Merge branch 'master' into for-next

Conflicts:
	fs/exofs/inode.c

13 files changed, 214 insertions, 229 deletions

diff --git a/block/blk-barrier.c b/block/blk-barrier.c
index 0d710c9d403..f0faefca032 100644
--- a/block/blk-barrier.c
+++ b/block/blk-barrier.c
@@ -13,7 +13,6 @@
  * blk_queue_ordered - does this queue support ordered writes
  * @q:        the request queue
  * @ordered:  one of QUEUE_ORDERED_*
- * @prepare_flush_fn: rq setup helper for cache flush ordered writes
  *
  * Description:
  *   For journalled file systems, doing ordered writes on a commit
@@ -22,15 +21,8 @@
  *   feature should call this function and indicate so.
  *
  **/
-int blk_queue_ordered(struct request_queue *q, unsigned ordered,
-		      prepare_flush_fn *prepare_flush_fn)
+int blk_queue_ordered(struct request_queue *q, unsigned ordered)
 {
-	if (!prepare_flush_fn && (ordered & (QUEUE_ORDERED_DO_PREFLUSH |
-					     QUEUE_ORDERED_DO_POSTFLUSH))) {
-		printk(KERN_ERR "%s: prepare_flush_fn required\n", __func__);
-		return -EINVAL;
-	}
-
 	if (ordered != QUEUE_ORDERED_NONE &&
 	    ordered != QUEUE_ORDERED_DRAIN &&
 	    ordered != QUEUE_ORDERED_DRAIN_FLUSH &&
@@ -44,7 +36,6 @@ int blk_queue_ordered(struct request_queue *q, unsigned ordered,
 
 	q->ordered = ordered;
 	q->next_ordered = ordered;
-	q->prepare_flush_fn = prepare_flush_fn;
 
 	return 0;
 }
@@ -79,7 +70,7 @@ unsigned blk_ordered_req_seq(struct request *rq)
 	 *
 	 * http://thread.gmane.org/gmane.linux.kernel/537473
 	 */
-	if (!blk_fs_request(rq))
+	if (rq->cmd_type != REQ_TYPE_FS)
 		return QUEUE_ORDSEQ_DRAIN;
 
 	if ((rq->cmd_flags & REQ_ORDERED_COLOR) ==
@@ -143,10 +134,10 @@ static void queue_flush(struct request_queue *q, unsigned which)
 	}
 
 	blk_rq_init(q, rq);
-	rq->cmd_flags = REQ_HARDBARRIER;
-	rq->rq_disk = q->bar_rq.rq_disk;
+	rq->cmd_type = REQ_TYPE_FS;
+	rq->cmd_flags = REQ_HARDBARRIER | REQ_FLUSH;
+	rq->rq_disk = q->orig_bar_rq->rq_disk;
 	rq->end_io = end_io;
-	q->prepare_flush_fn(q, rq);
 
 	elv_insert(q, rq, ELEVATOR_INSERT_FRONT);
 }
@@ -203,7 +194,7 @@ static inline bool start_ordered(struct request_queue *q, struct request **rqp)
 		/* initialize proxy request and queue it */
 		blk_rq_init(q, rq);
 		if (bio_data_dir(q->orig_bar_rq->bio) == WRITE)
-			rq->cmd_flags |= REQ_RW;
+			rq->cmd_flags |= REQ_WRITE;
 		if (q->ordered & QUEUE_ORDERED_DO_FUA)
 			rq->cmd_flags |= REQ_FUA;
 		init_request_from_bio(rq, q->orig_bar_rq->bio);
@@ -236,7 +227,8 @@ static inline bool start_ordered(struct request_queue *q, struct request **rqp)
 bool blk_do_ordered(struct request_queue *q, struct request **rqp)
 {
 	struct request *rq = *rqp;
-	const int is_barrier = blk_fs_request(rq) && blk_barrier_rq(rq);
+	const int is_barrier = rq->cmd_type == REQ_TYPE_FS &&
+				(rq->cmd_flags & REQ_HARDBARRIER);
 
 	if (!q->ordseq) {
 		if (!is_barrier)
@@ -261,7 +253,7 @@ bool blk_do_ordered(struct request_queue *q, struct request **rqp)
 	 */
 
 	/* Special requests are not subject to ordering rules. */
-	if (!blk_fs_request(rq) &&
+	if (rq->cmd_type != REQ_TYPE_FS &&
 	    rq != &q->pre_flush_rq && rq != &q->post_flush_rq)
 		return true;
 
@@ -319,6 +311,15 @@ int blkdev_issue_flush(struct block_device *bdev, gfp_t gfp_mask,
 	if (!q)
 		return -ENXIO;
 
+	/*
+	 * some block devices may not have their queue correctly set up here
+	 * (e.g. loop device without a backing file) and so issuing a flush
+	 * here will panic. Ensure there is a request function before issuing
+	 * the barrier.
+	 */
+	if (!q->make_request_fn)
+		return -ENXIO;
+
 	bio = bio_alloc(gfp_mask, 0);
 	bio->bi_end_io = bio_end_empty_barrier;
 	bio->bi_bdev = bdev;
diff --git a/block/blk-core.c b/block/blk-core.c
index 7ac24fa71f7..77411486b11 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -184,7 +184,7 @@ void blk_dump_rq_flags(struct request *rq, char *msg)
 	printk(KERN_INFO "  bio %p, biotail %p, buffer %p, len %u\n",
 	       rq->bio, rq->biotail, rq->buffer, blk_rq_bytes(rq));
 
