bfqq->entity.parent != last_bfqq_created->entity.parent || bfqq->ioprio != last_bfqq_created->ioprio || - bfqq->ioprio_class != last_bfqq_created->ioprio_class) + bfqq->ioprio_class != last_bfqq_created->ioprio_class || + bfqq->actuator_idx != last_bfqq_created->actuator_idx) *source_bfqq = bfqq; else if (time_after_eq(last_bfqq_created->creation_time + bfqd->bfq_burst_interval, @@ -5557,7 +5817,8 @@ static struct bfq_queue *bfq_do_or_sched_stable_merge(struct bfq_data *bfqd, /* * Record the bfqq to merge to. */ - bic->stable_merge_bfqq = last_bfqq_created; + bic->bfqq_data[last_bfqq_created->actuator_idx].stable_merge_bfqq = + last_bfqq_created; } } @@ -5570,35 +5831,28 @@ static struct bfq_queue *bfq_get_queue(struct bfq_data *bfqd, struct bfq_io_cq *bic, bool respawn) { - const int ioprio = IOPRIO_PRIO_DATA(bic->ioprio); + const int ioprio = IOPRIO_PRIO_LEVEL(bic->ioprio); const int ioprio_class = IOPRIO_PRIO_CLASS(bic->ioprio); struct bfq_queue **async_bfqq = NULL; struct bfq_queue *bfqq; struct bfq_group *bfqg; - rcu_read_lock(); - - bfqg = bfq_find_set_group(bfqd, __bio_blkcg(bio)); - if (!bfqg) { - bfqq = &bfqd->oom_bfqq; - goto out; - } - + bfqg = bfq_bio_bfqg(bfqd, bio); if (!is_sync) { async_bfqq = bfq_async_queue_prio(bfqd, bfqg, ioprio_class, - ioprio); + ioprio, + bfq_actuator_index(bfqd, bio)); bfqq = *async_bfqq; if (bfqq) goto out; } - bfqq = kmem_cache_alloc_node(bfq_pool, - GFP_NOWAIT | __GFP_ZERO | __GFP_NOWARN, + bfqq = kmem_cache_alloc_node(bfq_pool, GFP_NOWAIT | __GFP_ZERO, bfqd->queue->node); if (bfqq) { bfq_init_bfqq(bfqd, bfqq, bic, current->pid, - is_sync); + is_sync, bfq_actuator_index(bfqd, bio)); bfq_init_entity(&bfqq->entity, bfqg); bfq_log_bfqq(bfqd, bfqq, "allocated"); } else { @@ -5629,8 +5883,6 @@ out: if (bfqq != &bfqd->oom_bfqq && is_sync && !respawn) bfqq = bfq_do_or_sched_stable_merge(bfqd, bfqq, bic); - - rcu_read_unlock(); return bfqq; } @@ -5647,7 +5899,7 @@ static void bfq_update_io_thinktime(struct bfq_data *bfqd, */ if (bfqq->dispatched || bfq_bfqq_busy(bfqq)) return; - elapsed = ktime_get_ns() - bfqq->ttime.last_end_request; + elapsed = blk_time_get_ns() - bfqq->ttime.last_end_request; elapsed = min_t(u64, elapsed, 2ULL * bfqd->bfq_slice_idle); ttime->ttime_samples = (7*ttime->ttime_samples + 256) / 8; @@ -5875,6 +6127,22 @@ static void bfq_rq_enqueued(struct bfq_data *bfqd, struct bfq_queue *bfqq, } } +static void bfqq_request_allocated(struct bfq_queue *bfqq) +{ + struct bfq_entity *entity = &bfqq->entity; + + for_each_entity(entity) + entity->allocated++; +} + +static void bfqq_request_freed(struct bfq_queue *bfqq) +{ + struct bfq_entity *entity = &bfqq->entity; + + for_each_entity(entity) + entity->allocated--; +} + /* returns true if it causes the idle timer to be disabled */ static bool __bfq_insert_request(struct bfq_data *bfqd, struct request *rq) { @@ -5884,12 +6152,13 @@ static bool __bfq_insert_request(struct bfq_data *bfqd, struct request *rq) bool waiting, idle_timer_disabled = false; if (new_bfqq) { + struct bfq_queue *old_bfqq = bfqq; /* * Release the request's reference to the old bfqq * and make sure one is taken to the shared queue. */ - new_bfqq->allocated++; - bfqq->allocated--; + bfqq_request_allocated(new_bfqq); + bfqq_request_freed(bfqq); new_bfqq->ref++; /* * If the bic associated with the process @@ -5899,18 +6168,19 @@ static bool __bfq_insert_request(struct bfq_data *bfqd, struct request *rq) * then complete the merge and redirect it to * new_bfqq. */ - if (bic_to_bfqq(RQ_BIC(rq), 1) == bfqq) - bfq_merge_bfqqs(bfqd, RQ_BIC(rq), - bfqq, new_bfqq); + if (bic_to_bfqq(RQ_BIC(rq), true, + bfq_actuator_index(bfqd, rq->bio)) == bfqq) { + while (bfqq != new_bfqq) + bfqq = bfq_merge_bfqqs(bfqd, RQ_BIC(rq), bfqq); + } - bfq_clear_bfqq_just_created(bfqq); + bfq_clear_bfqq_just_created(old_bfqq); /* * rq is about to be enqueued into new_bfqq, * release rq reference on bfqq */ - bfq_put_queue(bfqq); + bfq_put_queue(old_bfqq); rq->elv.priv[1] = new_bfqq; - bfqq = new_bfqq; } bfq_update_io_thinktime(bfqd, bfqq); @@ -5921,7 +6191,7 @@ static bool __bfq_insert_request(struct bfq_data *bfqd, struct request *rq) bfq_add_request(rq); idle_timer_disabled = waiting && !bfq_bfqq_wait_request(bfqq); - rq->fifo_time = ktime_get_ns() + bfqd->bfq_fifo_expire[rq_is_sync(rq)]; + rq->fifo_time = blk_time_get_ns() + bfqd->bfq_fifo_expire[rq_is_sync(rq)]; list_add_tail(&rq->queuelist, &bfqq->fifo); bfq_rq_enqueued(bfqd, bfqq, rq); @@ -5933,7 +6203,7 @@ static bool __bfq_insert_request(struct bfq_data *bfqd, struct request *rq) static void bfq_update_insert_stats(struct request_queue *q, struct bfq_queue *bfqq, bool idle_timer_disabled, - unsigned int cmd_flags) + blk_opf_t cmd_flags) { if (!bfqq) return; @@ -5958,17 +6228,19 @@ static void bfq_update_insert_stats(struct request_queue *q, static inline void bfq_update_insert_stats(struct request_queue *q, struct bfq_queue *bfqq, bool idle_timer_disabled, - unsigned int cmd_flags) {} + blk_opf_t cmd_flags) {} #endif /* CONFIG_BFQ_CGROUP_DEBUG */ +static struct bfq_queue *bfq_init_rq(struct request *rq); + static void bfq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, - bool at_head) + blk_insert_t flags) { struct request_queue *q = hctx->queue; struct bfq_data *bfqd = q->elevator->elevator_data; struct bfq_queue *bfqq; bool idle_timer_disabled = false; - unsigned int cmd_flags; + blk_opf_t cmd_flags; LIST_HEAD(free); #ifdef CONFIG_BFQ_GROUP_IOSCHED @@ -5976,64 +6248,19 @@ static void bfq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, bfqg_stats_update_legacy_io(q, rq); #endif spin_lock_irq(&bfqd->lock); + bfqq = bfq_init_rq(rq); if (blk_mq_sched_try_insert_merge(q, rq, &free)) { spin_unlock_irq(&bfqd->lock); blk_mq_free_requests(&free); return; } - spin_unlock_irq(&bfqd->lock); - trace_block_rq_insert(rq); - spin_lock_irq(&bfqd->lock); - bfqq = bfq_init_rq(rq); - - /* - * Reqs with at_head or passthrough flags set are to be put - * directly into dispatch list. Additional case for putting rq - * directly into the dispatch queue: the only active - * bfq_queues are bfqq and either its waker bfq_queue or one - * of its woken bfq_queues. The rationale behind this - * additional condition is as follows: - * - consider a bfq_queue, say Q1, detected as a waker of - * another bfq_queue, say Q2 - * - by definition of a waker, Q1 blocks the I/O of Q2, i.e., - * some I/O of Q1 needs to be completed for new I/O of Q2 - * to arrive. A notable example of waker is journald - * - so, Q1 and Q2 are in any respect the queues of two - * cooperating processes (or of two cooperating sets of - * processes): the goal of Q1's I/O is doing what needs to - * be done so that new Q2's I/O can finally be - * issued. Therefore, if the service of Q1's I/O is delayed, - * then Q2's I/O is delayed too. Conversely, if Q2's I/O is - * delayed, the goal of Q1's I/O is hindered. - * - as a consequence, if some I/O of Q1/Q2 arrives while - * Q2/Q1 is the only queue in service, there is absolutely - * no point in delaying the service of such an I/O. The - * only possible result is a throughput loss - * - so, when the above condition holds, the best option is to - * have the new I/O dispatched as soon as possible - * - the most effective and efficient way to attain the above - * goal is to put the new I/O directly in the dispatch - * list - * - as an additional restriction, Q1 and Q2 must be the only - * busy queues for this commit to put the I/O of Q2/Q1 in - * the dispatch list. This is necessary, because, if