From b0b78f81a5a3bfe9442fcc3a9c13e298a742556a Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 8 Apr 2009 10:56:08 +0200 Subject: cfq-iosched: use rw_is_sync() to see if rw flags are sync or not Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'block/cfq-iosched.c') diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index a4809de6fea6..f36f8b4d04cf 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -2062,7 +2062,7 @@ static int cfq_may_queue(struct request_queue *q, int rw) if (!cic) return ELV_MQUEUE_MAY; - cfqq = cic_to_cfqq(cic, rw & REQ_RW_SYNC); + cfqq = cic_to_cfqq(cic, rw_is_sync(rw)); if (cfqq) { cfq_init_prio_data(cfqq, cic->ioc); cfq_prio_boost(cfqq); -- cgit v1.2.3 From ff6657c6c8ac99444e5dd4c4f7c1dc9271173382 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 8 Apr 2009 10:58:57 +0200 Subject: cfq-iosched: get rid of private SYNC/ASYNC defines We can just use the block layer BLK_RW_SYNC/ASYNC defines now. Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 24 +++++++++++------------- 1 file changed, 11 insertions(+), 13 deletions(-) (limited to 'block/cfq-iosched.c') diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index f36f8b4d04cf..cba8a5d91bfa 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -56,9 +56,6 @@ static DEFINE_SPINLOCK(ioc_gone_lock); #define cfq_class_idle(cfqq) ((cfqq)->ioprio_class == IOPRIO_CLASS_IDLE) #define cfq_class_rt(cfqq) ((cfqq)->ioprio_class == IOPRIO_CLASS_RT) -#define ASYNC (0) -#define SYNC (1) - #define sample_valid(samples) ((samples) > 80) /* @@ -1333,14 +1330,14 @@ static void __cfq_exit_single_io_context(struct cfq_data *cfqd, if (ioc->ioc_data == cic) rcu_assign_pointer(ioc->ioc_data, NULL); - if (cic->cfqq[ASYNC]) { - cfq_exit_cfqq(cfqd, cic->cfqq[ASYNC]); - cic->cfqq[ASYNC] = NULL; + if (cic->cfqq[BLK_RW_ASYNC]) { + cfq_exit_cfqq(cfqd, cic->cfqq[BLK_RW_ASYNC]); + cic->cfqq[BLK_RW_ASYNC] = NULL; } - if (cic->cfqq[SYNC]) { - cfq_exit_cfqq(cfqd, cic->cfqq[SYNC]); - cic->cfqq[SYNC] = NULL; + if (cic->cfqq[BLK_RW_SYNC]) { + cfq_exit_cfqq(cfqd, cic->cfqq[BLK_RW_SYNC]); + cic->cfqq[BLK_RW_SYNC] = NULL; } } @@ -1449,17 +1446,18 @@ static void changed_ioprio(struct io_context *ioc, struct cfq_io_context *cic) spin_lock_irqsave(cfqd->queue->queue_lock, flags); - cfqq = cic->cfqq[ASYNC]; + cfqq = cic->cfqq[BLK_RW_ASYNC]; if (cfqq) { struct cfq_queue *new_cfqq; - new_cfqq = cfq_get_queue(cfqd, ASYNC, cic->ioc, GFP_ATOMIC); + new_cfqq = cfq_get_queue(cfqd, BLK_RW_ASYNC, cic->ioc, + GFP_ATOMIC); if (new_cfqq) { - cic->cfqq[ASYNC] = new_cfqq; + cic->cfqq[BLK_RW_ASYNC] = new_cfqq; cfq_put_queue(cfqq); } } - cfqq = cic->cfqq[SYNC]; + cfqq = cic->cfqq[BLK_RW_SYNC]; if (cfqq) cfq_mark_cfqq_prio_changed(cfqq); -- cgit v1.2.3 From d6ceb25e8d8bccf826848c2621a50d02c0a7f4ae Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 14 Apr 2009 14:18:16 +0200 Subject: cfq-iosched: don't delay queue kick for a merged request "Zhang, Yanmin" reports that commit b029195dda0129b427c6e579a3bb3ae752da3a93 introduced a regression of about 50% with sequential threaded read workloads. The test case is: tiotest -k0 -k1 -k3 -f 80 -t 32 which starts 32 threads each reading a 80MB file. Twiddle the kick queue logic so that we do start IO immediately, if it appears to be a fully merged request. We can't really detect that, so just check if the request is bigger than a page or not. The assumption is that since single bio issues will first queue a single request with just one page attached and then later do merges on that, if we already have more than a page worth of data in the request, then the request is most likely good to go. Verified that this doesn't cause a regression with the test case that commit b029195dda0129b427c6e579a3bb3ae752da3a93 was fixing. It does not, we still see maximum sized requests for the queue-then-merge cases. Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'block/cfq-iosched.c') diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index cba8a5d91bfa..56e9d8503cf1 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -1903,10 +1903,17 @@ cfq_rq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq, * Remember that we saw a request from this process, but * don't start queuing just yet. Otherwise we risk seeing lots * of tiny requests, because we disrupt the normal plugging - * and merging. + * and merging. If the request is already larger than a single + * page, let it rip immediately. For that case we assume that + * merging is already done. */ - if (cfq_cfqq_wait_request(cfqq)) + if (cfq_cfqq_wait_request(cfqq)) { + if (blk_rq_bytes(rq) > PAGE_CACHE_SIZE) { + del_timer(&cfqd->idle_slice_timer); + blk_start_queueing(cfqd->queue); + } cfq_mark_cfqq_must_dispatch(cfqq); + } } else if (cfq_should_preempt(cfqd, cfqq, rq)) { /* * not the active queue - expire current slice if it is -- cgit v1.2.3 From 40bb54d197a10fe86d73073ee8202ba2812a05fa Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 15 Apr 2009 12:11:10 +0200 Subject: cfq-iosched: no need to save interrupts in cfq_kick_queue() It's called from the workqueue handlers from process context, so we always have irqs enabled when entered. Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'block/cfq-iosched.c') diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 56e9d8503cf1..e59f2108272f 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -2157,11 +2157,10 @@ static void cfq_kick_queue(struct work_struct *work) struct cfq_data *cfqd = container_of(work, struct cfq_data, unplug_work); struct request_queue *q = cfqd->queue; - unsigned long flags; - spin_lock_irqsave(q->queue_lock, flags); + spin_lock_irq(q->queue_lock); blk_start_queueing(q); - spin_unlock_irqrestore(q->queue_lock, flags); + spin_unlock_irq(q->queue_lock); } /* -- cgit v1.2.3 From 2d870722965211de072bb36b446a4df99dae07e1 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 15 Apr 2009 12:12:46 +0200 Subject: cfq-iosched: tweak kick logic a bit more We only kick the dispatch for an idling queue, if we think it's a (somewhat) fully merged request. Also allow a kick if we have other busy queues in the system, since we don't want to risk waiting for a potential merge in that case. It's better to get some work done and proceed. Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'block/cfq-iosched.c') diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index e59f2108272f..579f31822469 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -1905,10 +1905,13 @@ cfq_rq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq, * of tiny requests, because we disrupt the normal plugging * and merging. If the request is already larger than a single * page, let it rip immediately. For that case we assume that - * merging is already done. + * merging is already done. Ditto for a busy system that + * has other work pending, don't risk delaying until the + * idle timer unplug to continue working. */ if (cfq_cfqq_wait_request(cfqq)) { - if (blk_rq_bytes(rq) > PAGE_CACHE_SIZE) { + if (blk_rq_bytes(rq) > PAGE_CACHE_SIZE || + cfqd->busy_queues > 1) { del_timer(&cfqd->idle_slice_timer); blk_start_queueing(cfqd->queue); } -- cgit v1.2.3 From 9481ffdc61738a91baf0f8b7fb20922768ae1b8e Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 15 Apr 2009 12:14:13 +0200 Subject: cfq-iosched: log responsible 'cfqq' in idle timer arm Makes it easier to read the traces. Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'block/cfq-iosched.c') diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 579f31822469..090a4ee75b9d 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -936,7 +936,7 @@ static void cfq_arm_slice_timer(struct cfq_data *cfqd) sl = min(sl, msecs_to_jiffies(CFQ_MIN_TT)); mod_timer(&cfqd->idle_slice_timer, jiffies + sl); - cfq_log(cfqd, "arm_idle: %lu", sl); + cfq_log_cfqq(cfqd, cfqq, "arm_idle: %lu", sl); } /* -- cgit v1.2.3 From a36e71f996e25d6213f57951f7ae1874086ec57e Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 15 Apr 2009 12:15:11 +0200 Subject: cfq-iosched: add close cooperator code If we have processes that are working in close proximity to each other on disk, we don't want to idle wait. Instead allow the close process to issue a request, getting better aggregate bandwidth. The anticipatory scheduler has similar checks, noop and deadline do not need it since they don't care about process <-> io mappings. The code for CFQ is a little more involved though, since we split request queues into per-process contexts. This fixes a performance problem with eg dump(8), since it uses several processes in some silly attempt to speed IO up. Even if dump(8) isn't really a valid case (it should be fixed by using CLONE_IO), there are other cases where we see close processes and where idling ends up hurting performance. Credit goes to Jeff Moyer for writing the initial implementation. Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 223 ++++++++++++++++++++++++++++++++++++++++++++++------ 1 file changed, 198 insertions(+), 25 deletions(-) (limited to 'block/cfq-iosched.c') diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 090a4ee75b9d..0d3b70de3d80 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -80,6 +80,14 @@ struct cfq_data { * rr list of queues with requests and the count of them */ struct cfq_rb_root service_tree; + + /* + * Each priority tree is sorted by next_request position. These + * trees are used when determining if two or more queues are + * interleaving requests (see cfq_close_cooperator). + */ + struct rb_root prio_trees[CFQ_PRIO_LISTS]; + unsigned int busy_queues; /* * Used to track any pending rt requests so we can pre-empt current @@ -144,6 +152,8 @@ struct cfq_queue { struct rb_node rb_node; /* service_tree key */ unsigned long rb_key; + /* prio tree member */ + struct rb_node p_node; /* sorted list of pending requests */ struct rb_root sort_list; /* if fifo isn't expired, next request to serve */ @@ -182,6 +192,7 @@ enum cfqq_state_flags { CFQ_CFQQ_FLAG_prio_changed, /* task priority has changed */ CFQ_CFQQ_FLAG_slice_new, /* no requests dispatched in slice */ CFQ_CFQQ_FLAG_sync, /* synchronous queue */ + CFQ_CFQQ_FLAG_coop, /* has done a coop jump of the queue */ }; #define CFQ_CFQQ_FNS(name) \ @@ -208,6 +219,7 @@ CFQ_CFQQ_FNS(idle_window); CFQ_CFQQ_FNS(prio_changed); CFQ_CFQQ_FNS(slice_new); CFQ_CFQQ_FNS(sync); +CFQ_CFQQ_FNS(coop); #undef CFQ_CFQQ_FNS #define cfq_log_cfqq(cfqd, cfqq, fmt, args...) \ @@ -416,13 +428,17 @@ static struct cfq_queue *cfq_rb_first(struct cfq_rb_root *root) return NULL; } +static void rb_erase_init(struct rb_node *n, struct rb_root *root) +{ + rb_erase(n, root); + RB_CLEAR_NODE(n); +} + static void cfq_rb_erase(struct rb_node *n, struct cfq_rb_root *root) { if (root->left == n) root->left = NULL; - - rb_erase(n, &root->rb); - RB_CLEAR_NODE(n); + rb_erase_init(n, &root->rb); } /* @@ -467,8 +483,8 @@ static unsigned long cfq_slice_offset(struct cfq_data *cfqd, * requests waiting to be processed. It is sorted in the order that * we will service the queues. */ -static void cfq_service_tree_add(struct cfq_data *cfqd, - struct cfq_queue *cfqq, int add_front) +static void cfq_service_tree_add(struct cfq_data *cfqd, struct cfq_queue *cfqq, + int add_front) { struct rb_node **p, *parent; struct cfq_queue *__cfqq; @@ -541,6 +557,63 @@ static void cfq_service_tree_add(struct cfq_data *cfqd, rb_insert_color(&cfqq->rb_node, &cfqd->service_tree.rb); } +static struct cfq_queue * +cfq_prio_tree_lookup(struct cfq_data *cfqd, int ioprio, sector_t sector, + struct rb_node **ret_parent, struct rb_node ***rb_link) +{ + struct rb_root *root = &cfqd->prio_trees[ioprio]; + struct rb_node **p, *parent; + struct cfq_queue *cfqq = NULL; + + parent = NULL; + p = &root->rb_node; + while (*p) { + struct rb_node **n; + + parent = *p; + cfqq = rb_entry(parent, struct cfq_queue, p_node); + + /* + * Sort strictly based on sector. Smallest to the left, + * largest to the right. + */ + if (sector > cfqq->next_rq->sector) + n = &(*p)->rb_right; + else if (sector < cfqq->next_rq->sector) + n = &(*p)->rb_left; + else + break; + p = n; + } + + *ret_parent = parent; + if (rb_link) + *rb_link = p; + return NULL; +} + +static void cfq_prio_tree_add(struct cfq_data *cfqd, struct cfq_queue *cfqq) +{ + struct rb_root *root = &cfqd->prio_trees[cfqq->ioprio]; + struct rb_node **p, *parent; + struct cfq_queue *__cfqq; + + if (!RB_EMPTY_NODE(&cfqq->p_node)) + rb_erase_init(&cfqq->p_node, root); + + if (cfq_class_idle(cfqq)) + return; + if (!cfqq->next_rq) + return; + + __cfqq = cfq_prio_tree_lookup(cfqd, cfqq->ioprio, cfqq->next_rq->sector, + &parent, &p); + BUG_ON(__cfqq); + + rb_link_node(&cfqq->p_node, parent, p); + rb_insert_color(&cfqq->p_node, root); +} + /* * Update cfqq's position in the service tree. */ @@ -549,8 +622,10 @@ static void cfq_resort_rr_list(struct cfq_data *cfqd, struct cfq_queue *cfqq) /* * Resorting requires the cfqq to be on the RR list already. */ - if (cfq_cfqq_on_rr(cfqq)) + if (cfq_cfqq_on_rr(cfqq)) { cfq_service_tree_add(cfqd, cfqq, 0); + cfq_prio_tree_add(cfqd, cfqq); + } } /* @@ -581,6 +656,8 @@ static void cfq_del_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq) if (!RB_EMPTY_NODE(&cfqq->rb_node)) cfq_rb_erase(&cfqq->rb_node, &cfqd->service_tree); + if (!RB_EMPTY_NODE(&cfqq->p_node)) + rb_erase_init(&cfqq->p_node, &cfqd->prio_trees[cfqq->ioprio]); BUG_ON(!cfqd->busy_queues); cfqd->busy_queues--; @@ -610,7 +687,7 @@ static void cfq_add_rq_rb(struct request *rq) { struct cfq_queue *cfqq = RQ_CFQQ(rq); struct cfq_data *cfqd = cfqq->cfqd; - struct request *__alias; + struct request *__alias, *prev; cfqq->queued[rq_is_sync(rq)]++; @@ -627,7 +704,15 @@ static void cfq_add_rq_rb(struct request *rq) /* * check if this request is a better next-serve candidate */ + prev = cfqq->next_rq; cfqq->next_rq = cfq_choose_req(cfqd, cfqq->next_rq, rq); + + /* + * adjust priority tree position, if ->next_rq changes + */ + if (prev != cfqq->next_rq) + cfq_prio_tree_add(cfqd, cfqq); + BUG_ON(!cfqq->next_rq); } @@ -840,11 +925,15 @@ static struct cfq_queue *cfq_get_next_queue(struct cfq_data *cfqd) /* * Get and set a new active queue for service. */ -static struct cfq_queue *cfq_set_active_queue(struct cfq_data *cfqd) +static struct cfq_queue *cfq_set_active_queue(struct cfq_data *cfqd, + struct