-	if (blk_pc_request(rq)) {
+	if (rq->cmd_type == REQ_TYPE_BLOCK_PC) {
 		printk(KERN_INFO "  cdb: ");
 		for (bit = 0; bit < BLK_MAX_CDB; bit++)
 			printk("%02x ", rq->cmd[bit]);
@@ -608,6 +608,7 @@ blk_init_allocated_queue_node(struct request_queue *q, request_fn_proc *rfn,
 
 	q->request_fn		= rfn;
 	q->prep_rq_fn		= NULL;
+	q->unprep_rq_fn		= NULL;
 	q->unplug_fn		= generic_unplug_device;
 	q->queue_flags		= QUEUE_FLAG_DEFAULT;
 	q->queue_lock		= lock;
@@ -1135,30 +1136,46 @@ void blk_put_request(struct request *req)
 }
 EXPORT_SYMBOL(blk_put_request);
 
+/**
+ * blk_add_request_payload - add a payload to a request
+ * @rq: request to update
+ * @page: page backing the payload
+ * @len: length of the payload.
+ *
+ * This allows to later add a payload to an already submitted request by
+ * a block driver.  The driver needs to take care of freeing the payload
+ * itself.
+ *
+ * Note that this is a quite horrible hack and nothing but handling of
+ * discard requests should ever use it.
+ */
+void blk_add_request_payload(struct request *rq, struct page *page,
+		unsigned int len)
+{
+	struct bio *bio = rq->bio;
+
+	bio->bi_io_vec->bv_page = page;
+	bio->bi_io_vec->bv_offset = 0;
+	bio->bi_io_vec->bv_len = len;
+
+	bio->bi_size = len;
+	bio->bi_vcnt = 1;
+	bio->bi_phys_segments = 1;
+
+	rq->__data_len = rq->resid_len = len;
+	rq->nr_phys_segments = 1;
+	rq->buffer = bio_data(bio);
+}
+EXPORT_SYMBOL_GPL(blk_add_request_payload);
+
 void init_request_from_bio(struct request *req, struct bio *bio)
 {
 	req->cpu = bio->bi_comp_cpu;
 	req->cmd_type = REQ_TYPE_FS;
 
-	/*
-	 * Inherit FAILFAST from bio (for read-ahead, and explicit
-	 * FAILFAST).  FAILFAST flags are identical for req and bio.
-	 */
-	if (bio_rw_flagged(bio, BIO_RW_AHEAD))
+	req->cmd_flags |= bio->bi_rw & REQ_COMMON_MASK;
+	if (bio->bi_rw & REQ_RAHEAD)
 		req->cmd_flags |= REQ_FAILFAST_MASK;
-	else
-		req->cmd_flags |= bio->bi_rw & REQ_FAILFAST_MASK;
-
-	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_HARDBARRIER;
-	if (bio_rw_flagged(bio, BIO_RW_SYNCIO))
-		req->cmd_flags |= REQ_RW_SYNC;
-	if (bio_rw_flagged(bio, BIO_RW_META))
-		req->cmd_flags |= REQ_RW_META;
-	if (bio_rw_flagged(bio, BIO_RW_NOIDLE))
-		req->cmd_flags |= REQ_NOIDLE;
 
 	req->errors = 0;
 	req->__sector = bio->bi_sector;
@@ -1181,12 +1198,12 @@ static int __make_request(struct request_queue *q, struct bio *bio)
 	int el_ret;
 	unsigned int bytes = bio->bi_size;
 	const unsigned short prio = bio_prio(bio);
-	const bool sync = bio_rw_flagged(bio, BIO_RW_SYNCIO);
-	const bool unplug = bio_rw_flagged(bio, BIO_RW_UNPLUG);
+	const bool sync = (bio->bi_rw & REQ_SYNC);
+	const bool unplug = (bio->bi_rw & REQ_UNPLUG);
 	const unsigned int ff = bio->bi_rw & REQ_FAILFAST_MASK;
 	int rw_flags;
 
-	if (bio_rw_flagged(bio, BIO_RW_BARRIER) &&
+	if ((bio->bi_rw & REQ_HARDBARRIER) &&
 	    (q->next_ordered == QUEUE_ORDERED_NONE)) {
 		bio_endio(bio, -EOPNOTSUPP);
 		return 0;
@@ -1200,7 +1217,7 @@ static int __make_request(struct request_queue *q, struct bio *bio)
 
 	spin_lock_irq(q->queue_lock);
 
-	if (unlikely(bio_rw_flagged(bio, BIO_RW_BARRIER)) || elv_queue_empty(q))
+	if (unlikely((bio->bi_rw & REQ_HARDBARRIER)) || elv_queue_empty(q))
 		goto get_rq;
 
 	el_ret = elv_merge(q, &req, bio);
@@ -1275,7 +1292,7 @@ get_rq:
 	 */
 	rw_flags = bio_data_dir(bio);
 	if (sync)
-		rw_flags |= REQ_RW_SYNC;
+		rw_flags |= REQ_SYNC;
 
 	/*
 	 * Grab a free request. This is might sleep but can not fail.
@@ -1464,7 +1481,7 @@ static inline void __generic_make_request(struct bio *bio)
 			goto end_io;
 		}
 
-		if (unlikely(!bio_rw_flagged(bio, BIO_RW_DISCARD) &&
+		if (unlikely(!(bio->bi_rw & REQ_DISCARD) &&
 			     nr_sectors > queue_max_hw_sectors(q))) {
 			printk(KERN_ERR "bio too big device %s (%u > %u)\n",
 			       bdevname(bio->bi_bdev, b),
@@ -1497,8 +1514,7 @@ static inline void __generic_make_request(struct bio *bio)
 		if (bio_check_eod(bio, nr_sectors))
 			goto end_io;
 
