Merge branch 'for-linus' of git://git.kernel.dk/linux-2.6-block

* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
  block: add blk_run_queue_async
  block: blk_delay_queue() should use kblockd workqueue
  md: fix up raid1/raid10 unplugging.
  md: incorporate new plugging into raid5.
  md: provide generic support for handling unplug callbacks.
  md - remove old plugging code.
  md/dm - remove remains of plug_fn callback.
  md: use new plugging interface for RAID IO.
  block: drop queue lock before calling __blk_run_queue() for kblockd punt
  Revert "block: add callback function for unplug notification"
  block: Enhance new plugging support to support general callbacks

block/blk-core.c

@@ -204,7 +204,7 @@ static void blk_delay_work(struct work_struct *work)
 
 	q = container_of(work, struct request_queue, delay_work.work);
 	spin_lock_irq(q->queue_lock);
-	__blk_run_queue(q, false);
+	__blk_run_queue(q);
 	spin_unlock_irq(q->queue_lock);
 }
 
@@ -220,7 +220,8 @@ static void blk_delay_work(struct work_struct *work)
  */
 void blk_delay_queue(struct request_queue *q, unsigned long msecs)
 {
-	schedule_delayed_work(&q->delay_work, msecs_to_jiffies(msecs));
+	queue_delayed_work(kblockd_workqueue, &q->delay_work,
+				msecs_to_jiffies(msecs));
 }
 EXPORT_SYMBOL(blk_delay_queue);
 
@@ -238,7 +239,7 @@ void blk_start_queue(struct request_queue *q)
 	WARN_ON(!irqs_disabled());
 
 	queue_flag_clear(QUEUE_FLAG_STOPPED, q);
-	__blk_run_queue(q, false);
+	__blk_run_queue(q);
 }
 EXPORT_SYMBOL(blk_start_queue);
 
@@ -296,9 +297,8 @@ EXPORT_SYMBOL(blk_sync_queue);
  * Description:
  *    See @blk_run_queue. This variant must be called with the queue lock
  *    held and interrupts disabled.
- *
  */
-void __blk_run_queue(struct request_queue *q, bool force_kblockd)
+void __blk_run_queue(struct request_queue *q)
 {
 	if (unlikely(blk_queue_stopped(q)))
 		return;
@@ -307,14 +307,28 @@ void __blk_run_queue(struct request_queue *q, bool force_kblockd)
 	 * Only recurse once to avoid overrunning the stack, let the unplug
 	 * handling reinvoke the handler shortly if we already got there.
 	 */
-	if (!force_kblockd && !queue_flag_test_and_set(QUEUE_FLAG_REENTER, q)) {
+	if (!queue_flag_test_and_set(QUEUE_FLAG_REENTER, q)) {
 		q->request_fn(q);
 		queue_flag_clear(QUEUE_FLAG_REENTER, q);
 	} else
 		queue_delayed_work(kblockd_workqueue, &q->delay_work, 0);
 }
 EXPORT_SYMBOL(__blk_run_queue);
 
+/**
+ * blk_run_queue_async - run a single device queue in workqueue context
+ * @q:	The queue to run
+ *
+ * Description:
+ *    Tells kblockd to perform the equivalent of @blk_run_queue on behalf
+ *    of us.
+ */
+void blk_run_queue_async(struct request_queue *q)
+{
+	if (likely(!blk_queue_stopped(q)))
+		queue_delayed_work(kblockd_workqueue, &q->delay_work, 0);
+}
+
 /**
  * blk_run_queue - run a single device queue
  * @q: The queue to run
@@ -328,7 +342,7 @@ void blk_run_queue(struct request_queue *q)
 	unsigned long flags;
 
 	spin_lock_irqsave(q->queue_lock, flags);
-	__blk_run_queue(q, false);
+	__blk_run_queue(q);
 	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 EXPORT_SYMBOL(blk_run_queue);
@@ -977,7 +991,7 @@ void blk_insert_request(struct request_queue *q, struct request *rq,
 		blk_queue_end_tag(q, rq);
 