also - * other queues are waiting for service, then putting new - * I/O directly in the dispatch list may evidently cause a - * violation of service guarantees for the other queues - */ - if (!bfqq || - (bfqq != bfqd->in_service_queue && - bfqd->in_service_queue != NULL && - bfq_tot_busy_queues(bfqd) == 1 + bfq_bfqq_busy(bfqq) && - (bfqq->waker_bfqq == bfqd->in_service_queue || - bfqd->in_service_queue->waker_bfqq == bfqq)) || at_head) { - if (at_head) - list_add(&rq->queuelist, &bfqd->dispatch); - else - list_add_tail(&rq->queuelist, &bfqd->dispatch); + if (flags & BLK_MQ_INSERT_AT_HEAD) { + list_add(&rq->queuelist, &bfqd->dispatch); + } else if (!bfqq) { + list_add_tail(&rq->queuelist, &bfqd->dispatch); } else { idle_timer_disabled = __bfq_insert_request(bfqd, rq); /* @@ -6056,7 +6283,6 @@ static void bfq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, * merge). */ cmd_flags = rq->cmd_flags; - spin_unlock_irq(&bfqd->lock); bfq_update_insert_stats(q, bfqq, idle_timer_disabled, @@ -6064,14 +6290,15 @@ static void bfq_insert_request(struct blk_mq_hw_ctx *hctx, struct request *rq, } static void bfq_insert_requests(struct blk_mq_hw_ctx *hctx, - struct list_head *list, bool at_head) + struct list_head *list, + blk_insert_t flags) { while (!list_empty(list)) { struct request *rq; rq = list_first_entry(list, struct request, queuelist); list_del_init(&rq->queuelist); - bfq_insert_request(hctx, rq, at_head); + bfq_insert_request(hctx, rq, flags); } } @@ -6080,7 +6307,7 @@ static void bfq_update_hw_tag(struct bfq_data *bfqd) struct bfq_queue *bfqq = bfqd->in_service_queue; bfqd->max_rq_in_driver = max_t(int, bfqd->max_rq_in_driver, - bfqd->rq_in_driver); + bfqd->tot_rq_in_driver); if (bfqd->hw_tag == 1) return; @@ -6091,7 +6318,7 @@ static void bfq_update_hw_tag(struct bfq_data *bfqd) * sum is not exact, as it's not taking into account deactivated * requests. */ - if (bfqd->rq_in_driver + bfqd->queued <= BFQ_HW_QUEUE_THRESHOLD) + if (bfqd->tot_rq_in_driver + bfqd->queued <= BFQ_HW_QUEUE_THRESHOLD) return; /* @@ -6102,7 +6329,7 @@ static void bfq_update_hw_tag(struct bfq_data *bfqd) if (bfqq && bfq_bfqq_has_short_ttime(bfqq) && bfqq->dispatched + bfqq->queued[0] + bfqq->queued[1] < BFQ_HW_QUEUE_THRESHOLD && - bfqd->rq_in_driver < BFQ_HW_QUEUE_THRESHOLD) + bfqd->tot_rq_in_driver < BFQ_HW_QUEUE_THRESHOLD) return; if (bfqd->hw_tag_samples++ < BFQ_HW_QUEUE_SAMPLES) @@ -6123,7 +6350,8 @@ static void bfq_completed_request(struct bfq_queue *bfqq, struct bfq_data *bfqd) bfq_update_hw_tag(bfqd); - bfqd->rq_in_driver--; + bfqd->rq_in_driver[bfqq->actuator_idx]--; + bfqd->tot_rq_in_driver--; bfqq->dispatched--; if (!bfqq->dispatched && !bfq_bfqq_busy(bfqq)) { @@ -6135,10 +6363,11 @@ static void bfq_completed_request(struct bfq_queue *bfqq, struct bfq_data *bfqd) */ bfqq->budget_timeout = jiffies; - bfq_weights_tree_remove(bfqd, bfqq); + bfq_del_bfqq_in_groups_with_pending_reqs(bfqq); + bfq_weights_tree_remove(bfqq); } - now_ns = ktime_get_ns(); + now_ns = blk_time_get_ns(); bfqq->ttime.last_end_request = now_ns; @@ -6242,17 +6471,10 @@ static void bfq_completed_request(struct bfq_queue *bfqq, struct bfq_data *bfqd) BFQQE_NO_MORE_REQUESTS); } - if (!bfqd->rq_in_driver) + if (!bfqd->tot_rq_in_driver) bfq_schedule_dispatch(bfqd); } -static void bfq_finish_requeue_request_body(struct bfq_queue *bfqq) -{ - bfqq->allocated--; - - bfq_put_queue(bfqq); -} - /* * The processes associated with bfqq may happen to generate their * cumulative I/O at a lower rate than the rate at which the device @@ -6360,7 +6582,7 @@ static void bfq_finish_requeue_request_body(struct bfq_queue *bfqq) static void bfq_update_inject_limit(struct bfq_data *bfqd, struct bfq_queue *bfqq) { - u64 tot_time_ns = ktime_get_ns() - bfqd->last_empty_occupied_ns; + u64 