diff --git a/block/blk-core.c b/block/blk-core.c
index f84cce4..f0640d7 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1149,13 +1149,10 @@
 	else
 		req->cmd_flags |= bio->bi_rw & REQ_FAILFAST_MASK;
 
-	if (unlikely(bio_rw_flagged(bio, BIO_RW_DISCARD))) {
+	if (bio_rw_flagged(bio, BIO_RW_DISCARD))
 		req->cmd_flags |= REQ_DISCARD;
-		if (bio_rw_flagged(bio, BIO_RW_BARRIER))
-			req->cmd_flags |= REQ_SOFTBARRIER;
-	} else if (unlikely(bio_rw_flagged(bio, BIO_RW_BARRIER)))
+	if (bio_rw_flagged(bio, BIO_RW_BARRIER))
 		req->cmd_flags |= REQ_HARDBARRIER;
-
 	if (bio_rw_flagged(bio, BIO_RW_SYNCIO))
 		req->cmd_flags |= REQ_RW_SYNC;
 	if (bio_rw_flagged(bio, BIO_RW_META))
@@ -1586,7 +1583,7 @@
 	 * If it's a regular read/write or a barrier with data attached,
 	 * go through the normal accounting stuff before submission.
 	 */
-	if (bio_has_data(bio)) {
+	if (bio_has_data(bio) && !(rw & (1 << BIO_RW_DISCARD))) {
 		if (rw & WRITE) {
 			count_vm_events(PGPGOUT, count);
 		} else {
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 5ff4f48..7982b83 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -14,7 +14,7 @@
 #include <linux/rbtree.h>
 #include <linux/ioprio.h>
 #include <linux/blktrace_api.h>
-#include "blk-cgroup.h"
+#include "cfq.h"
 
 /*
  * tunables
@@ -879,7 +879,7 @@
 	if (!RB_EMPTY_NODE(&cfqg->rb_node))
 		cfq_rb_erase(&cfqg->rb_node, st);
 	cfqg->saved_workload_slice = 0;
-	blkiocg_update_dequeue_stats(&cfqg->blkg, 1);
+	cfq_blkiocg_update_dequeue_stats(&cfqg->blkg, 1);
 }
 
 static inline unsigned int cfq_cfqq_slice_usage(struct cfq_queue *cfqq)
@@ -939,8 +939,8 @@
 
 	cfq_log_cfqg(cfqd, cfqg, "served: vt=%llu min_vt=%llu", cfqg->vdisktime,
 					st->min_vdisktime);
-	blkiocg_update_timeslice_used(&cfqg->blkg, used_sl);
-	blkiocg_set_start_empty_time(&cfqg->blkg);
+	cfq_blkiocg_update_timeslice_used(&cfqg->blkg, used_sl);
+	cfq_blkiocg_set_start_empty_time(&cfqg->blkg);
 }
 
 #ifdef CONFIG_CFQ_GROUP_IOSCHED
@@ -995,7 +995,7 @@
 
 	/* Add group onto cgroup list */
 	sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor);
-	blkiocg_add_blkio_group(blkcg, &cfqg->blkg, (void *)cfqd,
+	cfq_blkiocg_add_blkio_group(blkcg, &cfqg->blkg, (void *)cfqd,
 					MKDEV(major, minor));
 	cfqg->weight = blkcg_get_weight(blkcg, cfqg->blkg.dev);
 
@@ -1079,7 +1079,7 @@
 		 * it from cgroup list, then it will take care of destroying
 		 * cfqg also.
 		 */
-		if (!blkiocg_del_blkio_group(&cfqg->blkg))
+		if (!cfq_blkiocg_del_blkio_group(&cfqg->blkg))
 			cfq_destroy_cfqg(cfqd, cfqg);
 	}
 }
@@ -1421,10 +1421,10 @@
 {
 	elv_rb_del(&cfqq->sort_list, rq);
 	cfqq->queued[rq_is_sync(rq)]--;
-	blkiocg_update_io_remove_stats(&(RQ_CFQG(rq))->blkg, rq_data_dir(rq),
-						rq_is_sync(rq));
+	cfq_blkiocg_update_io_remove_stats(&(RQ_CFQG(rq))->blkg,
+					rq_data_dir(rq), rq_is_sync(rq));
 	cfq_add_rq_rb(rq);
-	blkiocg_update_io_add_stats(&(RQ_CFQG(rq))->blkg,
+	cfq_blkiocg_update_io_add_stats(&(RQ_CFQG(rq))->blkg,
 			&cfqq->cfqd->serving_group->blkg, rq_data_dir(rq),
 			rq_is_sync(rq));
 }
@@ -1482,8 +1482,8 @@
 	cfq_del_rq_rb(rq);
 