 	add_acct_request(q, rq, where);
-	__blk_run_queue(q, false);
+	__blk_run_queue(q);
 	spin_unlock_irqrestore(q->queue_lock, flags);
 }
 EXPORT_SYMBOL(blk_insert_request);
@@ -1321,7 +1335,7 @@ static int __make_request(struct request_queue *q, struct bio *bio)
 	} else {
 		spin_lock_irq(q->queue_lock);
 		add_acct_request(q, req, where);
-		__blk_run_queue(q, false);
+		__blk_run_queue(q);
 out_unlock:
 		spin_unlock_irq(q->queue_lock);
 	}
@@ -2638,6 +2652,7 @@ void blk_start_plug(struct blk_plug *plug)
 
 	plug->magic = PLUG_MAGIC;
 	INIT_LIST_HEAD(&plug->list);
+	INIT_LIST_HEAD(&plug->cb_list);
 	plug->should_sort = 0;
 
 	/*
@@ -2670,12 +2685,41 @@ static int plug_rq_cmp(void *priv, struct list_head *a, struct list_head *b)
  */
 static void queue_unplugged(struct request_queue *q, unsigned int depth,
 			    bool from_schedule)
+	__releases(q->queue_lock)
 {
 	trace_block_unplug(q, depth, !from_schedule);
-	__blk_run_queue(q, from_schedule);
 
-	if (q->unplugged_fn)
-		q->unplugged_fn(q);
+	/*
+	 * If we are punting this to kblockd, then we can safely drop
+	 * the queue_lock before waking kblockd (which needs to take
+	 * this lock).
+	 */
+	if (from_schedule) {
+		spin_unlock(q->queue_lock);
+		blk_run_queue_async(q);
+	} else {
+		__blk_run_queue(q);
+		spin_unlock(q->queue_lock);
+	}
+
+}
+
+static void flush_plug_callbacks(struct blk_plug *plug)
+{
+	LIST_HEAD(callbacks);
+
+	if (list_empty(&plug->cb_list))
+		return;
+
+	list_splice_init(&plug->cb_list, &callbacks);
+
+	while (!list_empty(&callbacks)) {
+		struct blk_plug_cb *cb = list_first_entry(&callbacks,
+							  struct blk_plug_cb,
+							  list);
+		list_del(&cb->list);
+		cb->callback(cb);
+	}
 }
 
 void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
@@ -2688,6 +2732,7 @@ void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
 
 	BUG_ON(plug->magic != PLUG_MAGIC);
 
+	flush_plug_callbacks(plug);
 	if (list_empty(&plug->list))
 		return;
 
@@ -2712,10 +2757,11 @@ void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
 		BUG_ON(!(rq->cmd_flags & REQ_ON_PLUG));
 		BUG_ON(!rq->q);
 		if (rq->q != q) {
-			if (q) {
+			/*
+			 * This drops the queue lock
+			 */
+			if (q)
 				queue_unplugged(q, depth, from_schedule);
-				spin_unlock(q->queue_lock);
-			}
 			q = rq->q;
 			depth = 0;
 			spin_lock(q->queue_lock);
@@ -2733,10 +2779,11 @@ void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
 		depth++;
 	}
 
-	if (q) {
+	/*
+	 * This drops the queue lock
+	 */
+	if (q)
 		queue_unplugged(q, depth, from_schedule);
-		spin_unlock(q->queue_lock);
-	}
 
 	local_irq_restore(flags);
 }

block/blk-exec.c

@@ -55,7 +55,7 @@ void blk_execute_rq_nowait(struct request_queue *q, struct gendisk *bd_disk,
 	WARN_ON(irqs_disabled());
 	spin_lock_irq(q->queue_lock);
 	__elv_add_request(q, rq, where);
-	__blk_run_queue(q, false);
+	__blk_run_queue(q);
 	/* the queue is stopped so it won't be plugged+unplugged */
 	if (rq->cmd_type == REQ_TYPE_PM_RESUME)
 		q->request_fn(q);

block/blk-flush.c

@@ -218,7 +218,7 @@ static void flush_end_io(struct request *flush_rq, int error)
 	 * request_fn may confuse the driver.  Always use kblockd.
 	 */
 	if (queued)
-		__blk_run_queue(q, true);
+		blk_run_queue_async(q);
 }
 
 /**
@@ -274,7 +274,7 @@ static void flush_data_end_io(struct request *rq, int error)
 	 * the comment in flush_end_io().
 	 */
 	if (blk_flush_complete_seq(rq, REQ_FSEQ_DATA, error))
-		__blk_run_queue(q, true);
+		blk_run_queue_async(q);
 }
 