cfq_queue *cfqq) { - struct cfq_queue *cfqq; + if (!cfqq) { + cfqq = cfq_get_next_queue(cfqd); + if (cfqq) + cfq_clear_cfqq_coop(cfqq); + } - cfqq = cfq_get_next_queue(cfqd); __cfq_set_active_queue(cfqd, cfqq); return cfqq; } @@ -868,17 +957,89 @@ static inline int cfq_rq_close(struct cfq_data *cfqd, struct request *rq) return cfq_dist_from_last(cfqd, rq) <= cic->seek_mean; } -static int cfq_close_cooperator(struct cfq_data *cfq_data, - struct cfq_queue *cfqq) +static struct cfq_queue *cfqq_close(struct cfq_data *cfqd, + struct cfq_queue *cur_cfqq) +{ + struct rb_root *root = &cfqd->prio_trees[cur_cfqq->ioprio]; + struct rb_node *parent, *node; + struct cfq_queue *__cfqq; + sector_t sector = cfqd->last_position; + + if (RB_EMPTY_ROOT(root)) + return NULL; + + /* + * First, if we find a request starting at the end of the last + * request, choose it. + */ + __cfqq = cfq_prio_tree_lookup(cfqd, cur_cfqq->ioprio, + sector, &parent, NULL); + if (__cfqq) + return __cfqq; + + /* + * If the exact sector wasn't found, the parent of the NULL leaf + * will contain the closest sector. + */ + __cfqq = rb_entry(parent, struct cfq_queue, p_node); + if (cfq_rq_close(cfqd, __cfqq->next_rq)) + return __cfqq; + + if (__cfqq->next_rq->sector < sector) + node = rb_next(&__cfqq->p_node); + else + node = rb_prev(&__cfqq->p_node); + if (!node) + return NULL; + + __cfqq = rb_entry(node, struct cfq_queue, p_node); + if (cfq_rq_close(cfqd, __cfqq->next_rq)) + return __cfqq; + + return NULL; +} + +/* + * cfqd - obvious + * cur_cfqq - passed in so that we don't decide that the current queue is + * closely cooperating with itself. + * + * So, basically we're assuming that that cur_cfqq has dispatched at least + * one request, and that cfqd->last_position reflects a position on the disk + * associated with the I/O issued by cur_cfqq. I'm not sure this is a valid + * assumption. + */ +static struct cfq_queue *cfq_close_cooperator(struct cfq_data *cfqd, + struct cfq_queue *cur_cfqq, + int probe) { + struct cfq_queue *cfqq; + + /* + * A valid cfq_io_context is necessary to compare requests against + * the seek_mean of the current cfqq. + */ + if (!cfqd->active_cic) + return NULL; + /* * We should notice if some of the queues are cooperating, eg * working closely on the same area of the disk. In that case, * we can group them together and don't waste time idling. */ - return 0; + cfqq = cfqq_close(cfqd, cur_cfqq); + if (!cfqq) + return NULL; + + if (cfq_cfqq_coop(cfqq)) + return NULL; + + if (!probe) + cfq_mark_cfqq_coop(cfqq); + return cfqq; } + #define CIC_SEEKY(cic) ((cic)->seek_mean > (8 * 1024)) static void cfq_arm_slice_timer(struct cfq_data *cfqd) @@ -917,13 +1078,6 @@ static void cfq_arm_slice_timer(struct cfq_data *cfqd) if (!cic || !atomic_read(&cic->ioc->nr_tasks)) return; - /* - * See if this prio level has a good candidate - */ - if (cfq_close_cooperator(cfqd, cfqq) && - (sample_valid(cic->ttime_samples) && cic->ttime_mean > 2)) - return; - cfq_mark_cfqq_wait_request(cfqq); /* @@ -1000,7 +1154,7 @@ cfq_prio_to_maxrq(struct cfq_data *cfqd, struct cfq_queue *cfqq) */ static struct cfq_queue *cfq_select_queue(struct cfq_data *cfqd) { - struct cfq_queue *cfqq; + struct cfq_queue *cfqq, *new_cfqq = NULL; cfqq = cfqd->active_queue; if (!cfqq) @@ -1033,6 +1187,16 @@ static struct cfq_queue *cfq_select_queue(struct cfq_data *cfqd) if (!RB_EMPTY_ROOT(&cfqq->sort_list)) goto keep_queue; + /* + * If another queue has a request waiting within our mean seek + * distance, let it run. The expire code will check for close + * cooperators and