 	cfqq->cfqd->rq_queued--;
-	blkiocg_update_io_remove_stats(&(RQ_CFQG(rq))->blkg, rq_data_dir(rq),
-						rq_is_sync(rq));
+	cfq_blkiocg_update_io_remove_stats(&(RQ_CFQG(rq))->blkg,
+					rq_data_dir(rq), rq_is_sync(rq));
 	if (rq_is_meta(rq)) {
 		WARN_ON(!cfqq->meta_pending);
 		cfqq->meta_pending--;
@@ -1518,8 +1518,8 @@
 static void cfq_bio_merged(struct request_queue *q, struct request *req,
 				struct bio *bio)
 {
-	blkiocg_update_io_merged_stats(&(RQ_CFQG(req))->blkg, bio_data_dir(bio),
-					cfq_bio_sync(bio));
+	cfq_blkiocg_update_io_merged_stats(&(RQ_CFQG(req))->blkg,
+					bio_data_dir(bio), cfq_bio_sync(bio));
 }
 
 static void
@@ -1539,8 +1539,8 @@
 	if (cfqq->next_rq == next)
 		cfqq->next_rq = rq;
 	cfq_remove_request(next);
-	blkiocg_update_io_merged_stats(&(RQ_CFQG(rq))->blkg, rq_data_dir(next),
-					rq_is_sync(next));
+	cfq_blkiocg_update_io_merged_stats(&(RQ_CFQG(rq))->blkg,
+					rq_data_dir(next), rq_is_sync(next));
 }
 
 static int cfq_allow_merge(struct request_queue *q, struct request *rq,
@@ -1571,7 +1571,7 @@
 static inline void cfq_del_timer(struct cfq_data *cfqd, struct cfq_queue *cfqq)
 {
 	del_timer(&cfqd->idle_slice_timer);
-	blkiocg_update_idle_time_stats(&cfqq->cfqg->blkg);
+	cfq_blkiocg_update_idle_time_stats(&cfqq->cfqg->blkg);
 }
 
 static void __cfq_set_active_queue(struct cfq_data *cfqd,
@@ -1580,7 +1580,7 @@
 	if (cfqq) {
 		cfq_log_cfqq(cfqd, cfqq, "set_active wl_prio:%d wl_type:%d",
 				cfqd->serving_prio, cfqd->serving_type);
-		blkiocg_update_avg_queue_size_stats(&cfqq->cfqg->blkg);
+		cfq_blkiocg_update_avg_queue_size_stats(&cfqq->cfqg->blkg);
 		cfqq->slice_start = 0;
 		cfqq->dispatch_start = jiffies;
 		cfqq->allocated_slice = 0;
@@ -1911,7 +1911,7 @@
 	sl = cfqd->cfq_slice_idle;
 
 	mod_timer(&cfqd->idle_slice_timer, jiffies + sl);
-	blkiocg_update_set_idle_time_stats(&cfqq->cfqg->blkg);
+	cfq_blkiocg_update_set_idle_time_stats(&cfqq->cfqg->blkg);
 	cfq_log_cfqq(cfqd, cfqq, "arm_idle: %lu", sl);
 }
 
@@ -1931,7 +1931,7 @@
 	elv_dispatch_sort(q, rq);
 
 	cfqd->rq_in_flight[cfq_cfqq_sync(cfqq)]++;
-	blkiocg_update_dispatch_stats(&cfqq->cfqg->blkg, blk_rq_bytes(rq),
+	cfq_blkiocg_update_dispatch_stats(&cfqq->cfqg->blkg, blk_rq_bytes(rq),
 					rq_data_dir(rq), rq_is_sync(rq));
 }
 
@@ -1986,6 +1986,15 @@
 	int process_refs, new_process_refs;
 	struct cfq_queue *__cfqq;
 
+	/*
+	 * If there are no process references on the new_cfqq, then it is
+	 * unsafe to follow the ->new_cfqq chain as other cfqq's in the
+	 * chain may have dropped their last reference (not just their
+	 * last process reference).
+	 */
+	if (!cfqq_process_refs(new_cfqq))
+		return;
+
 	/* Avoid a circular list and skip interim queue merges */
 	while ((__cfqq = new_cfqq->new_cfqq)) {
 		if (__cfqq == cfqq)
@@ -1994,17 +2003,17 @@
 	}
 
 	process_refs = cfqq_process_refs(cfqq);
+	new_process_refs = cfqq_process_refs(new_cfqq);
 	/*
 	 * If the process for the cfqq has gone away, there is no
 	 * sense in merging the queues.
 	 */
-	if (process_refs == 0)
+	if (process_refs == 0 || new_process_refs == 0)
 		return;
 