 /**

block/blk-settings.c

@@ -790,22 +790,6 @@ void blk_queue_flush(struct request_queue *q, unsigned int flush)
 }
 EXPORT_SYMBOL_GPL(blk_queue_flush);
 
-/**
- * blk_queue_unplugged - register a callback for an unplug event
- * @q:		the request queue for the device
- * @fn:		the function to call
- *
- * Some stacked drivers may need to know when IO is dispatched on an
- * unplug event. By registrering a callback here, they will be notified
- * when someone flushes their on-stack queue plug. The function will be
- * called with the queue lock held.
- */
-void blk_queue_unplugged(struct request_queue *q, unplugged_fn *fn)
-{
-	q->unplugged_fn = fn;
-}
-EXPORT_SYMBOL(blk_queue_unplugged);
-
 static int __init blk_settings_init(void)
 {
 	blk_max_low_pfn = max_low_pfn - 1;

block/blk.h

@@ -22,6 +22,7 @@ void blk_rq_timed_out_timer(unsigned long data);
 void blk_delete_timer(struct request *);
 void blk_add_timer(struct request *);
 void __generic_unplug_device(struct request_queue *);
+void blk_run_queue_async(struct request_queue *q);
 
 /*
  * Internal atomic flags for request handling

block/cfq-iosched.c

@@ -3368,7 +3368,7 @@ cfq_rq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 			    cfqd->busy_queues > 1) {
 				cfq_del_timer(cfqd, cfqq);
 				cfq_clear_cfqq_wait_request(cfqq);
-				__blk_run_queue(cfqd->queue, false);
+				__blk_run_queue(cfqd->queue);
 			} else {
 				cfq_blkiocg_update_idle_time_stats(
 						&cfqq->cfqg->blkg);
@@ -3383,7 +3383,7 @@ cfq_rq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 		 * this new queue is RT and the current one is BE
 		 */
 		cfq_preempt_queue(cfqd, cfqq);
-		__blk_run_queue(cfqd->queue, false);
+		__blk_run_queue(cfqd->queue);
 	}
 }
 
@@ -3743,7 +3743,7 @@ static void cfq_kick_queue(struct work_struct *work)
 	struct request_queue *q = cfqd->queue;
 
 	spin_lock_irq(q->queue_lock);
-	__blk_run_queue(cfqd->queue, false);
+	__blk_run_queue(cfqd->queue);
 	spin_unlock_irq(q->queue_lock);
 }
 

block/elevator.c

@@ -642,7 +642,7 @@ void elv_quiesce_start(struct request_queue *q)
 	 */
 	elv_drain_elevator(q);
 	while (q->rq.elvpriv) {
-		__blk_run_queue(q, false);
+		__blk_run_queue(q);
 		spin_unlock_irq(q->queue_lock);
 		msleep(10);
 		spin_lock_irq(q->queue_lock);
@@ -695,7 +695,7 @@ void __elv_add_request(struct request_queue *q, struct request *rq, int where)
 		 *   with anything.  There's no point in delaying queue
 		 *   processing.
 		 */
-		__blk_run_queue(q, false);
+		__blk_run_queue(q);
 		break;
 
 	case ELEVATOR_INSERT_SORT_MERGE:

drivers/md/dm-raid.c

@@ -390,13 +390,6 @@ static int raid_is_congested(struct dm_target_callbacks *cb, int bits)
 	return md_raid5_congested(&rs->md, bits);
 }
 
-static void raid_unplug(struct dm_target_callbacks *cb)
-{
-	struct raid_set *rs = container_of(cb, struct raid_set, callbacks);
-
-	md_raid5_kick_device(rs->md.private);
-}
-
 /*
  * Construct a RAID4/5/6 mapping:
  * Args:
@@ -487,7 +480,6 @@ static int raid_ctr(struct dm_target *ti, unsigned argc, char **argv)
 	}
 
 	rs->callbacks.congested_fn = raid_is_congested;
-	rs->callbacks.unplug_fn = raid_unplug;
 	dm_table_add_target_callbacks(ti->table, &rs->callbacks);
 
 	return 0;