put the close queue at the front of the service + * tree. + */ + new_cfqq = cfq_close_cooperator(cfqd, cfqq, 0); + if (new_cfqq) + goto expire; + /* * No requests pending. If the active queue still has requests in * flight or is idling for a new request, allow either of these @@ -1047,7 +1211,7 @@ static struct cfq_queue *cfq_select_queue(struct cfq_data *cfqd) expire: cfq_slice_expired(cfqd, 0); new_queue: - cfqq = cfq_set_active_queue(cfqd); + cfqq = cfq_set_active_queue(cfqd, new_cfqq); keep_queue: return cfqq; } @@ -1508,6 +1672,7 @@ retry: } RB_CLEAR_NODE(&cfqq->rb_node); + RB_CLEAR_NODE(&cfqq->p_node); INIT_LIST_HEAD(&cfqq->fifo); atomic_set(&cfqq->ref, 0); @@ -2000,16 +2165,24 @@ static void cfq_completed_request(struct request_queue *q, struct request *rq) * or if we want to idle in case it has no pending requests. */ if (cfqd->active_queue == cfqq) { + const bool cfqq_empty = RB_EMPTY_ROOT(&cfqq->sort_list); + if (cfq_cfqq_slice_new(cfqq)) { cfq_set_prio_slice(cfqd, cfqq); cfq_clear_cfqq_slice_new(cfqq); } + /* + * If there are no requests waiting in this queue, and + * there are other queues ready to issue requests, AND + * those other queues are issuing requests within our + * mean seek distance, give them a chance to run instead + * of idling. + */ if (cfq_slice_used(cfqq) || cfq_class_idle(cfqq)) cfq_slice_expired(cfqd, 1); - else if (sync && !rq_noidle(rq) && - RB_EMPTY_ROOT(&cfqq->sort_list)) { + else if (cfqq_empty && !cfq_close_cooperator(cfqd, cfqq, 1) && + sync && !rq_noidle(rq)) cfq_arm_slice_timer(cfqd); - } } if (!cfqd->rq_in_driver) -- cgit v1.2.3 From 4d00aa47e2337dcfe2d8a7215dbde3765b507167 Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Tue, 21 Apr 2009 07:25:04 +0200 Subject: cfq-iosched: make seek_mean converge more quickly Right now, depending on the first sector to which a process issues I/O, the seek time may start out way out of whack. So make sure we start with 0 sectors in seek, instead of the offset of the first request issued. Signed-off-by: Jeff Moyer Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'block/cfq-iosched.c') diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 0d3b70de3d80..0eb4aff9df68 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -1908,7 +1908,9 @@ cfq_update_io_seektime(struct cfq_data *cfqd, struct cfq_io_context *cic, sector_t sdist; u64 total; - if (cic->last_request_pos < rq->sector) + if (!cic->last_request_pos) + sdist = 0; + else if (cic->last_request_pos < rq->sector) sdist = rq->sector - cic->last_request_pos; else sdist = cic->last_request_pos - rq->sector; -- cgit v1.2.3 From 04dc6e71a28d4815bf9431efcafc107bb0ad2792 Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Tue, 21 Apr 2009 07:31:56 +0200 Subject: cfq-iosched: use the default seek distance when there aren't enough seek samples If the cfq io context doesn't have enough samples yet to provide a mean seek distance, then use the default threshold we have for seeky IO instead of defaulting to 0. Signed-off-by: Jeff Moyer Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'block/cfq-iosched.c') diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 0eb4aff9df68..7e13f04b5ed4 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -947,14 +947,18 @@ static inline sector_t cfq_dist_from_last(struct cfq_data *cfqd, return cfqd->last_position - rq->sector; } +#define CIC_SEEK_THR 8 * 1024 +#define CIC_SEEKY(cic) ((cic)->seek_mean > CIC_SEEK_THR) + static inline int cfq_rq_close(struct cfq_data *cfqd, struct request *rq) { struct cfq_io_context *cic = cfqd->active_cic; + sector_t