 	/*
 	 * Merge in the direction of the lesser amount of work.
 	 */
-	new_process_refs = cfqq_process_refs(new_cfqq);
 	if (new_process_refs >= process_refs) {
 		cfqq->new_cfqq = new_cfqq;
 		atomic_add(process_refs, &new_cfqq->ref);
@@ -3248,7 +3257,7 @@
 				cfq_clear_cfqq_wait_request(cfqq);
 				__blk_run_queue(cfqd->queue);
 			} else {
-				blkiocg_update_idle_time_stats(
+				cfq_blkiocg_update_idle_time_stats(
 						&cfqq->cfqg->blkg);
 				cfq_mark_cfqq_must_dispatch(cfqq);
 			}
@@ -3276,7 +3285,7 @@
 	rq_set_fifo_time(rq, jiffies + cfqd->cfq_fifo_expire[rq_is_sync(rq)]);
 	list_add_tail(&rq->queuelist, &cfqq->fifo);
 	cfq_add_rq_rb(rq);
-	blkiocg_update_io_add_stats(&(RQ_CFQG(rq))->blkg,
+	cfq_blkiocg_update_io_add_stats(&(RQ_CFQG(rq))->blkg,
 			&cfqd->serving_group->blkg, rq_data_dir(rq),
 			rq_is_sync(rq));
 	cfq_rq_enqueued(cfqd, cfqq, rq);
@@ -3364,9 +3373,9 @@
 	WARN_ON(!cfqq->dispatched);
 	cfqd->rq_in_driver--;
 	cfqq->dispatched--;
-	blkiocg_update_completion_stats(&cfqq->cfqg->blkg, rq_start_time_ns(rq),
-			rq_io_start_time_ns(rq), rq_data_dir(rq),
-			rq_is_sync(rq));
+	cfq_blkiocg_update_completion_stats(&cfqq->cfqg->blkg,
+			rq_start_time_ns(rq), rq_io_start_time_ns(rq),
+			rq_data_dir(rq), rq_is_sync(rq));
 
 	cfqd->rq_in_flight[cfq_cfqq_sync(cfqq)]--;
 
@@ -3730,7 +3739,7 @@
 
 	cfq_put_async_queues(cfqd);
 	cfq_release_cfq_groups(cfqd);
-	blkiocg_del_blkio_group(&cfqd->root_group.blkg);
+	cfq_blkiocg_del_blkio_group(&cfqd->root_group.blkg);
 
 	spin_unlock_irq(q->queue_lock);
 