-		if (bio_rw_flagged(bio, BIO_RW_DISCARD) &&
-		    !blk_queue_discard(q)) {
+		if ((bio->bi_rw & REQ_DISCARD) && !blk_queue_discard(q)) {
 			err = -EOPNOTSUPP;
 			goto end_io;
 		}
@@ -1583,7 +1599,7 @@ void submit_bio(int rw, struct bio *bio)
 	 * 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) && !(rw & (1 << BIO_RW_DISCARD))) {
+	if (bio_has_data(bio) && !(rw & REQ_DISCARD)) {
 		if (rw & WRITE) {
 			count_vm_events(PGPGOUT, count);
 		} else {
@@ -1628,6 +1644,9 @@ EXPORT_SYMBOL(submit_bio);
  */
 int blk_rq_check_limits(struct request_queue *q, struct request *rq)
 {
+	if (rq->cmd_flags & REQ_DISCARD)
+		return 0;
+
 	if (blk_rq_sectors(rq) > queue_max_sectors(q) ||
 	    blk_rq_bytes(rq) > queue_max_hw_sectors(q) << 9) {
 		printk(KERN_ERR "%s: over max size limit.\n", __func__);
@@ -1796,7 +1815,7 @@ struct request *blk_peek_request(struct request_queue *q)
 			 * sees this request (possibly after
 			 * requeueing).  Notify IO scheduler.
 			 */
-			if (blk_sorted_rq(rq))
+			if (rq->cmd_flags & REQ_SORTED)
 				elv_activate_rq(q, rq);
 
 			/*
@@ -1984,10 +2003,11 @@ bool blk_update_request(struct request *req, int error, unsigned int nr_bytes)
 	 * TODO: tj: This is too subtle.  It would be better to let
 	 * low level drivers do what they see fit.
 	 */
-	if (blk_fs_request(req))
+	if (req->cmd_type == REQ_TYPE_FS)
 		req->errors = 0;
 
-	if (error && (blk_fs_request(req) && !(req->cmd_flags & REQ_QUIET))) {
+	if (error && req->cmd_type == REQ_TYPE_FS &&
+	    !(req->cmd_flags & REQ_QUIET)) {
 		printk(KERN_ERR "end_request: I/O error, dev %s, sector %llu\n",
 				req->rq_disk ? req->rq_disk->disk_name : "?",
 				(unsigned long long)blk_rq_pos(req));
@@ -2074,7 +2094,7 @@ bool blk_update_request(struct request *req, int error, unsigned int nr_bytes)
 	req->buffer = bio_data(req->bio);
 
 	/* update sector only for requests with clear definition of sector */
-	if (blk_fs_request(req) || blk_discard_rq(req))
+	if (req->cmd_type == REQ_TYPE_FS || (req->cmd_flags & REQ_DISCARD))
 		req->__sector += total_bytes >> 9;
 
 	/* mixed attributes always follow the first bio */
@@ -2111,11 +2131,32 @@ static bool blk_update_bidi_request(struct request *rq, int error,
 	    blk_update_request(rq->next_rq, error, bidi_bytes))
 		return true;
 
-	add_disk_randomness(rq->rq_disk);
+	if (blk_queue_add_random(rq->q))
+		add_disk_randomness(rq->rq_disk);
 
 	return false;
 }
 
+/**
+ * blk_unprep_request - unprepare a request
+ * @req:	the request
+ *
+ * This function makes a request ready for complete resubmission (or
+ * completion).  It happens only after all error handling is complete,
+ * so represents the appropriate moment to deallocate any resources
+ * that were allocated to the request in the prep_rq_fn.  The queue
+ * lock is held when calling this.
+ */
+void blk_unprep_request(struct request *req)
+{
+	struct request_queue *q = req->q;
+
+	req->cmd_flags &= ~REQ_DONTPREP;
+	if (q->unprep_rq_fn)
+		q->unprep_rq_fn(q, req);
+}
+EXPORT_SYMBOL_GPL(blk_unprep_request);
+
 /*
  * queue lock must be held
  */
@@ -2126,11 +2167,15 @@ static void blk_finish_request(struct request *req, int error)
 
 	BUG_ON(blk_queued_rq(req));
 
-	if (unlikely(laptop_mode) && blk_fs_request(req))
+	if (unlikely(laptop_mode) && req->cmd_type == REQ_TYPE_FS)
 		laptop_io_completion(&req->q->backing_dev_info);
 
 	blk_delete_timer(req);
 
+	if (req->cmd_flags & REQ_DONTPREP)
+		blk_unprep_request(req);
+
+
 	blk_account_io_done(req);
 
 	if (req->end_io)
@@ -2363,7 +2408,7 @@ void blk_rq_bio_prep(struct request_queue *q, struct request *rq,
 		     struct bio *bio)
 {
 	/* Bit 0 (R/W) is identical in rq->cmd_flags and bio->bi_rw */
-	rq->cmd_flags |= bio->bi_rw & REQ_RW;
+	rq->cmd_flags |= bio->bi_rw & REQ_WRITE;
 