sdist = cic->seek_mean; if (!sample_valid(cic->seek_samples)) - return 0; + sdist = CIC_SEEK_THR; - return cfq_dist_from_last(cfqd, rq) <= cic->seek_mean; + return cfq_dist_from_last(cfqd, rq) <= sdist; } static struct cfq_queue *cfqq_close(struct cfq_data *cfqd, @@ -1039,9 +1043,6 @@ static struct cfq_queue *cfq_close_cooperator(struct cfq_data *cfqd, return cfqq; } - -#define CIC_SEEKY(cic) ((cic)->seek_mean > (8 * 1024)) - static void cfq_arm_slice_timer(struct cfq_data *cfqd) { struct cfq_queue *cfqq = cfqd->active_queue; -- cgit v1.2.3 From 26a2ac009c2b07e1959c8864ca23486c1c485587 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 23 Apr 2009 12:13:27 +0200 Subject: cfq-iosched: clear ->prio_trees[] on cfqd alloc Not strictly needed, but we should make it clear that we init the rbtree roots here. Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'block/cfq-iosched.c') diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 7e13f04b5ed4..20a54b8e03e1 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -2446,12 +2446,22 @@ static void cfq_exit_queue(struct elevator_queue *e) static void *cfq_init_queue(struct request_queue *q) { struct cfq_data *cfqd; + int i; cfqd = kmalloc_node(sizeof(*cfqd), GFP_KERNEL | __GFP_ZERO, q->node); if (!cfqd) return NULL; cfqd->service_tree = CFQ_RB_ROOT; + + /* + * Not strictly needed (since RB_ROOT just clears the node and we + * zeroed cfqd on alloc), but better be safe in case someone decides + * to add magic to the rb code + */ + for (i = 0; i < CFQ_PRIO_LISTS; i++) + cfqd->prio_trees[i] = RB_ROOT; + INIT_LIST_HEAD(&cfqd->cic_list); cfqd->queue = q; -- cgit v1.2.3 From 3ac6c9f8a66726745136e46f63600550c3eb6cec Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 23 Apr 2009 12:14:56 +0200 Subject: cfq-iosched: fix bug with aliased request and cooperation detection cfq_prio_tree_lookup() should return the direct match, yet it always returns zero. Fix that. cfq_prio_tree_add() assumes that we don't get a direct match, while it is very possible that we do. Using O_DIRECT, you can have different cfqq with matching requests, since you don't have the page cache to serialize things for you. Fix this bug by only adding the cfqq if there isn't an existing match. Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'block/cfq-iosched.c') diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 20a54b8e03e1..b0b754a6882b 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -584,12 +584,13 @@ cfq_prio_tree_lookup(struct cfq_data *cfqd, int ioprio, sector_t sector, else break; p = n; + cfqq = NULL; } *ret_parent = parent; if (rb_link) *rb_link = p; - return NULL; + return cfqq; } static void cfq_prio_tree_add(struct cfq_data *cfqd, struct cfq_queue *cfqq) @@ -608,10 +609,10 @@ static void cfq_prio_tree_add(struct cfq_data *cfqd, struct cfq_queue *cfqq) __cfqq = cfq_prio_tree_lookup(cfqd, cfqq->ioprio, cfqq->next_rq->sector, &parent, &p); - BUG_ON(__cfqq); - - rb_link_node(&cfqq->p_node, parent, p); - rb_insert_color(&cfqq->p_node, root); + if (!