@@ -3798,8 +3807,8 @@
 	 */
 	atomic_set(&cfqg->ref, 1);
 	rcu_read_lock();
-	blkiocg_add_blkio_group(&blkio_root_cgroup, &cfqg->blkg, (void *)cfqd,
-					0);
+	cfq_blkiocg_add_blkio_group(&blkio_root_cgroup, &cfqg->blkg,
+					(void *)cfqd, 0);
 	rcu_read_unlock();
 #endif
 	/*
diff --git a/block/cfq.h b/block/cfq.h
new file mode 100644
index 0000000..93448e5
--- /dev/null
+++ b/block/cfq.h
@@ -0,0 +1,115 @@
+#ifndef _CFQ_H
+#define _CFQ_H
+#include "blk-cgroup.h"
+
+#ifdef CONFIG_CFQ_GROUP_IOSCHED
+static inline void cfq_blkiocg_update_io_add_stats(struct blkio_group *blkg,
+	struct blkio_group *curr_blkg, bool direction, bool sync)
+{
+	blkiocg_update_io_add_stats(blkg, curr_blkg, direction, sync);
+}
+
+static inline void cfq_blkiocg_update_dequeue_stats(struct blkio_group *blkg,
+			unsigned long dequeue)
+{
+	blkiocg_update_dequeue_stats(blkg, dequeue);
+}
+
+static inline void cfq_blkiocg_update_timeslice_used(struct blkio_group *blkg,
+			unsigned long time)
+{
+	blkiocg_update_timeslice_used(blkg, time);
+}
+
+static inline void cfq_blkiocg_set_start_empty_time(struct blkio_group *blkg)
+{
+	blkiocg_set_start_empty_time(blkg);
+}
+
+static inline void cfq_blkiocg_update_io_remove_stats(struct blkio_group *blkg,
+				bool direction, bool sync)
+{
+	blkiocg_update_io_remove_stats(blkg, direction, sync);
+}
+
+static inline void cfq_blkiocg_update_io_merged_stats(struct blkio_group *blkg,
+		bool direction, bool sync)
+{
+	blkiocg_update_io_merged_stats(blkg, direction, sync);
+}
+
+static inline void cfq_blkiocg_update_idle_time_stats(struct blkio_group *blkg)
+{
+	blkiocg_update_idle_time_stats(blkg);
+}
+
+static inline void
+cfq_blkiocg_update_avg_queue_size_stats(struct blkio_group *blkg)
+{
+	blkiocg_update_avg_queue_size_stats(blkg);
+}
+
+static inline void
+cfq_blkiocg_update_set_idle_time_stats(struct blkio_group *blkg)
+{
+	blkiocg_update_set_idle_time_stats(blkg);
+}
+
+static inline void cfq_blkiocg_update_dispatch_stats(struct blkio_group *blkg,
+				uint64_t bytes, bool direction, bool sync)
+{
+	blkiocg_update_dispatch_stats(blkg, bytes, direction, sync);
+}
+
+static inline void cfq_blkiocg_update_completion_stats(struct blkio_group *blkg, uint64_t start_time, uint64_t io_start_time, bool direction, bool sync)
+{
+	blkiocg_update_completion_stats(blkg, start_time, io_start_time,
+				direction, sync);
+}
+
+static inline void cfq_blkiocg_add_blkio_group(struct blkio_cgroup *blkcg,
+			struct blkio_group *blkg, void *key, dev_t dev) {
+	blkiocg_add_blkio_group(blkcg, blkg, key, dev);
+}
+
+static inline int cfq_blkiocg_del_blkio_group(struct blkio_group *blkg)
+{
+	return blkiocg_del_blkio_group(blkg);
+}
+
+#else /* CFQ_GROUP_IOSCHED */
+static inline void cfq_blkiocg_update_io_add_stats(struct blkio_group *blkg,
+	struct blkio_group *curr_blkg, bool direction, bool sync) {}
+
+static inline void cfq_blkiocg_update_dequeue_stats(struct blkio_group *blkg,
+			unsigned long dequeue) {}
+
+static inline void cfq_blkiocg_update_timeslice_used(struct blkio_group *blkg,
+			unsigned long time) {}
+static inline void cfq_blkiocg_set_start_empty_time(struct blkio_group *blkg) {}
+static inline void cfq_blkiocg_update_io_remove_stats(struct blkio_group *blkg,
+				bool direction, bool sync) {}
+static inline void cfq_blkiocg_update_io_merged_stats(struct blkio_group *blkg,
+		bool direction, bool sync) {}
+static inline void cfq_blkiocg_update_idle_time_stats(struct blkio_group *blkg)
+{
+}
+static inline void
+cfq_blkiocg_update_avg_queue_size_stats(struct blkio_group *blkg) {}
+
+static inline void
+cfq_blkiocg_update_set_idle_time_stats(struct blkio_group *blkg) {}
+
+static inline void cfq_blkiocg_update_dispatch_stats(struct blkio_group *blkg,
+				uint64_t bytes, bool direction, bool sync) {}
+static inline void cfq_blkiocg_update_completion_stats(struct blkio_group *blkg, uint64_t start_time, uint64_t io_start_time, bool direction, bool sync) {}
+
+static inline void cfq_blkiocg_add_blkio_group(struct blkio_cgroup *blkcg,
+			struct blkio_group *blkg, void *key, dev_t dev) {}
+static inline int cfq_blkiocg_del_blkio_group(struct blkio_group *blkg)
+{
+	return 0;
+}
+
+#endif /* CFQ_GROUP_IOSCHED */
+#endif
diff --git a/drivers/block/cciss_scsi.c b/drivers/block/cciss_scsi.c
index 3381505..72dae92 100644
--- a/drivers/block/cciss_scsi.c
+++ b/drivers/block/cciss_scsi.c
@@ -861,6 +861,7 @@
 	sh->n_io_port = 0;	// I don't think we use these two...
 	sh->this_id = SELF_SCSI_ID;  
 	sh->sg_tablesize = hba[ctlr]->maxsgentries;
+	sh->max_cmd_len = MAX_COMMAND_SIZE;
 
 	((struct cciss_scsi_adapter_data_t *) 
 		hba[ctlr]->scsi_ctlr)->scsi_host = sh;
diff --git a/drivers/block/cpqarray.c b/drivers/block/cpqarray.c
index 91d1163..abb4ec6 100644
--- a/drivers/block/cpqarray.c
+++ b/drivers/block/cpqarray.c
@@ -386,7 +386,7 @@
 }
 
 /* pdev is NULL for eisa */
-static int __init cpqarray_register_ctlr( int i, struct pci_dev *pdev)
+static int __devinit cpqarray_register_ctlr( int i, struct pci_dev *pdev)
 {
 	struct request_queue *q;
 	int j;
@@ -503,7 +503,7 @@
 	return -1;
 }
 