tot_time_ns = blk_time_get_ns() - bfqd->last_empty_occupied_ns; unsigned int old_limit = bfqq->inject_limit; if (bfqq->last_serv_time_ns > 0 && bfqd->rqs_injected) { @@ -6380,13 +6602,13 @@ static void bfq_update_inject_limit(struct bfq_data *bfqd, * conditions to do it, or we can lower the last base value * computed. * - * NOTE: (bfqd->rq_in_driver == 1) means that there is no I/O + * NOTE: (bfqd->tot_rq_in_driver == 1) means that there is no I/O * request in flight, because this function is in the code * path that handles the completion of a request of bfqq, and, * in particular, this function is executed before - * bfqd->rq_in_driver is decremented in such a code path. + * bfqd->tot_rq_in_driver is decremented in such a code path. */ - if ((bfqq->last_serv_time_ns == 0 && bfqd->rq_in_driver == 1) || + if ((bfqq->last_serv_time_ns == 0 && bfqd->tot_rq_in_driver == 1) || tot_time_ns < bfqq->last_serv_time_ns) { if (bfqq->last_serv_time_ns == 0) { /* @@ -6396,7 +6618,7 @@ static void bfq_update_inject_limit(struct bfq_data *bfqd, bfqq->inject_limit = max_t(unsigned int, 1, old_limit); } bfqq->last_serv_time_ns = tot_time_ns; - } else if (!bfqd->rqs_injected && bfqd->rq_in_driver == 1) + } else if (!bfqd->rqs_injected && bfqd->tot_rq_in_driver == 1) /* * No I/O injected and no request still in service in * the drive: these are the exact conditions for @@ -6449,7 +6671,9 @@ static void bfq_finish_requeue_request(struct request *rq) bfq_completed_request(bfqq, bfqd); } - bfq_finish_requeue_request_body(bfqq); + bfqq_request_freed(bfqq); + bfq_put_queue(bfqq); + RQ_BIC(rq)->requests--; spin_unlock_irqrestore(&bfqd->lock, flags); /* @@ -6473,6 +6697,16 @@ static void bfq_finish_requeue_request(struct request *rq) rq->elv.priv[1] = NULL; } +static void bfq_finish_request(struct request *rq) +{ + bfq_finish_requeue_request(rq); + + if (rq->elv.icq) { + put_io_context(rq->elv.icq->ioc); + rq->elv.icq = NULL; + } +} + /* * Removes the association between the current task and bfqq, assuming * that bic points to the bfq iocontext of the task. @@ -6484,14 +6718,14 @@ bfq_split_bfqq(struct bfq_io_cq *bic, struct bfq_queue *bfqq) { bfq_log_bfqq(bfqq->bfqd, bfqq, "splitting queue"); - if (bfqq_process_refs(bfqq) == 1) { + if (bfqq_process_refs(bfqq) == 1 && !bfqq->new_bfqq) { bfqq->pid = current->pid; bfq_clear_bfqq_coop(bfqq); bfq_clear_bfqq_split_coop(bfqq); return bfqq; } - bic_set_bfqq(bic, NULL, 1); + bic_set_bfqq(bic, NULL, true, bfqq->actuator_idx); bfq_put_cooperator(bfqq); @@ -6499,13 +6733,14 @@ bfq_split_bfqq(struct bfq_io_cq *bic, struct bfq_queue *bfqq) return NULL; } -static struct bfq_queue *bfq_get_bfqq_handle_split(struct bfq_data *bfqd, - struct bfq_io_cq *bic, - struct bio *bio, - bool split, bool is_sync, - bool *new_queue) +static struct bfq_queue * +__bfq_get_bfqq_handle_split(struct bfq_data *bfqd, struct bfq_io_cq *bic, + struct bio *bio, bool split, bool is_sync, + bool *new_queue) { - struct bfq_queue *bfqq = bic_to_bfqq(bic, is_sync); + unsigned int act_idx = bfq_actuator_index(bfqd, bio); + struct bfq_queue *bfqq = bic_to_bfqq(bic, is_sync, act_idx); + struct bfq_iocq_bfqq_data *bfqq_data = &bic->bfqq_data[act_idx]; if (likely(bfqq && bfqq != &bfqd->oom_bfqq)) return bfqq; @@ -6517,14 +6752,14 @@ static struct bfq_queue *bfq_get_bfqq_handle_split(struct bfq_data *bfqd, bfq_put_queue(bfqq); bfqq = bfq_get_queue(bfqd, bio, is_sync, bic, split); - bic_set_bfqq(bic, bfqq, is_sync); + bic_set_bfqq(bic, bfqq, is_sync, act_idx); if (split && is_sync) { - if ((bic->was_in_burst_list && bfqd->large_burst) || - bic->saved_in_large_burst) + if ((bfqq_data->was_in_burst_list && bfqd->large_burst) || + bfqq_data->saved_in_large_burst) bfq_mark_bfqq_in_large_burst(bfqq); else { bfq_clear_bfqq_in_large_burst(bfqq); - if (bic->was_in_burst_list) + if (bfqq_data->was_in_burst_list) /* * If bfqq was in the current * burst list before being @@ -6570,6 +6805,8 @@ static struct bfq_queue *bfq_get_bfqq_handle_split(struct bfq_data *bfqd, */ static void bfq_prepare_request(struct request *rq) { + rq->elv.icq = ioc_find_get_icq(rq->q); + /* * Regardless of whether we have an icq attached, we have to * clear the scheduler pointers, as they might point to @@ -6578,6 +6815,92 @@ static void bfq_prepare_request(struct request *rq) rq->elv.priv[0] = rq->elv.priv[1] = NULL; } +static struct bfq_queue *bfq_waker_bfqq(struct bfq_queue *bfqq) +{ + struct bfq_queue *new_bfqq = bfqq->new_bfqq; + struct bfq_queue *waker_bfqq = bfqq->waker_bfqq; + + if (!waker_bfqq) + return NULL; + + while (new_bfqq) { + if (new_bfqq == waker_bfqq) { + /* + * If waker_bfqq is in the merge chain, and current + * is the only process, waker_bfqq can be freed. + */ + if (bfqq_process_refs(waker_bfqq) == 1) + return NULL; + + return waker_bfqq; + } + + new_bfqq = new_bfqq->new_bfqq; + } + + /* + * If waker_bfqq is not in the merge chain, and it's procress reference + * is 0, waker_bfqq can be freed. + */ + if (bfqq_process_refs(waker_bfqq) == 0) + return NULL; + + return waker_bfqq; +} + +static struct bfq_queue *bfq_get_bfqq_handle_split(struct bfq_data *bfqd, + struct bfq_io_cq *bic, + struct bio *bio, + unsigned int idx, + bool is_sync) +{ + struct bfq_queue *waker_bfqq; + struct bfq_queue *bfqq; + bool new_queue = false; + + bfqq = __bfq_get_bfqq_handle_split(bfqd, bic, bio, false, is_sync, + &new_queue); + if (unlikely(new_queue)) + return bfqq; + + /* If the queue was seeky for too long, break it apart. */ + if (!bfq_bfqq_coop(bfqq) || !bfq_bfqq_split_coop(bfqq) || + bic->bfqq_data[idx].stably_merged) + return bfqq; + + waker_bfqq = bfq_waker_bfqq(bfqq); + + /* Update bic before losing reference to bfqq */ + if (bfq_bfqq_in_large_burst(bfqq)) + bic->bfqq_data[idx].saved_in_large_burst = true; + + bfqq = bfq_split_bfqq(bic, bfqq); + if (bfqq) { + bfq_bfqq_resume_state(bfqq, bfqd, bic, true); + return bfqq; + } + + bfqq = __bfq_get_bfqq_handle_split(bfqd, bic, bio, true, is_sync, NULL); + if (unlikely(bfqq == &bfqd->oom_bfqq)) + return bfqq; + + bfq_bfqq_resume_state(bfqq, bfqd, bic, false); + bfqq->waker_bfqq = waker_bfqq; + bfqq->tentative_waker_bfqq = NULL; + + /* + * If the waker queue disappears, then new_bfqq->waker_bfqq must be + * reset. So insert new_bfqq into the + * woken_list of the waker. See + * bfq_check_waker for details. + */ + if (waker_bfqq) + hlist_add_head(&bfqq->woken_list_node, + &bfqq->waker_bfqq->woken_list); + + return bfqq; +} + /* * If needed, init rq, allocate bfq data structures associated with * rq, and increment reference counters in the destination bfq_queue @@ -6609,68 +6932,29 @@ static struct bfq_queue *bfq_init_rq(struct request *rq) struct bfq_io_cq *bic; const int is_sync = rq_is_sync(rq); struct bfq_queue *bfqq; - bool new_queue = false; - bool bfqq_already_existing = false, split = false; + unsigned int a_idx = bfq_actuator_index(bfqd, bio); if (unlikely(!rq->elv.icq)) return NULL; /* - * Assuming that elv.priv[1] is set only if everything is set + * Assuming that RQ_BFQQ(rq) is set only if everything is set * for this rq. This holds true, because this function is * invoked only for insertion or merging, and, after such * events, a request cannot be manipulated any longer before * being removed from bfq. */ - if (rq->elv.priv[1]) - return rq->elv.priv[1]; + if (RQ_BFQQ(rq)) + return RQ_BFQQ(rq); bic = icq_to_bic(rq->elv.icq); - bfq_check_ioprio_change(bic, bio); - bfq_bic_update_cgroup(bic, bio); + bfqq = bfq_get_bfqq_handle_split(bfqd, bic, bio, a_idx, is_sync); - bfqq = bfq_get_bfqq_handle_split(bfqd, bic, bio, false, is_sync, - &new_queue); - - if (likely(!new_queue)) { - /* If the queue was seeky for too long, break it apart. */ - if (bfq_bfqq_coop(bfqq) && bfq_bfqq_split_coop(bfqq) && - !bic->stably_merged) { - struct bfq_queue *old_bfqq = bfqq; - - /* Update bic before losing reference to bfqq */ - if (bfq_bfqq_in_large_burst(bfqq)) - bic->saved_in_large_burst = true; - - bfqq = bfq_split_bfqq(bic, bfqq); - split = true; - - if (!bfqq) { - bfqq = bfq_get_bfqq_handle_split(bfqd, bic, bio, - true, is_sync, - NULL); - bfqq->waker_bfqq = old_bfqq->waker_bfqq; - bfqq->tentative_waker_bfqq = NULL; - - /* - * If the waker queue disappears, then - * new_bfqq->waker_bfqq must be - * reset. So insert new_bfqq into the - * woken_list of the waker. See - * bfq_check_waker for details. - */ - if (bfqq->waker_bfqq) - hlist_add_head(&bfqq->woken_list_node, - &bfqq->waker_bfqq->woken_list); - } else - bfqq_already_existing = true; - } - } - - bfqq->allocated++; + bfqq_request_allocated(bfqq); bfqq->ref++; + bic->requests++; bfq_log_bfqq(bfqd, bfqq, "get_request %p: bfqq %p, %d", rq, bfqq, bfqq->ref); @@ -6683,18 +6967,9 @@ static struct bfq_queue *bfq_init_rq(struct request *rq) * addition, if the queue has also just been split, we have to * resume its state. */ - if (likely(bfqq != &bfqd->oom_bfqq) && bfqq_process_refs(bfqq) == 1) { + if (likely(bfqq != &bfqd->oom_bfqq) && !bfqq->new_bfqq && + bfqq_process_refs(bfqq) == 1) bfqq->bic = bic; - if (split) { - /* - * The queue has just been split from a shared - * queue: restore the idle window and the - * possible weight raising period. - */ - bfq_bfqq_resume_state(bfqq, bfqd, bic, - bfqq_already_existing); - } - } /* * Consider bfqq as possibly belonging to a burst of newly @@ -6767,8 +7042,8 @@ bfq_idle_slice_timer_body(struct bfq_data *bfqd, struct bfq_queue *bfqq) bfq_bfqq_expire(bfqd, bfqq, true, reason); schedule_dispatch: - spin_unlock_irqrestore(&bfqd->lock, flags); bfq_schedule_dispatch(bfqd); + spin_unlock_irqrestore(&bfqd->lock, flags); } /* @@ -6819,23 +7094,25 @@ static void __bfq_put_async_bfqq(struct bfq_data *bfqd, */ void bfq_put_async_queues(struct bfq_data *bfqd, struct bfq_group *bfqg) { - int i, j; + int i, j, k; - for (i = 0; i < 2; i++) - for (j = 0; j < IOPRIO_BE_NR; j++) - __bfq_put_async_bfqq(bfqd, &bfqg->async_bfqq[i][j]); + for (k = 0; k < bfqd->num_actuators; k++) { + for (i = 0; i < 2; i++) + for (j = 0; j < IOPRIO_NR_LEVELS; j++) + __bfq_put_async_bfqq(bfqd, &bfqg->async_bfqq[i][j][k]); - __bfq_put_async_bfqq(bfqd, &bfqg->async_idle_bfqq); + __bfq_put_async_bfqq(bfqd, &bfqg->async_idle_bfqq[k]); + } } /* * See the comments on bfq_limit_depth for the purpose of * the depths set in the function. Return minimum shallow depth we'll use. */ -static unsigned int bfq_update_depths(struct bfq_data *bfqd, - struct sbitmap_queue *bt) +static void bfq_depth_updated(struct request_queue *q) { - unsigned int i, j, min_shallow = UINT_MAX; + struct bfq_data *bfqd = q->elevator->elevator_data; + unsigned int nr_requests = q->nr_requests; /* * In-word depths if no bfq_queue is being weight-raised: @@ -6848,13 +7125,13 @@ static unsigned int bfq_update_depths(struct bfq_data *bfqd, * limit 'something'. */ /* no more than 50% of tags for async I/O */ - bfqd->word_depths[0][0] = max((1U << bt->sb.shift) >> 1, 1U); + bfqd->async_depths[0][0] = max(nr_requests >> 1, 1U); /* * no more than 75% of tags for sync writes (25% extra tags * w.r.t. async I/O, to prevent async I/O from starving sync * writes) */ - bfqd->word_depths[0][1] = max(((1U << bt->sb.shift) * 3) >> 2, 1U); + bfqd->async_depths[0][1] = max((nr_requests * 3) >> 2, 1U); /* * In-word depths in case some bfq_queue is