__cfqq) { + rb_link_node(&cfqq->p_node, parent, p); + rb_insert_color(&cfqq->p_node, root); + } } /* -- cgit v1.2.3 From f2d1f0ae7851be5ebd9613a80dac139270938809 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 23 Apr 2009 12:19:38 +0200 Subject: cfq-iosched: cache prio_tree root in cfqq->p_root Currently we look it up from ->ioprio, but ->ioprio can change if either the process gets its IO priority changed explicitly, or if cfq decides to temporarily boost it. So if we are unlucky, we can end up attempting to remove a node from a different rbtree root than where it was added. Fix this by using ->org_ioprio as the prio_tree index, since that will only change for explicit IO priority settings (not for a boost). Additionally cache the rbtree root inside the cfqq, then we don't have to add code to reinsert the cfqq in the prio_tree if IO priority changes. Signed-off-by: Jens Axboe --- block/cfq-iosched.c | 34 ++++++++++++++++++++-------------- 1 file changed, 20 insertions(+), 14 deletions(-) (limited to 'block/cfq-iosched.c') diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index b0b754a6882b..a55a9bd75bd1 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -154,6 +154,8 @@ struct cfq_queue { unsigned long rb_key; /* prio tree member */ struct rb_node p_node; + /* prio tree root we belong to, if any */ + struct rb_root *p_root; /* sorted list of pending requests */ struct rb_root sort_list; /* if fifo isn't expired, next request to serve */ @@ -558,10 +560,10 @@ static void cfq_service_tree_add(struct cfq_data *cfqd, struct cfq_queue *cfqq, } static struct cfq_queue * -cfq_prio_tree_lookup(struct cfq_data *cfqd, int ioprio, sector_t sector, - struct rb_node **ret_parent, struct rb_node ***rb_link) +cfq_prio_tree_lookup(struct cfq_data *cfqd, struct rb_root *root, + sector_t sector, struct rb_node **ret_parent, + struct rb_node ***rb_link) { - struct rb_root *root = &cfqd->prio_trees[ioprio]; struct rb_node **p, *parent; struct cfq_queue *cfqq = NULL; @@ -595,24 +597,27 @@ cfq_prio_tree_lookup(struct cfq_data *cfqd, int ioprio, sector_t sector, static void cfq_prio_tree_add(struct cfq_data *cfqd, struct cfq_queue *cfqq) { - struct rb_root *root = &cfqd->prio_trees[cfqq->ioprio]; struct rb_node **p, *parent; struct cfq_queue *__cfqq; - if (!RB_EMPTY_NODE(&cfqq->p_node)) - rb_erase_init(&cfqq->p_node, root); + if (cfqq->p_root) { + rb_erase(&cfqq->p_node, cfqq->p_root); + cfqq->p_root = NULL; + } if (cfq_class_idle(cfqq)) return; if (!cfqq->next_rq) return; - __cfqq = cfq_prio_tree_lookup(cfqd, cfqq->ioprio, cfqq->next_rq->sector, + cfqq->p_root = &cfqd->prio_trees[cfqq->org_ioprio]; + __cfqq = cfq_prio_tree_lookup(cfqd, cfqq->p_root, cfqq->next_rq->sector, &parent, &p); if (!__cfqq) { rb_link_node(&cfqq->p_node, parent, p); - rb_insert_color(&cfqq->p_node, root); - } + rb_insert_color(&cfqq->p_node, cfqq->p_root); + } else + cfqq->p_root = NULL; } /* @@ -657,8 +662,10 @@ static void cfq_del_cfqq_rr(struct cfq_data *cfqd, struct cfq_queue *cfqq) if (!RB_EMPTY_NODE(&cfqq->rb_node)) cfq_rb_erase(&cfqq->rb_node, &cfqd->service_tree); - if (!RB_EMPTY_NODE(&cfqq->p_node)) - rb_erase_init(&cfqq->p_node, &cfqd->prio_trees[cfqq->ioprio]); + if (cfqq->p_root) { + rb_erase(&cfqq->p_node, cfqq->p_root); + cfqq->p_root = NULL; + } BUG_ON(!cfqd->busy_queues); cfqd->busy_queues--; @@ -965,7 +972,7 @@ static inline int cfq_rq_close(struct cfq_data *cfqd, struct request *rq) static struct cfq_queue *cfqq_close(struct cfq_data *cfqd, struct cfq_queue *cur_cfqq) { - struct rb_root *root = &cfqd->prio_trees[cur_cfqq->ioprio]; + struct rb_root *root = &cfqd->prio_trees[cur_cfqq->org_ioprio]; struct rb_node *parent, *node; struct cfq_queue *__cfqq; sector_t sector = cfqd->last_position; @@ -977,8 +984,7 @@ static struct cfq_queue *cfqq_close(struct cfq_data *cfqd, * First, if we find a request starting at the end of the last * request, choose it. */ - __cfqq = cfq_prio_tree_lookup(cfqd, cur_cfqq->ioprio, - sector, &parent, NULL); + __cfqq = cfq_prio_tree_lookup(cfqd, root, sector, &parent, NULL); if (__cfqq) return __cfqq; -- cgit v1.2.3