-static int __init cpqarray_init_one( struct pci_dev *pdev,
+static int __devinit cpqarray_init_one( struct pci_dev *pdev,
 	const struct pci_device_id *ent)
 {
 	int i;
@@ -740,7 +740,7 @@
 /*
  * Find an EISA controller's signature.  Set up an hba if we find it.
  */
-static int __init cpqarray_eisa_detect(void)
+static int __devinit cpqarray_eisa_detect(void)
 {
 	int i=0, j;
 	__u32 board_id;
diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index 6b077f9..7258c95 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -1236,8 +1236,6 @@
 	/* Last part of the attaching process ... */
 	if (ns.conn >= C_CONNECTED &&
 	    os.disk == D_ATTACHING && ns.disk == D_NEGOTIATING) {
-		kfree(mdev->p_uuid); /* We expect to receive up-to-date UUIDs soon. */
-		mdev->p_uuid = NULL; /* ...to not use the old ones in the mean time */
 		drbd_send_sizes(mdev, 0, 0);  /* to start sync... */
 		drbd_send_uuids(mdev);
 		drbd_send_state(mdev);
diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c
index 632e324..2151f18 100644
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -1114,6 +1114,12 @@
 		mdev->new_state_tmp.i = ns.i;
 		ns.i = os.i;
 		ns.disk = D_NEGOTIATING;
+
+		/* We expect to receive up-to-date UUIDs soon.
+		   To avoid a race in receive_state, free p_uuid while
+		   holding req_lock. I.e. atomic with the state change */
+		kfree(mdev->p_uuid);
+		mdev->p_uuid = NULL;
 	}
 
 	rv = _drbd_set_state(mdev, ns, CS_VERBOSE, NULL);
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 1d1088f..0609607 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -63,24 +63,16 @@
 };
 
 enum {
-	WS_USED_B = 0,
-	WS_ONSTACK_B,
+	WS_INPROGRESS = 0,
+	WS_ONSTACK,
 };
 
-#define WS_USED (1 << WS_USED_B)
-#define WS_ONSTACK (1 << WS_ONSTACK_B)
-
-static inline bool bdi_work_on_stack(struct bdi_work *work)
-{
-	return test_bit(WS_ONSTACK_B, &work->state);
-}
-
 static inline void bdi_work_init(struct bdi_work *work,
 				 struct wb_writeback_args *args)
 {
 	INIT_RCU_HEAD(&work->rcu_head);
 	work->args = *args;
-	work->state = WS_USED;
+	__set_bit(WS_INPROGRESS, &work->state);
 }
 
 /**
@@ -95,43 +87,16 @@
 	return !list_empty(&bdi->work_list);
 }
 
-static void bdi_work_clear(struct bdi_work *work)
-{
-	clear_bit(WS_USED_B, &work->state);
-	smp_mb__after_clear_bit();
-	/*
-	 * work can have disappeared at this point. bit waitq functions
-	 * should be able to tolerate this, provided bdi_sched_wait does
-	 * not dereference it's pointer argument.
-	*/
-	wake_up_bit(&work->state, WS_USED_B);
-}
-
 static void bdi_work_free(struct rcu_head *head)
 {
 	struct bdi_work *work = container_of(head, struct bdi_work, rcu_head);
 
-	if (!bdi_work_on_stack(work))
+	clear_bit(WS_INPROGRESS, &work->state);
+	smp_mb__after_clear_bit();
+	wake_up_bit(&work->state, WS_INPROGRESS);
+
+	if (!test_bit(WS_ONSTACK, &work->state))
 		kfree(work);
-	else
-		bdi_work_clear(work);
-}
-
-static void wb_work_complete(struct bdi_work *work)
-{
-	const enum writeback_sync_modes sync_mode = work->args.sync_mode;
-	int onstack = bdi_work_on_stack(work);
-
-	/*
-	 * For allocated work, we can clear the done/seen bit right here.
-	 * For on-stack work, we need to postpone both the clear and free
-	 * to after the RCU grace period, since the stack could be invalidated
-	 * as soon as bdi_work_clear() has done the wakeup.
-	 */
-	if (!onstack)
-		bdi_work_clear(work);
-	if (sync_mode == WB_SYNC_NONE || onstack)
-		call_rcu(&work->rcu_head, bdi_work_free);
 }
 
 static void wb_clear_pending(struct bdi_writeback *wb, struct bdi_work *work)
@@ -147,7 +112,7 @@
 		list_del_rcu(&work->list);
 		spin_unlock(&bdi->wb_lock);
 