 	if (bio_has_data(bio)) {
 		rq->nr_phys_segments = bio_phys_segments(q, bio);
@@ -2450,6 +2495,8 @@ static void __blk_rq_prep_clone(struct request *dst, struct request *src)
 {
 	dst->cpu = src->cpu;
 	dst->cmd_flags = (rq_data_dir(src) | REQ_NOMERGE);
+	if (src->cmd_flags & REQ_DISCARD)
+		dst->cmd_flags |= REQ_DISCARD;
 	dst->cmd_type = src->cmd_type;
 	dst->__sector = blk_rq_pos(src);
 	dst->__data_len = blk_rq_bytes(src);
diff --git a/block/blk-exec.c b/block/blk-exec.c
index 49557e91f0d..e1672f14840 100644
--- a/block/blk-exec.c
+++ b/block/blk-exec.c
@@ -57,7 +57,7 @@ void blk_execute_rq_nowait(struct request_queue *q, struct gendisk *bd_disk,
 	__elv_add_request(q, rq, where, 1);
 	__generic_unplug_device(q);
 	/* the queue is stopped so it won't be plugged+unplugged */
-	if (blk_pm_resume_request(rq))
+	if (rq->cmd_type == REQ_TYPE_PM_RESUME)
 		q->request_fn(q);
 	spin_unlock_irq(q->queue_lock);
 }
diff --git a/block/blk-lib.c b/block/blk-lib.c
index d0216b9f22d..c1fc55a83ba 100644
--- a/block/blk-lib.c
+++ b/block/blk-lib.c
@@ -19,7 +19,6 @@ static void blkdev_discard_end_io(struct bio *bio, int err)
 
 	if (bio->bi_private)
 		complete(bio->bi_private);
-	__free_page(bio_page(bio));
 
 	bio_put(bio);
 }
@@ -42,8 +41,8 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 	struct request_queue *q = bdev_get_queue(bdev);
 	int type = flags & BLKDEV_IFL_BARRIER ?
 		DISCARD_BARRIER : DISCARD_NOBARRIER;
+	unsigned int max_discard_sectors;
 	struct bio *bio;
-	struct page *page;
 	int ret = 0;
 
 	if (!q)
@@ -52,36 +51,30 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 	if (!blk_queue_discard(q))
 		return -EOPNOTSUPP;
 
-	while (nr_sects && !ret) {
-		unsigned int sector_size = q->limits.logical_block_size;
-		unsigned int max_discard_sectors =
-			min(q->limits.max_discard_sectors, UINT_MAX >> 9);
+	/*
+	 * Ensure that max_discard_sectors is of the proper
+	 * granularity
+	 */
+	max_discard_sectors = min(q->limits.max_discard_sectors, UINT_MAX >> 9);
+	if (q->limits.discard_granularity) {
+		unsigned int disc_sects = q->limits.discard_granularity >> 9;
 
+		max_discard_sectors &= ~(disc_sects - 1);
+	}
+
+	while (nr_sects && !ret) {
 		bio = bio_alloc(gfp_mask, 1);
-		if (!bio)
-			goto out;
+		if (!bio) {
+			ret = -ENOMEM;
+			break;
+		}
+
 		bio->bi_sector = sector;
 		bio->bi_end_io = blkdev_discard_end_io;
 		bio->bi_bdev = bdev;
 		if (flags & BLKDEV_IFL_WAIT)
 			bio->bi_private = &wait;
 
-		/*
-		 * Add a zeroed one-sector payload as that's what
-		 * our current implementations need.  If we'll ever need
-		 * more the interface will need revisiting.
-		 */
-		page = alloc_page(gfp_mask | __GFP_ZERO);
-		if (!page)
-			goto out_free_bio;
-		if (bio_add_pc_page(q, bio, page, sector_size, 0) < sector_size)
-			goto out_free_page;
-
-		/*
-		 * And override the bio size - the way discard works we
-		 * touch many more blocks on disk than the actual payload
-		 * length.
-		 */
 		if (nr_sects > max_discard_sectors) {
 			bio->bi_size = max_discard_sectors << 9;
 			nr_sects -= max_discard_sectors;
@@ -103,13 +96,8 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 			ret = -EIO;
 		bio_put(bio);
 	}
+
 	return ret;
-out_free_page:
-	__free_page(page);
-out_free_bio:
-	bio_put(bio);
-out:
-	return -ENOMEM;
 }
 EXPORT_SYMBOL(blkdev_issue_discard);
 
@@ -157,7 +145,7 @@ static void bio_batch_end_io(struct bio *bio, int err)
 int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
 			sector_t nr_sects, gfp_t gfp_mask, unsigned long flags)
 {
-	int ret = 0;
+	int ret;
 	struct bio *bio;
 	struct bio_batch bb;
 	unsigned int sz, issued = 0;
@@ -175,11 +163,14 @@ int blkdev_issue_zeroout(struct block_device *bdev, sector_t sector,
 			return ret;
 	}
 submit:
+	ret = 0;
 	while (nr_sects != 0) {
 		bio = bio_alloc(gfp_mask,
 				min(nr_sects, (sector_t)BIO_MAX_PAGES));
-		if (!bio)
+		if (!bio) {
+			ret = -ENOMEM;
 			break;
+		}
 
 		bio->bi_sector = sector;
 		bio->bi_bdev   = bdev;
@@ -198,6 +189,7 @@ submit:
 			if (ret < (sz << 9))
 				break;
 		}
+		ret = 0;
 		issued++;
 		submit_bio(WRITE, bio);
 	}
diff --git a/block/blk-map.c b/block/blk-map.c
index 9083cf0180c..c65d7593f7f 100644
--- a/block/blk-map.c
+++ b/block/blk-map.c
@@ -307,7 +307,7 @@ int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf,
 		return PTR_ERR(bio);
 