being weight- @@ -6864,37 +7141,18 @@ static unsigned int bfq_update_depths(struct bfq_data *bfqd, * shortage. */ /* no more than ~18% of tags for async I/O */ - bfqd->word_depths[1][0] = max(((1U << bt->sb.shift) * 3) >> 4, 1U); + bfqd->async_depths[1][0] = max((nr_requests * 3) >> 4, 1U); /* no more than ~37% of tags for sync writes (~20% extra tags) */ - bfqd->word_depths[1][1] = max(((1U << bt->sb.shift) * 6) >> 4, 1U); - - for (i = 0; i < 2; i++) - for (j = 0; j < 2; j++) - min_shallow = min(min_shallow, bfqd->word_depths[i][j]); + bfqd->async_depths[1][1] = max((nr_requests * 6) >> 4, 1U); - return min_shallow; -} - -static void bfq_depth_updated(struct blk_mq_hw_ctx *hctx) -{ - struct bfq_data *bfqd = hctx->queue->elevator->elevator_data; - struct blk_mq_tags *tags = hctx->sched_tags; - unsigned int min_shallow; - - min_shallow = bfq_update_depths(bfqd, tags->bitmap_tags); - sbitmap_queue_min_shallow_depth(tags->bitmap_tags, min_shallow); -} - -static int bfq_init_hctx(struct blk_mq_hw_ctx *hctx, unsigned int index) -{ - bfq_depth_updated(hctx); - return 0; + blk_mq_set_min_shallow_depth(q, 1); } static void bfq_exit_queue(struct elevator_queue *e) { struct bfq_data *bfqd = e->elevator_data; struct bfq_queue *bfqq, *n; + unsigned int actuator; hrtimer_cancel(&bfqd->idle_slice_timer); @@ -6903,13 +7161,17 @@ static void bfq_exit_queue(struct elevator_queue *e) bfq_deactivate_bfqq(bfqd, bfqq, false, false); spin_unlock_irq(&bfqd->lock); + for (actuator = 0; actuator < bfqd->num_actuators; actuator++) + WARN_ON_ONCE(bfqd->rq_in_driver[actuator]); + WARN_ON_ONCE(bfqd->tot_rq_in_driver); + hrtimer_cancel(&bfqd->idle_slice_timer); /* release oom-queue reference to root group */ bfqg_and_blkg_put(bfqd->root_group); #ifdef CONFIG_BFQ_GROUP_IOSCHED - blkcg_deactivate_policy(bfqd->queue, &blkcg_policy_bfq); + blkcg_deactivate_policy(bfqd->queue->disk, &blkcg_policy_bfq); #else spin_lock_irq(&bfqd->lock); bfq_put_async_queues(bfqd, bfqd->root_group); @@ -6917,6 +7179,10 @@ static void bfq_exit_queue(struct elevator_queue *e) spin_unlock_irq(&bfqd->lock); #endif + blk_stat_disable_accounting(bfqd->queue); + blk_queue_flag_clear(QUEUE_FLAG_DISABLE_WBT_DEF, bfqd->queue); + set_bit(ELEVATOR_FLAG_ENABLE_WBT_ON_EXIT, &e->flags); + kfree(bfqd); } @@ -6936,20 +7202,16 @@ static void bfq_init_root_group(struct bfq_group *root_group, root_group->sched_data.bfq_class_idle_last_service = jiffies; } -static int bfq_init_queue(struct request_queue *q, struct elevator_type *e) +static int bfq_init_queue(struct request_queue *q, struct elevator_queue *eq) { struct bfq_data *bfqd; - struct elevator_queue *eq; - - eq = elevator_alloc(q, e); - if (!eq) - return -ENOMEM; + unsigned int i; + struct blk_independent_access_ranges *ia_ranges = q->disk->ia_ranges; bfqd = kzalloc_node(sizeof(*bfqd), GFP_KERNEL, q->node); - if (!bfqd) { - kobject_put(&eq->kobj); + if (!bfqd) return -ENOMEM; - } + eq->elevator_data = bfqd; spin_lock_irq(&q->queue_lock); @@ -6960,8 +7222,10 @@ static int bfq_init_queue(struct request_queue *q, struct elevator_type *e) * Our fallback bfqq if bfq_find_alloc_queue() runs into OOM issues. * Grab a permanent reference to it, so that the normal code flow * will not attempt to free it. + * Set zero as actuator index: we will pretend that + * all I/O requests are for the same actuator. */ - bfq_init_bfqq(bfqd, &bfqd->oom_bfqq, NULL, 1, 0); + bfq_init_bfqq(bfqd, &bfqd->oom_bfqq, NULL, 1, 0, 0); bfqd->oom_bfqq.ref++; bfqd->oom_bfqq.new_ioprio = BFQ_DEFAULT_QUEUE_IOPRIO; bfqd->oom_bfqq.new_ioprio_class = IOPRIO_CLASS_BE; @@ -6980,16 +7244,51 @@ static int bfq_init_queue(struct request_queue *q, struct elevator_type *e) bfqd->queue = q; + bfqd->num_actuators = 1; + /* + * If the disk supports multiple actuators, copy independent + * access ranges from the request queue structure. + */ + spin_lock_irq(&q->queue_lock); + if (ia_ranges) { + /* + * Check if the disk ia_ranges size exceeds the current bfq + * actuator limit. + */ + if (ia_ranges->nr_ia_ranges > BFQ_MAX_ACTUATORS) { + pr_crit("nr_ia_ranges higher than act limit: iars=%d, max=%d.