-		wb_work_complete(work);
+		call_rcu(&work->rcu_head, bdi_work_free);
 	}
 }
 
@@ -185,9 +150,9 @@
  * Used for on-stack allocated work items. The caller needs to wait until
  * the wb threads have acked the work before it's safe to continue.
  */
-static void bdi_wait_on_work_clear(struct bdi_work *work)
+static void bdi_wait_on_work_done(struct bdi_work *work)
 {
-	wait_on_bit(&work->state, WS_USED_B, bdi_sched_wait,
+	wait_on_bit(&work->state, WS_INPROGRESS, bdi_sched_wait,
 		    TASK_UNINTERRUPTIBLE);
 }
 
@@ -213,37 +178,28 @@
 }
 
 /**
- * bdi_sync_writeback - start and wait for writeback
- * @bdi: the backing device to write from
+ * bdi_queue_work_onstack - start and wait for writeback
  * @sb: write inodes from this super_block
  *
  * Description:
- *   This does WB_SYNC_ALL data integrity writeback and waits for the
- *   IO to complete. Callers must hold the sb s_umount semaphore for
+ *   This function initiates writeback and waits for the operation to
+ *   complete. Callers must hold the sb s_umount semaphore for
  *   reading, to avoid having the super disappear before we are done.
  */
-static void bdi_sync_writeback(struct backing_dev_info *bdi,
-			       struct super_block *sb)
+static void bdi_queue_work_onstack(struct wb_writeback_args *args)
 {
-	struct wb_writeback_args args = {
-		.sb		= sb,
-		.sync_mode	= WB_SYNC_ALL,
-		.nr_pages	= LONG_MAX,
-		.range_cyclic	= 0,
-	};
 	struct bdi_work work;
 
-	bdi_work_init(&work, &args);
-	work.state |= WS_ONSTACK;
+	bdi_work_init(&work, args);
+	__set_bit(WS_ONSTACK, &work.state);
 
-	bdi_queue_work(bdi, &work);
-	bdi_wait_on_work_clear(&work);
+	bdi_queue_work(args->sb->s_bdi, &work);
+	bdi_wait_on_work_done(&work);
 }
 
 /**
  * bdi_start_writeback - start writeback
  * @bdi: the backing device to write from
- * @sb: write inodes from this super_block
  * @nr_pages: the number of pages to write
  *
  * Description:
@@ -252,25 +208,34 @@
  *   completion. Caller need not hold sb s_umount semaphore.
  *
  */
-void bdi_start_writeback(struct backing_dev_info *bdi, struct super_block *sb,
-			 long nr_pages)
+void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages)
 {
 	struct wb_writeback_args args = {
-		.sb		= sb,
 		.sync_mode	= WB_SYNC_NONE,
 		.nr_pages	= nr_pages,
 		.range_cyclic	= 1,
 	};
 
-	/*
-	 * We treat @nr_pages=0 as the special case to do background writeback,
-	 * ie. to sync pages until the background dirty threshold is reached.
-	 */
-	if (!nr_pages) {
-		args.nr_pages = LONG_MAX;
-		args.for_background = 1;
-	}
+	bdi_alloc_queue_work(bdi, &args);
+}
 
+/**
+ * bdi_start_background_writeback - start background writeback
+ * @bdi: the backing device to write from
+ *
+ * Description:
+ *   This does WB_SYNC_NONE background writeback. The IO is only
+ *   started when this function returns, we make no guarentees on
+ *   completion. Caller need not hold sb s_umount semaphore.
+ */
+void bdi_start_background_writeback(struct backing_dev_info *bdi)
+{
+	struct wb_writeback_args args = {
+		.sync_mode	= WB_SYNC_NONE,
+		.nr_pages	= LONG_MAX,
+		.for_background = 1,
+		.range_cyclic	= 1,
+	};
 	bdi_alloc_queue_work(bdi, &args);
 }
 
@@ -561,48 +526,30 @@
 	return ret;
 }
 
-static void unpin_sb_for_writeback(struct super_block *sb)
-{
-	up_read(&sb->s_umount);
-	put_super(sb);
-}
-
-enum sb_pin_state {
-	SB_PINNED,
-	SB_NOT_PINNED,
-	SB_PIN_FAILED
-};
-
 /*
- * For WB_SYNC_NONE writeback, the caller does not have the sb pinned
+ * For background writeback the caller does not have the sb pinned
  * before calling writeback. So make sure that we do pin it, so it doesn't
  * go away while we are writing inodes from it.
  */
-static enum sb_pin_state pin_sb_for_writeback(struct writeback_control *wbc,
-					      struct super_block *sb)
+static bool pin_sb_for_writeback(struct super_block *sb)
 {
-	/*
-	 * Caller must already hold the ref for this
-	 */
-	if (wbc->sync_mode == WB_SYNC_ALL) {
-		WARN_ON(!rwsem_is_locked(&sb->s_umount));
-		return SB_NOT_PINNED;
-	}
 	spin_lock(&sb_lock);
+	if (list_empty(&sb->s_instances)) {
+		spin_unlock(&sb_lock);
+		return false;
+	}
+
 	sb->s_count++;
+	spin_unlock(&sb_lock);
+
 	if (down_read_trylock(&sb->s_umount)) {
-		if (sb->s_root) {
-			spin_unlock(&sb_lock);
-			return SB_PINNED;
-		}
-		/*
-		 * umounted, drop rwsem again and fall through to failure
-		 */
+		if (sb->s_root)
+			return true;
 		up_read(&sb->s_umount);
 	}
-	sb->s_count--;
-	spin_unlock(&sb_lock);
-	return SB_PIN_FAILED;
+
+	put_super(sb);
+	return false;
 }
 