 	if (rq_data_dir(rq) == WRITE)
-		bio->bi_rw |= (1 << BIO_RW);
+		bio->bi_rw |= (1 << REQ_WRITE);
 
 	if (do_copy)
 		rq->cmd_flags |= REQ_COPY_USER;
diff --git a/block/blk-merge.c b/block/blk-merge.c
index 5e7dc997345..3b0cd424967 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -12,7 +12,6 @@
 static unsigned int __blk_recalc_rq_segments(struct request_queue *q,
 					     struct bio *bio)
 {
-	unsigned int phys_size;
 	struct bio_vec *bv, *bvprv = NULL;
 	int cluster, i, high, highprv = 1;
 	unsigned int seg_size, nr_phys_segs;
@@ -24,7 +23,7 @@ static unsigned int __blk_recalc_rq_segments(struct request_queue *q,
 	fbio = bio;
 	cluster = test_bit(QUEUE_FLAG_CLUSTER, &q->queue_flags);
 	seg_size = 0;
-	phys_size = nr_phys_segs = 0;
+	nr_phys_segs = 0;
 	for_each_bio(bio) {
 		bio_for_each_segment(bv, bio, i) {
 			/*
@@ -180,7 +179,7 @@ new_segment:
 	}
 
 	if (q->dma_drain_size && q->dma_drain_needed(rq)) {
-		if (rq->cmd_flags & REQ_RW)
+		if (rq->cmd_flags & REQ_WRITE)
 			memset(q->dma_drain_buffer, 0, q->dma_drain_size);
 
 		sg->page_link &= ~0x02;
@@ -226,7 +225,7 @@ int ll_back_merge_fn(struct request_queue *q, struct request *req,
 {
 	unsigned short max_sectors;
 
-	if (unlikely(blk_pc_request(req)))
+	if (unlikely(req->cmd_type == REQ_TYPE_BLOCK_PC))
 		max_sectors = queue_max_hw_sectors(q);
 	else
 		max_sectors = queue_max_sectors(q);
@@ -250,7 +249,7 @@ int ll_front_merge_fn(struct request_queue *q, struct request *req,
 {
 	unsigned short max_sectors;
 
-	if (unlikely(blk_pc_request(req)))
+	if (unlikely(req->cmd_type == REQ_TYPE_BLOCK_PC))
 		max_sectors = queue_max_hw_sectors(q);
 	else
 		max_sectors = queue_max_sectors(q);
diff --git a/block/blk-settings.c b/block/blk-settings.c
index f5ed5a1187b..a234f4bf1d6 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -37,6 +37,23 @@ void blk_queue_prep_rq(struct request_queue *q, prep_rq_fn *pfn)
 EXPORT_SYMBOL(blk_queue_prep_rq);
 
 /**
+ * blk_queue_unprep_rq - set an unprepare_request function for queue
+ * @q:		queue
+ * @ufn:	unprepare_request function
+ *
+ * It's possible for a queue to register an unprepare_request callback
+ * which is invoked before the request is finally completed. The goal
+ * of the function is to deallocate any data that was allocated in the
+ * prepare_request callback.
+ *
+ */
+void blk_queue_unprep_rq(struct request_queue *q, unprep_rq_fn *ufn)
+{
+	q->unprep_rq_fn = ufn;
+}
+EXPORT_SYMBOL(blk_queue_unprep_rq);
+
+/**
  * blk_queue_merge_bvec - set a merge_bvec function for queue
  * @q:		queue
  * @mbfn:	merge_bvec_fn
diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index 306759bbdf1..001ab18078f 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -180,26 +180,36 @@ static ssize_t queue_max_hw_sectors_show(struct request_queue *q, char *page)
 	return queue_var_show(max_hw_sectors_kb, (page));
 }
 
-static ssize_t queue_nonrot_show(struct request_queue *q, char *page)
-{
-	return queue_var_show(!blk_queue_nonrot(q), page);
+#define QUEUE_SYSFS_BIT_FNS(name, flag, neg)				\
+static ssize_t								\
+queue_show_##name(struct request_queue *q, char *page)			\
+{									\
+	int bit;							\
+	bit = test_bit(QUEUE_FLAG_##flag, &q->queue_flags);		\
+	return queue_var_show(neg ? !bit : bit, page);			\
+}									\
+static ssize_t								\
+queue_store_##name(struct request_queue *q, const char *page, size_t count) \
+{									\
+	unsigned long val;						\
+	ssize_t ret;							\
+	ret = queue_var_store(&val, page, count);			\
+	if (neg)							\
+		val = !val;						\
+									\
+	spin_lock_irq(q->queue_lock);					\
+	if (val)							\
+		queue_flag_set(QUEUE_FLAG_##flag, q);			\
+	else								\
+		queue_flag_clear(QUEUE_FLAG_##flag, q);			\
+	spin_unlock_irq(q->queue_lock);					\
+	return ret;							\
 }
 
-static ssize_t queue_nonrot_store(struct request_queue *q, const char *page,
-				  size_t count)
-{
-	unsigned long nm;
-	ssize_t ret = queue_var_store(&nm, page, count);
-
-	spin_lock_irq(q->queue_lock);
-	if (nm)
-		queue_flag_clear(QUEUE_FLAG_NONROT, q);
-	else
-		queue_flag_set(QUEUE_FLAG_NONROT, q);
-	spin_unlock_irq(q->queue_lock);
-
-	return ret;
-}
+QUEUE_SYSFS_BIT_FNS(nonrot, NONROT, 1);
+QUEUE_SYSFS_BIT_FNS(random, ADD_RANDOM, 0);
+QUEUE_SYSFS_BIT_FNS(iostats, IO_STAT, 0);
+#undef QUEUE_SYSFS_BIT_FNS
 
 static ssize_t queue_nomerges_show(struct request_queue *q, char *page)
 {
@@ -250,27 +260,6 @@ queue_rq_affinity_store(struct request_queue *q, const char *page, size_t count)
 	return ret;
 }
 