\n", + ia_ranges->nr_ia_ranges, BFQ_MAX_ACTUATORS); + pr_crit("Falling back to single actuator mode.\n"); + } else { + bfqd->num_actuators = ia_ranges->nr_ia_ranges; + + for (i = 0; i < bfqd->num_actuators; i++) { + bfqd->sector[i] = ia_ranges->ia_range[i].sector; + bfqd->nr_sectors[i] = + ia_ranges->ia_range[i].nr_sectors; + } + } + } + + /* Otherwise use single-actuator dev info */ + if (bfqd->num_actuators == 1) { + bfqd->sector[0] = 0; + bfqd->nr_sectors[0] = get_capacity(q->disk); + } + spin_unlock_irq(&q->queue_lock); + INIT_LIST_HEAD(&bfqd->dispatch); - hrtimer_init(&bfqd->idle_slice_timer, CLOCK_MONOTONIC, - HRTIMER_MODE_REL); - bfqd->idle_slice_timer.function = bfq_idle_slice_timer; + hrtimer_setup(&bfqd->idle_slice_timer, bfq_idle_slice_timer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL); bfqd->queue_weights_tree = RB_ROOT_CACHED; +#ifdef CONFIG_BFQ_GROUP_IOSCHED bfqd->num_groups_with_pending_reqs = 0; +#endif - INIT_LIST_HEAD(&bfqd->active_list); + INIT_LIST_HEAD(&bfqd->active_list[0]); + INIT_LIST_HEAD(&bfqd->active_list[1]); INIT_LIST_HEAD(&bfqd->idle_list); INIT_HLIST_HEAD(&bfqd->burst_list); @@ -7015,7 +7314,6 @@ static int bfq_init_queue(struct request_queue *q, struct elevator_type *e) */ bfqd->bfq_wr_coeff = 30; bfqd->bfq_wr_rt_max_time = msecs_to_jiffies(300); - bfqd->bfq_wr_max_time = 0; bfqd->bfq_wr_min_idle_time = msecs_to_jiffies(2000); bfqd->bfq_wr_min_inter_arr_async = msecs_to_jiffies(500); bfqd->bfq_wr_max_softrt_rate = 7000; /* @@ -7034,6 +7332,9 @@ static int bfq_init_queue(struct request_queue *q, struct elevator_type *e) ref_wr_duration[blk_queue_nonrot(bfqd->queue)]; bfqd->peak_rate = ref_rate[blk_queue_nonrot(bfqd->queue)] * 2 / 3; + /* see comments on the definition of next field inside bfq_data */ + bfqd->actuator_load_threshold = 4; + spin_lock_init(&bfqd->lock); /* @@ -7056,13 +7357,19 @@ static int bfq_init_queue(struct request_queue *q, struct elevator_type *e) goto out_free; bfq_init_root_group(bfqd->root_group, bfqd); bfq_init_entity(&bfqd->oom_bfqq.entity, bfqd->root_group); + bfq_depth_updated(q); + + /* We dispatch from request queue wide instead of hw queue */ + blk_queue_flag_set(QUEUE_FLAG_SQ_SCHED, q); + + blk_queue_flag_set(QUEUE_FLAG_DISABLE_WBT_DEF, q); + wbt_disable_default(q->disk); + blk_stat_enable_accounting(q); - wbt_disable_default(q); return 0; out_free: kfree(bfqd); - kobject_put(&eq->kobj); return -ENOMEM; } @@ -7278,7 +7585,7 @@ static ssize_t bfq_low_latency_store(struct elevator_queue *e, #define BFQ_ATTR(name) \ __ATTR(name, 0644, bfq_##name##_show, bfq_##name##_store) -static struct elv_fs_entry bfq_attrs[] = { +static const struct elv_fs_entry bfq_attrs[] = { BFQ_ATTR(fifo_expire_sync), BFQ_ATTR(fifo_expire_async), BFQ_ATTR(back_seek_max), @@ -7297,7 +7604,7 @@ static struct elevator_type iosched_bfq_mq = { .limit_depth = bfq_limit_depth, .prepare_request = bfq_prepare_request, .requeue_request = bfq_finish_requeue_request, - .finish_request = bfq_finish_requeue_request, + .finish_request = bfq_finish_request, .exit_icq = bfq_exit_icq, .insert_requests = bfq_insert_requests, .dispatch_request = bfq_dispatch_request, @@ -7310,7 +7617,6 @@ static struct elevator_type iosched_bfq_mq = { .request_merged = bfq_request_merged, .has_work = bfq_has_work, .depth_updated = bfq_depth_updated, - .init_hctx = bfq_init_hctx, .init_sched = bfq_init_queue, .exit_sched = bfq_exit_queue, }, |