 /*
@@ -681,24 +628,31 @@
 		struct inode *inode = list_entry(wb->b_io.prev,
 						 struct inode, i_list);
 		struct super_block *sb = inode->i_sb;
-		enum sb_pin_state state;
 
-		if (wbc->sb && sb != wbc->sb) {
-			/* super block given and doesn't
-			   match, skip this inode */
-			redirty_tail(inode);
-			continue;
+		if (wbc->sb) {
+			/*
+			 * We are requested to write out inodes for a specific
+			 * superblock.  This means we already have s_umount
+			 * taken by the caller which also waits for us to
+			 * complete the writeout.
+			 */
+			if (sb != wbc->sb) {
+				redirty_tail(inode);
+				continue;
+			}
+
+			WARN_ON(!rwsem_is_locked(&sb->s_umount));
+
+			ret = writeback_sb_inodes(sb, wb, wbc);
+		} else {
+			if (!pin_sb_for_writeback(sb)) {
+				requeue_io(inode);
+				continue;
+			}
+			ret = writeback_sb_inodes(sb, wb, wbc);
+			drop_super(sb);
 		}
-		state = pin_sb_for_writeback(wbc, sb);
 
-		if (state == SB_PIN_FAILED) {
-			requeue_io(inode);
-			continue;
-		}
-		ret = writeback_sb_inodes(sb, wb, wbc);
-
-		if (state == SB_PINNED)
-			unpin_sb_for_writeback(sb);
 		if (ret)
 			break;
 	}
@@ -911,7 +865,7 @@
 		 * If this isn't a data integrity operation, just notify
 		 * that we have seen this work and we are now starting it.
 		 */
-		if (args.sync_mode == WB_SYNC_NONE)
+		if (!test_bit(WS_ONSTACK, &work->state))
 			wb_clear_pending(wb, work);
 
 		wrote += wb_writeback(wb, &args);
@@ -920,7 +874,7 @@
 		 * This is a data integrity writeback, so only do the
 		 * notification when we have completed the work.
 		 */
-		if (args.sync_mode == WB_SYNC_ALL)
+		if (test_bit(WS_ONSTACK, &work->state))
 			wb_clear_pending(wb, work);
 	}
 
@@ -978,40 +932,30 @@
 }
 
 /*
- * Schedule writeback for all backing devices. This does WB_SYNC_NONE
- * writeback, for integrity writeback see bdi_sync_writeback().
- */
-static void bdi_writeback_all(struct super_block *sb, long nr_pages)
-{
-	struct wb_writeback_args args = {
-		.sb		= sb,
-		.nr_pages	= nr_pages,
-		.sync_mode	= WB_SYNC_NONE,
-	};
-	struct backing_dev_info *bdi;
-
-	rcu_read_lock();
-
-	list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) {
-		if (!bdi_has_dirty_io(bdi))
-			continue;
-
-		bdi_alloc_queue_work(bdi, &args);
-	}
-
-	rcu_read_unlock();
-}
-
-/*
  * Start writeback of `nr_pages' pages.  If `nr_pages' is zero, write back
  * the whole world.
  */
 void wakeup_flusher_threads(long nr_pages)
 {
-	if (nr_pages == 0)
-		nr_pages = global_page_state(NR_FILE_DIRTY) +
+	struct backing_dev_info *bdi;
+	struct wb_writeback_args args = {
+		.sync_mode	= WB_SYNC_NONE,
+	};
+
+	if (nr_pages) {
+		args.nr_pages = nr_pages;
+	} else {
+		args.nr_pages = global_page_state(NR_FILE_DIRTY) +
 				global_page_state(NR_UNSTABLE_NFS);
-	bdi_writeback_all(NULL, nr_pages);
+	}
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(bdi, &bdi_list, bdi_list) {
+		if (!bdi_has_dirty_io(bdi))
+			continue;
+		bdi_alloc_queue_work(bdi, &args);
+	}
+	rcu_read_unlock();
 }
 
 static noinline void block_dump___mark_inode_dirty(struct inode *inode)
@@ -1218,12 +1162,17 @@
 {
 	unsigned long nr_dirty = global_page_state(NR_FILE_DIRTY);
 	unsigned long nr_unstable = global_page_state(NR_UNSTABLE_NFS);
-	long nr_to_write;
+	struct wb_writeback_args args = {
+		.sb		= sb,
+		.sync_mode	= WB_SYNC_NONE,
+	};
 