-static ssize_t queue_iostats_show(struct request_queue *q, char *page)
-{
-	return queue_var_show(blk_queue_io_stat(q), page);
-}
-
-static ssize_t queue_iostats_store(struct request_queue *q, const char *page,
-				   size_t count)
-{
-	unsigned long stats;
-	ssize_t ret = queue_var_store(&stats, page, count);
-
-	spin_lock_irq(q->queue_lock);
-	if (stats)
-		queue_flag_set(QUEUE_FLAG_IO_STAT, q);
-	else
-		queue_flag_clear(QUEUE_FLAG_IO_STAT, q);
-	spin_unlock_irq(q->queue_lock);
-
-	return ret;
-}
-
 static struct queue_sysfs_entry queue_requests_entry = {
 	.attr = {.name = "nr_requests", .mode = S_IRUGO | S_IWUSR },
 	.show = queue_requests_show,
@@ -352,8 +341,8 @@ static struct queue_sysfs_entry queue_discard_zeroes_data_entry = {
 
 static struct queue_sysfs_entry queue_nonrot_entry = {
 	.attr = {.name = "rotational", .mode = S_IRUGO | S_IWUSR },
-	.show = queue_nonrot_show,
-	.store = queue_nonrot_store,
+	.show = queue_show_nonrot,
+	.store = queue_store_nonrot,
 };
 
 static struct queue_sysfs_entry queue_nomerges_entry = {
@@ -370,8 +359,14 @@ static struct queue_sysfs_entry queue_rq_affinity_entry = {
 
 static struct queue_sysfs_entry queue_iostats_entry = {
 	.attr = {.name = "iostats", .mode = S_IRUGO | S_IWUSR },
-	.show = queue_iostats_show,
-	.store = queue_iostats_store,
+	.show = queue_show_iostats,
+	.store = queue_store_iostats,
+};
+
+static struct queue_sysfs_entry queue_random_entry = {
+	.attr = {.name = "add_random", .mode = S_IRUGO | S_IWUSR },
+	.show = queue_show_random,
+	.store = queue_store_random,
 };
 
 static struct attribute *default_attrs[] = {
@@ -394,6 +389,7 @@ static struct attribute *default_attrs[] = {
 	&queue_nomerges_entry.attr,
 	&queue_rq_affinity_entry.attr,
 	&queue_iostats_entry.attr,
+	&queue_random_entry.attr,
 	NULL,
 };
 
diff --git a/block/blk.h b/block/blk.h
index 5ee3d7e72fe..6e7dc87141e 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -161,8 +161,10 @@ static inline int blk_cpu_to_group(int cpu)
  */
 static inline int blk_do_io_stat(struct request *rq)
 {
-	return rq->rq_disk && blk_rq_io_stat(rq) &&
-	       (blk_fs_request(rq) || blk_discard_rq(rq));
+	return rq->rq_disk &&
+	       (rq->cmd_flags & REQ_IO_STAT) &&
+	       (rq->cmd_type == REQ_TYPE_FS ||
+	        (rq->cmd_flags & REQ_DISCARD));
 }
 
 #endif
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 7982b830db5..eb4086f7dfe 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -458,7 +458,7 @@ static inline struct cfq_data *cic_to_cfqd(struct cfq_io_context *cic)
  */
 static inline bool cfq_bio_sync(struct bio *bio)
 {
-	return bio_data_dir(bio) == READ || bio_rw_flagged(bio, BIO_RW_SYNCIO);
+	return bio_data_dir(bio) == READ || (bio->bi_rw & REQ_SYNC);
 }
 
 /*
@@ -646,9 +646,10 @@ cfq_choose_req(struct cfq_data *cfqd, struct request *rq1, struct request *rq2,
 		return rq1;
 	else if (rq_is_sync(rq2) && !rq_is_sync(rq1))
 		return rq2;
-	if (rq_is_meta(rq1) && !rq_is_meta(rq2))
+	if ((rq1->cmd_flags & REQ_META) && !(rq2->cmd_flags & REQ_META))
 		return rq1;
-	else if (rq_is_meta(rq2) && !rq_is_meta(rq1))
+	else if ((rq2->cmd_flags & REQ_META) &&
+		 !(rq1->cmd_flags & REQ_META))
 		return rq2;
 
 	s1 = blk_rq_pos(rq1);
@@ -1484,7 +1485,7 @@ static void cfq_remove_request(struct request *rq)
 	cfqq->cfqd->rq_queued--;
 	cfq_blkiocg_update_io_remove_stats(&(RQ_CFQG(rq))->blkg,
 					rq_data_dir(rq), rq_is_sync(rq));
-	if (rq_is_meta(rq)) {
+	if (rq->cmd_flags & REQ_META) {
 		WARN_ON(!cfqq->meta_pending);
 		cfqq->meta_pending--;
 	}
@@ -3176,7 +3177,7 @@ cfq_should_preempt(struct cfq_data *cfqd, struct cfq_queue *new_cfqq,
 	 * So both queues are sync. Let the new request get disk time if
 	 * it's a metadata request and the current queue is doing regular IO.
 	 */
-	if (rq_is_meta(rq) && !cfqq->meta_pending)
+	if ((rq->cmd_flags & REQ_META) && !cfqq->meta_pending)
 		return true;
 
