Merge branch 'block'

Documentation/cgroups/blkio-controller.txt

@@ -89,6 +89,33 @@ Throttling/Upper Limit policy
 
  Limits for writes can be put using blkio.write_bps_device file.
 
+Hierarchical Cgroups
+====================
+- Currently none of the IO control policy supports hierarhical groups. But
+  cgroup interface does allow creation of hierarhical cgroups and internally
+  IO policies treat them as flat hierarchy.
+
+  So this patch will allow creation of cgroup hierarhcy but at the backend
+  everything will be treated as flat. So if somebody created a hierarchy like
+  as follows.
+
+			root
+			/  \
+		     test1 test2
+			|
+		     test3
+
+  CFQ and throttling will practically treat all groups at same level.
+
+				pivot
+			     /  |   \  \
+			root  test1 test2  test3
+
+  Down the line we can implement hierarchical accounting/control support
+  and also introduce a new cgroup file "use_hierarchy" which will control
+  whether cgroup hierarchy is viewed as flat or hierarchical by the policy..
+  This is how memory controller also has implemented the things.
+
 Various user visible config options
 ===================================
 CONFIG_BLK_CGROUP

block/blk-cgroup.c

@@ -1452,10 +1452,6 @@ blkiocg_create(struct cgroup_subsys *subsys, struct cgroup *cgroup)
 		goto done;
 	}
 
-	/* Currently we do not support hierarchy deeper than two level (0,1) */
-	if (parent != cgroup->top_cgroup)
-		return ERR_PTR(-EPERM);
-
 	blkcg = kzalloc(sizeof(*blkcg), GFP_KERNEL);
 	if (!blkcg)
 		return ERR_PTR(-ENOMEM);

block/blk-core.c

@@ -33,7 +33,7 @@
 
 #include "blk.h"
 
-EXPORT_TRACEPOINT_SYMBOL_GPL(block_remap);
+EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_remap);
 EXPORT_TRACEPOINT_SYMBOL_GPL(block_rq_remap);
 EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_complete);
 
@@ -1329,9 +1329,9 @@ static inline void blk_partition_remap(struct bio *bio)
 		bio->bi_sector += p->start_sect;
 		bio->bi_bdev = bdev->bd_contains;
 
-		trace_block_remap(bdev_get_queue(bio->bi_bdev), bio,
-				    bdev->bd_dev,
-				    bio->bi_sector - p->start_sect);
+		trace_block_bio_remap(bdev_get_queue(bio->bi_bdev), bio,
+				      bdev->bd_dev,
+				      bio->bi_sector - p->start_sect);
 	}
 }
 
@@ -1500,7 +1500,7 @@ static inline void __generic_make_request(struct bio *bio)
 			goto end_io;
 
 		if (old_sector != -1)
-			trace_block_remap(q, bio, old_dev, old_sector);
+			trace_block_bio_remap(q, bio, old_dev, old_sector);
 
 		old_sector = bio->bi_sector;
 		old_dev = bio->bi_bdev->bd_dev;

block/cfq-iosched.c

@@ -87,7 +87,6 @@ struct cfq_rb_root {
 	unsigned count;
 	unsigned total_weight;
 	u64 min_vdisktime;
-	struct rb_node *active;
 };
 #define CFQ_RB_ROOT	(struct cfq_rb_root) { .rb = RB_ROOT, .left = NULL, \
 			.count = 0, .min_vdisktime = 0, }
@@ -180,7 +179,6 @@ struct cfq_group {
 	/* group service_tree key */
 	u64 vdisktime;
 	unsigned int weight;
-	bool on_st;
 
 	/* number of cfqq currently on this group */
 	int nr_cfqq;
@@ -563,11 +561,6 @@ static void update_min_vdisktime(struct cfq_rb_root *st)
 	u64 vdisktime = st->min_vdisktime;
 	struct cfq_group *cfqg;
 
-	if (st->active) {
-		cfqg = rb_entry_cfqg(st->active);
-		vdisktime = cfqg->vdisktime;
-	}
-
 	if (st->left) {
 		cfqg = rb_entry_cfqg(st->left);
 		vdisktime = min_vdisktime(vdisktime, cfqg->vdisktime);
@@ -646,11 +639,11 @@ cfq_set_prio_slice(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 static inline bool cfq_slice_used(struct cfq_queue *cfqq)
 {
 	if (cfq_cfqq_slice_new(cfqq))
-		return 0;
+		return false;
 	if (time_before(jiffies, cfqq->slice_end))
-		return 0;
+		return false;
 
-	return 1;
+	return true;
 }
 
 /*
@@ -869,7 +862,7 @@ cfq_group_service_tree_add(struct cfq_data *cfqd, struct cfq_group *cfqg)
 	struct rb_node *n;
 
 	cfqg->nr_cfqq++;
-	if (cfqg->on_st)
+	if (!RB_EMPTY_NODE(&cfqg->rb_node))
 		return;
 
 	/*
@@ -885,7 +878,6 @@ cfq_group_service_tree_add(struct cfq_data *cfqd, struct cfq_group *cfqg)
 		cfqg->vdisktime = st->min_vdisktime;
 
 	__cfq_group_service_tree_add(st, cfqg);
-	cfqg->on_st = true;
 	st->total_weight += cfqg->weight;
 }
 
@@ -894,9 +886,6 @@ cfq_group_service_tree_del(struct cfq_data *cfqd, struct cfq_group *cfqg)
 {
 	struct cfq_rb_root *st = &cfqd->grp_service_tree;
 
-	if (st->active == &cfqg->rb_node)
-		st->active = NULL;
-
 	BUG_ON(cfqg->nr_cfqq < 1);
 	cfqg->nr_cfqq--;
 