-	nr_to_write = nr_dirty + nr_unstable +
+	WARN_ON(!rwsem_is_locked(&sb->s_umount));
+
+	args.nr_pages = nr_dirty + nr_unstable +
 			(inodes_stat.nr_inodes - inodes_stat.nr_unused);
 
-	bdi_start_writeback(sb->s_bdi, sb, nr_to_write);
+	bdi_queue_work_onstack(&args);
 }
 EXPORT_SYMBOL(writeback_inodes_sb);
 
@@ -1237,7 +1186,9 @@
 int writeback_inodes_sb_if_idle(struct super_block *sb)
 {
 	if (!writeback_in_progress(sb->s_bdi)) {
+		down_read(&sb->s_umount);
 		writeback_inodes_sb(sb);
+		up_read(&sb->s_umount);
 		return 1;
 	} else
 		return 0;
@@ -1253,7 +1204,16 @@
  */
 void sync_inodes_sb(struct super_block *sb)
 {
-	bdi_sync_writeback(sb->s_bdi, sb);
+	struct wb_writeback_args args = {
+		.sb		= sb,
+		.sync_mode	= WB_SYNC_ALL,
+		.nr_pages	= LONG_MAX,
+		.range_cyclic	= 0,
+	};
+
+	WARN_ON(!rwsem_is_locked(&sb->s_umount));
+
+	bdi_queue_work_onstack(&args);
 	wait_sb_inodes(sb);
 }
 EXPORT_SYMBOL(sync_inodes_sb);
diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c
index 076ca50..c8ff0d1 100644
--- a/fs/ubifs/budget.c
+++ b/fs/ubifs/budget.c
@@ -62,7 +62,9 @@
  */
 static void shrink_liability(struct ubifs_info *c, int nr_to_write)
 {
+	down_read(&c->vfs_sb->s_umount);
 	writeback_inodes_sb(c->vfs_sb);
+	up_read(&c->vfs_sb->s_umount);
 }
 
 /**
diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h
index aee5f6c..9ae2889 100644
--- a/include/linux/backing-dev.h
+++ b/include/linux/backing-dev.h
@@ -105,8 +105,8 @@
 int bdi_register_dev(struct backing_dev_info *bdi, dev_t dev);
 void bdi_unregister(struct backing_dev_info *bdi);
 int bdi_setup_and_register(struct backing_dev_info *, char *, unsigned int);
-void bdi_start_writeback(struct backing_dev_info *bdi, struct super_block *sb,
-				long nr_pages);
+void bdi_start_writeback(struct backing_dev_info *bdi, long nr_pages);
+void bdi_start_background_writeback(struct backing_dev_info *bdi);
 int bdi_writeback_task(struct bdi_writeback *wb);
 int bdi_has_dirty_io(struct backing_dev_info *bdi);
 void bdi_arm_supers_timer(void);
diff --git a/include/linux/drbd.h b/include/linux/drbd.h
index 30da4ae..b8d2516 100644
--- a/include/linux/drbd.h
+++ b/include/linux/drbd.h
@@ -53,7 +53,7 @@
 
 
 extern const char *drbd_buildtag(void);
-#define REL_VERSION "8.3.8rc2"
+#define REL_VERSION "8.3.8"
 #define API_VERSION 88
 #define PRO_VERSION_MIN 86
 #define PRO_VERSION_MAX 94
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index bbd396a..54f28bd 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -597,7 +597,7 @@
 	    (!laptop_mode && ((global_page_state(NR_FILE_DIRTY)
 			       + global_page_state(NR_UNSTABLE_NFS))
 					  > background_thresh)))
-		bdi_start_writeback(bdi, NULL, 0);
+		bdi_start_background_writeback(bdi);
 }
 
 void set_page_dirty_balance(struct page *page, int page_mkwrite)
@@ -705,9 +705,8 @@
 	 * We want to write everything out, not just down to the dirty
 	 * threshold
 	 */
-
 	if (bdi_has_dirty_io(&q->backing_dev_info))
-		bdi_start_writeback(&q->backing_dev_info, NULL, nr_pages);
+		bdi_start_writeback(&q->backing_dev_info, nr_pages);
 }
 
 /*