 	/*
@@ -3230,7 +3231,7 @@ cfq_rq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 	struct cfq_io_context *cic = RQ_CIC(rq);
 
 	cfqd->rq_queued++;
-	if (rq_is_meta(rq))
+	if (rq->cmd_flags & REQ_META)
 		cfqq->meta_pending++;
 
 	cfq_update_io_thinktime(cfqd, cic);
@@ -3365,7 +3366,8 @@ static void cfq_completed_request(struct request_queue *q, struct request *rq)
 	unsigned long now;
 
 	now = jiffies;
-	cfq_log_cfqq(cfqd, cfqq, "complete rqnoidle %d", !!rq_noidle(rq));
+	cfq_log_cfqq(cfqd, cfqq, "complete rqnoidle %d",
+		     !!(rq->cmd_flags & REQ_NOIDLE));
 
 	cfq_update_hw_tag(cfqd);
 
@@ -3419,11 +3421,12 @@ static void cfq_completed_request(struct request_queue *q, struct request *rq)
 			cfq_slice_expired(cfqd, 1);
 		else if (sync && cfqq_empty &&
 			 !cfq_close_cooperator(cfqd, cfqq)) {
-			cfqd->noidle_tree_requires_idle |= !rq_noidle(rq);
+			cfqd->noidle_tree_requires_idle |=
+				!(rq->cmd_flags & REQ_NOIDLE);
 			/*
 			 * Idling is enabled for SYNC_WORKLOAD.
 			 * SYNC_NOIDLE_WORKLOAD idles at the end of the tree
-			 * only if we processed at least one !rq_noidle request
+			 * only if we processed at least one !REQ_NOIDLE request
 			 */
 			if (cfqd->serving_type == SYNC_WORKLOAD
 			    || cfqd->noidle_tree_requires_idle
diff --git a/block/compat_ioctl.c b/block/compat_ioctl.c
index f26051f4468..d5308563773 100644
--- a/block/compat_ioctl.c
+++ b/block/compat_ioctl.c
@@ -535,56 +535,6 @@ out:
 	return err;
 }
 
-struct compat_blk_user_trace_setup {
-	char name[32];
-	u16 act_mask;
-	u32 buf_size;
-	u32 buf_nr;
-	compat_u64 start_lba;
-	compat_u64 end_lba;
-	u32 pid;
-};
-#define BLKTRACESETUP32 _IOWR(0x12, 115, struct compat_blk_user_trace_setup)
-
-static int compat_blk_trace_setup(struct block_device *bdev, char __user *arg)
-{
-	struct blk_user_trace_setup buts;
-	struct compat_blk_user_trace_setup cbuts;
-	struct request_queue *q;
-	char b[BDEVNAME_SIZE];
-	int ret;
-
-	q = bdev_get_queue(bdev);
-	if (!q)
-		return -ENXIO;
-
-	if (copy_from_user(&cbuts, arg, sizeof(cbuts)))
-		return -EFAULT;
-
-	bdevname(bdev, b);
-
-	buts = (struct blk_user_trace_setup) {
-		.act_mask = cbuts.act_mask,
-		.buf_size = cbuts.buf_size,
-		.buf_nr = cbuts.buf_nr,
-		.start_lba = cbuts.start_lba,
-		.end_lba = cbuts.end_lba,
-		.pid = cbuts.pid,
-	};
-	memcpy(&buts.name, &cbuts.name, 32);
-
-	mutex_lock(&bdev->bd_mutex);
-	ret = do_blk_trace_setup(q, b, bdev->bd_dev, bdev, &buts);
-	mutex_unlock(&bdev->bd_mutex);
-	if (ret)
-		return ret;
-
-	if (copy_to_user(arg, &buts.name, 32))
-		return -EFAULT;
-
-	return 0;
-}
-
 static int compat_blkdev_driver_ioctl(struct block_device *bdev, fmode_t mode,
 			unsigned cmd, unsigned long arg)
 {
@@ -802,16 +752,10 @@ long compat_blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 		return compat_put_u64(arg, bdev->bd_inode->i_size);
 
 	case BLKTRACESETUP32:
-		lock_kernel();
-		ret = compat_blk_trace_setup(bdev, compat_ptr(arg));
-		unlock_kernel();
-		return ret;
 	case BLKTRACESTART: /* compatible */
 	case BLKTRACESTOP:  /* compatible */
 	case BLKTRACETEARDOWN: /* compatible */
-		lock_kernel();
 		ret = blk_trace_ioctl(bdev, cmd, compat_ptr(arg));
-		unlock_kernel();
 		return ret;
 	default:
 		if (disk->fops->compat_ioctl)
diff --git a/block/elevator.c b/block/elevator.c
index 923a9139106..816a7c8d639 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -79,8 +79,7 @@ int elv_rq_merge_ok(struct request *rq, struct bio *bio)
 	/*
 	 * Don't merge file system requests and discard requests
 	 */
-	if (bio_rw_flagged(bio, BIO_RW_DISCARD) !=
-	    bio_rw_flagged(rq->bio, BIO_RW_DISCARD))
+	if ((bio->bi_rw & REQ_DISCARD) != (rq->bio->bi_rw & REQ_DISCARD))
 		return 0;
 
 	/*
@@ -428,7 +427,8 @@ void elv_dispatch_sort(struct request_queue *q, struct request *rq)
 	list_for_each_prev(entry, &q->queue_head) {
 		struct request *pos = list_entry_rq(entry);
 