@@ -905,7 +894,6 @@ cfq_group_service_tree_del(struct cfq_data *cfqd, struct cfq_group *cfqg)
 		return;
 
 	cfq_log_cfqg(cfqd, cfqg, "del_from_rr group");
-	cfqg->on_st = false;
 	st->total_weight -= cfqg->weight;
 	if (!RB_EMPTY_NODE(&cfqg->rb_node))
 		cfq_rb_erase(&cfqg->rb_node, st);
@@ -1095,7 +1083,7 @@ static void cfq_put_cfqg(struct cfq_group *cfqg)
 	if (!atomic_dec_and_test(&cfqg->ref))
 		return;
 	for_each_cfqg_st(cfqg, i, j, st)
-		BUG_ON(!RB_EMPTY_ROOT(&st->rb) || st->active != NULL);
+		BUG_ON(!RB_EMPTY_ROOT(&st->rb));
 	kfree(cfqg);
 }
 
@@ -1687,9 +1675,6 @@ __cfq_slice_expired(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 	if (cfqq == cfqd->active_queue)
 		cfqd->active_queue = NULL;
 
-	if (&cfqq->cfqg->rb_node == cfqd->grp_service_tree.active)
-		cfqd->grp_service_tree.active = NULL;
-
 	if (cfqd->active_cic) {
 		put_io_context(cfqd->active_cic->ioc);
 		cfqd->active_cic = NULL;
@@ -1901,10 +1886,10 @@ static bool cfq_should_idle(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 	 * in their service tree.
 	 */
 	if (service_tree->count == 1 && cfq_cfqq_sync(cfqq))
-		return 1;
+		return true;
 	cfq_log_cfqq(cfqd, cfqq, "Not idling. st->count:%d",
 			service_tree->count);
-	return 0;
+	return false;
 }
 
 static void cfq_arm_slice_timer(struct cfq_data *cfqd)
@@ -2116,12 +2101,7 @@ static void choose_service_tree(struct cfq_data *cfqd, struct cfq_group *cfqg)
 	unsigned count;
 	struct cfq_rb_root *st;
 	unsigned group_slice;
-
-	if (!cfqg) {
-		cfqd->serving_prio = IDLE_WORKLOAD;
-		cfqd->workload_expires = jiffies + 1;
-		return;
-	}
+	enum wl_prio_t original_prio = cfqd->serving_prio;
 
 	/* Choose next priority. RT > BE > IDLE */
 	if (cfq_group_busy_queues_wl(RT_WORKLOAD, cfqd, cfqg))
@@ -2134,6 +2114,9 @@ static void choose_service_tree(struct cfq_data *cfqd, struct cfq_group *cfqg)
 		return;
 	}
 
+	if (original_prio != cfqd->serving_prio)
+		goto new_workload;
+
 	/*
 	 * For RT and BE, we have to choose also the type
 	 * (SYNC, SYNC_NOIDLE, ASYNC), and to compute a workload
@@ -2148,6 +2131,7 @@ static void choose_service_tree(struct cfq_data *cfqd, struct cfq_group *cfqg)
 	if (count && !time_after(jiffies, cfqd->workload_expires))
 		return;
 
+new_workload:
 	/* otherwise select new workload type */
 	cfqd->serving_type =
 		cfq_choose_wl(cfqd, cfqg, cfqd->serving_prio);
@@ -2199,7 +2183,6 @@ static struct cfq_group *cfq_get_next_cfqg(struct cfq_data *cfqd)
 	if (RB_EMPTY_ROOT(&st->rb))
 		return NULL;
 	cfqg = cfq_rb_first_group(st);
-	st->active = &cfqg->rb_node;
 	update_min_vdisktime(st);
 	return cfqg;
 }
@@ -2293,6 +2276,17 @@ static struct cfq_queue *cfq_select_queue(struct cfq_data *cfqd)
 		goto keep_queue;
 	}
 
+	/*
+	 * This is a deep seek queue, but the device is much faster than
+	 * the queue can deliver, don't idle
+	 **/
+	if (CFQQ_SEEKY(cfqq) && cfq_cfqq_idle_window(cfqq) &&
+	    (cfq_cfqq_slice_new(cfqq) ||
+	    (cfqq->slice_end - jiffies > jiffies - cfqq->slice_start))) {
+		cfq_clear_cfqq_deep(cfqq);
+		cfq_clear_cfqq_idle_window(cfqq);
+	}
+
 	if (cfqq->dispatched && cfq_should_idle(cfqd, cfqq)) {
 		cfqq = NULL;
 		goto keep_queue;
@@ -2367,12 +2361,12 @@ static inline bool cfq_slice_used_soon(struct cfq_data *cfqd,
 {
 	/* the queue hasn't finished any request, can't estimate */
 	if (cfq_cfqq_slice_new(cfqq))
-		return 1;
+		return true;
 	if (time_after(jiffies + cfqd->cfq_slice_idle * cfqq->dispatched,
 		cfqq->slice_end))
-		return 1;
+		return true;
 
-	return 0;
+	return false;
 }
 
 static bool cfq_may_dispatch(struct cfq_data *cfqd, struct cfq_queue *cfqq)
@@ -3265,6 +3259,10 @@ cfq_should_preempt(struct cfq_data *cfqd, struct cfq_queue *new_cfqq,
 	if (cfq_class_rt(new_cfqq) && !cfq_class_rt(cfqq))
 		return true;
 
+	/* An idle queue should not be idle now for some reason */
+	if (RB_EMPTY_ROOT(&cfqq->sort_list) && !cfq_should_idle(cfqd, cfqq))
+		return true;
+
 	if (!cfqd->active_cic || !cfq_cfqq_wait_request(cfqq))
 		return false;