-		if (blk_discard_rq(rq) != blk_discard_rq(pos))
+		if ((rq->cmd_flags & REQ_DISCARD) !=
+		    (pos->cmd_flags & REQ_DISCARD))
 			break;
 		if (rq_data_dir(rq) != rq_data_dir(pos))
 			break;
@@ -558,7 +558,7 @@ void elv_requeue_request(struct request_queue *q, struct request *rq)
 	 */
 	if (blk_account_rq(rq)) {
 		q->in_flight[rq_is_sync(rq)]--;
-		if (blk_sorted_rq(rq))
+		if (rq->cmd_flags & REQ_SORTED)
 			elv_deactivate_rq(q, rq);
 	}
 
@@ -644,7 +644,8 @@ void elv_insert(struct request_queue *q, struct request *rq, int where)
 		break;
 
 	case ELEVATOR_INSERT_SORT:
-		BUG_ON(!blk_fs_request(rq) && !blk_discard_rq(rq));
+		BUG_ON(rq->cmd_type != REQ_TYPE_FS &&
+		       !(rq->cmd_flags & REQ_DISCARD));
 		rq->cmd_flags |= REQ_SORTED;
 		q->nr_sorted++;
 		if (rq_mergeable(rq)) {
@@ -716,7 +717,7 @@ void __elv_add_request(struct request_queue *q, struct request *rq, int where,
 		/*
 		 * toggle ordered color
 		 */
-		if (blk_barrier_rq(rq))
+		if (rq->cmd_flags & REQ_HARDBARRIER)
 			q->ordcolor ^= 1;
 
 		/*
@@ -729,7 +730,8 @@ void __elv_add_request(struct request_queue *q, struct request *rq, int where,
 		 * this request is scheduling boundary, update
 		 * end_sector
 		 */
-		if (blk_fs_request(rq) || blk_discard_rq(rq)) {
+		if (rq->cmd_type == REQ_TYPE_FS ||
+		    (rq->cmd_flags & REQ_DISCARD)) {
 			q->end_sector = rq_end_sector(rq);
 			q->boundary_rq = rq;
 		}
@@ -843,7 +845,8 @@ void elv_completed_request(struct request_queue *q, struct request *rq)
 	 */
 	if (blk_account_rq(rq)) {
 		q->in_flight[rq_is_sync(rq)]--;
-		if (blk_sorted_rq(rq) && e->ops->elevator_completed_req_fn)
+		if ((rq->cmd_flags & REQ_SORTED) &&
+		    e->ops->elevator_completed_req_fn)
 			e->ops->elevator_completed_req_fn(q, rq);
 	}
 
diff --git a/block/ioctl.c b/block/ioctl.c
index e8eb679f2f9..09fd7f1ef23 100644
--- a/block/ioctl.c
+++ b/block/ioctl.c
@@ -163,18 +163,10 @@ int __blkdev_driver_ioctl(struct block_device *bdev, fmode_t mode,
 			unsigned cmd, unsigned long arg)
 {
 	struct gendisk *disk = bdev->bd_disk;
-	int ret;
 
 	if (disk->fops->ioctl)
 		return disk->fops->ioctl(bdev, mode, cmd, arg);
 
-	if (disk->fops->locked_ioctl) {
-		lock_kernel();
-		ret = disk->fops->locked_ioctl(bdev, mode, cmd, arg);
-		unlock_kernel();
-		return ret;
-	}
-
 	return -ENOTTY;
 }
 /*
@@ -185,8 +177,7 @@ int __blkdev_driver_ioctl(struct block_device *bdev, fmode_t mode,
 EXPORT_SYMBOL_GPL(__blkdev_driver_ioctl);
 
 /*
- * always keep this in sync with compat_blkdev_ioctl() and
- * compat_blkdev_locked_ioctl()
+ * always keep this in sync with compat_blkdev_ioctl()
  */
 int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
 			unsigned long arg)
@@ -206,10 +197,8 @@ int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
 		if (ret != -EINVAL && ret != -ENOTTY)
 			return ret;
 
-		lock_kernel();
 		fsync_bdev(bdev);
 		invalidate_bdev(bdev);
-		unlock_kernel();
 		return 0;
 
 	case BLKROSET:
@@ -221,9 +210,7 @@ int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
 			return -EACCES;
 		if (get_user(n, (int __user *)(arg)))
 			return -EFAULT;
-		lock_kernel();
 		set_device_ro(bdev, n);
-		unlock_kernel();
 		return 0;
 
 	case BLKDISCARD: {
@@ -309,14 +296,10 @@ int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
 			bd_release(bdev);
 		return ret;
 	case BLKPG:
-		lock_kernel();
 		ret = blkpg_ioctl(bdev, (struct blkpg_ioctl_arg __user *) arg);
-		unlock_kernel();
 		break;
 	case BLKRRPART:
-		lock_kernel();
 		ret = blkdev_reread_part(bdev);
-		unlock_kernel();
 		break;
 	case BLKGETSIZE:
 		size = bdev->bd_inode->i_size;
@@ -329,9 +312,7 @@ int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
 	case BLKTRACESTOP:
 	case BLKTRACESETUP:
 	case BLKTRACETEARDOWN:
-		lock_kernel();
 		ret = blk_trace_ioctl(bdev, cmd, (char __user *) arg);
-		unlock_kernel();
 		break;
 	default:
 		ret = __blkdev_driver_ioctl(bdev, mode, cmd, arg);