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

* 'for-linus' of git://git.kernel.dk/linux-2.6-block: (41 commits)
  Revert "Seperate read and write statistics of in_flight requests"
  cfq-iosched: don't delay async queue if it hasn't dispatched at all
  block: Topology ioctls
  cfq-iosched: use assigned slice sync value, not default
  cfq-iosched: rename 'desktop' sysfs entry to 'low_latency'
  cfq-iosched: implement slower async initiate and queue ramp up
  cfq-iosched: delay async IO dispatch, if sync IO was just done
  cfq-iosched: add a knob for desktop interactiveness
  Add a tracepoint for block request remapping
  block: allow large discard requests
  block: use normal I/O path for discard requests
  swapfile: avoid NULL pointer dereference in swapon when s_bdev is NULL
  fs/bio.c: move EXPORT* macros to line after function
  Add missing blk_trace_remove_sysfs to be in pair with blk_trace_init_sysfs
  cciss: fix build when !PROC_FS
  block: Do not clamp max_hw_sectors for stacking devices
  block: Set max_sectors correctly for stacking devices
  cciss: cciss_host_attr_groups should be const
  cciss: Dynamically allocate the drive_info_struct for each logical drive.
  cciss: Add usage_count attribute to each logical drive in /sys
  ...

Documentation/ABI/testing/sysfs-bus-pci-devices-cciss

@@ -31,3 +31,31 @@ Date:		March 2009
 Kernel Version: 2.6.30
 Contact:	[email protected]
 Description:	A symbolic link to /sys/block/cciss!cXdY
+
+Where:		/sys/bus/pci/devices/<dev>/ccissX/rescan
+Date:		August 2009
+Kernel Version:	2.6.31
+Contact:	[email protected]
+Description:	Kicks of a rescan of the controller to discover logical
+		drive topology changes.
+
+Where:		/sys/bus/pci/devices/<dev>/ccissX/cXdY/lunid
+Date:		August 2009
+Kernel Version: 2.6.31
+Contact:	[email protected]
+Description:	Displays the 8-byte LUN ID used to address logical
+		drive Y of controller X.
+
+Where:		/sys/bus/pci/devices/<dev>/ccissX/cXdY/raid_level
+Date:		August 2009
+Kernel Version: 2.6.31
+Contact:	[email protected]
+Description:	Displays the RAID level of logical drive Y of
+		controller X.
+
+Where:		/sys/bus/pci/devices/<dev>/ccissX/cXdY/usage_count
+Date:		August 2009
+Kernel Version: 2.6.31
+Contact:	[email protected]
+Description:	Displays the usage count (number of opens) of logical drive Y
+		of controller X.

block/blk-barrier.c

@@ -350,6 +350,7 @@ 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);
 }
@@ -372,30 +373,50 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 	struct request_queue *q = bdev_get_queue(bdev);
 	int type = flags & DISCARD_FL_BARRIER ?
 		DISCARD_BARRIER : DISCARD_NOBARRIER;
+	struct bio *bio;
+	struct page *page;
 	int ret = 0;
 
 	if (!q)
 		return -ENXIO;
 
-	if (!q->prepare_discard_fn)
+	if (!blk_queue_discard(q))
 		return -EOPNOTSUPP;
 
 	while (nr_sects && !ret) {
-		struct bio *bio = bio_alloc(gfp_mask, 0);
-		if (!bio)
-			return -ENOMEM;
+		unsigned int sector_size = q->limits.logical_block_size;
+		unsigned int max_discard_sectors =
+			min(q->limits.max_discard_sectors, UINT_MAX >> 9);
 
+		bio = bio_alloc(gfp_mask, 1);
+		if (!bio)
+			goto out;
+		bio->bi_sector = sector;
 		bio->bi_end_io = blkdev_discard_end_io;
 		bio->bi_bdev = bdev;
 		if (flags & DISCARD_FL_WAIT)
 			bio->bi_private = &wait;
 
-		bio->bi_sector = sector;
+		/*
+		 * 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_KERNEL | __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;
 
-		if (nr_sects > queue_max_hw_sectors(q)) {
-			bio->bi_size = queue_max_hw_sectors(q) << 9;
-			nr_sects -= queue_max_hw_sectors(q);
-			sector += queue_max_hw_sectors(q);
+		/*
+		 * 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;
+			sector += max_discard_sectors;
 		} else {
 			bio->bi_size = nr_sects << 9;
 			nr_sects = 0;
@@ -414,5 +435,11 @@ int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
 		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);

block/blk-core.c

@@ -34,6 +34,7 @@
 #include "blk.h"
 
 EXPORT_TRACEPOINT_SYMBOL_GPL(block_remap);
+EXPORT_TRACEPOINT_SYMBOL_GPL(block_rq_remap);
 EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_complete);
 
 static int __make_request(struct request_queue *q, struct bio *bio);
@@ -69,7 +70,7 @@ static void drive_stat_acct(struct request *rq, int new_io)
 		part_stat_inc(cpu, part, merges[rw]);
 	else {
 		part_round_stats(cpu, part);
-		part_inc_in_flight(part, rw);
+		part_inc_in_flight(part);
 	}
 
 	part_stat_unlock();
@@ -1031,7 +1032,7 @@ static void part_round_stats_single(int cpu, struct hd_struct *part,
 
 	if (part->in_flight) {
 		__part_stat_add(cpu, part, time_in_queue,
-				part_in_flight(part) * (now - part->stamp));
+				part->in_flight * (now - part->stamp));
 		__part_stat_add(cpu, part, io_ticks, (now - part->stamp));
 	}
 	part->stamp = now;
@@ -1124,7 +1125,6 @@ void init_request_from_bio(struct request *req, struct bio *bio)
 		req->cmd_flags |= REQ_DISCARD;
 		if (bio_rw_flagged(bio, BIO_RW_BARRIER))
 			req->cmd_flags |= REQ_SOFTBARRIER;
-		req->q->prepare_discard_fn(req->q, req);
 	} else if (unlikely(bio_rw_flagged(bio, BIO_RW_BARRIER)))
 		req->cmd_flags |= REQ_HARDBARRIER;
 
@@ -1437,7 +1437,8 @@ static inline void __generic_make_request(struct bio *bio)
 			goto end_io;
 		}
 
-		if (unlikely(nr_sectors > queue_max_hw_sectors(q))) {
+		if (unlikely(!bio_rw_flagged(bio, BIO_RW_DISCARD) &&
+			     nr_sectors > queue_max_hw_sectors(q))) {
 			printk(KERN_ERR "bio too big device %s (%u > %u)\n",
 			       bdevname(bio->bi_bdev, b),
 			       bio_sectors(bio),
@@ -1470,7 +1471,7 @@ static inline void __generic_make_request(struct bio *bio)
 			goto end_io;
 
 		if (bio_rw_flagged(bio, BIO_RW_DISCARD) &&
-		    !q->prepare_discard_fn) {
+		    !blk_queue_discard(q)) {
 			err = -EOPNOTSUPP;
 			goto end_io;
 		}
@@ -1738,7 +1739,7 @@ static void blk_account_io_done(struct request *req)
 		part_stat_inc(cpu, part, ios[rw]);
 		part_stat_add(cpu, part, ticks[rw], duration);
 		part_round_stats(cpu, part);
-		part_dec_in_flight(part, rw);
+		part_dec_in_flight(part);
 
 		part_stat_unlock();
 	}
@@ -2491,6 +2492,14 @@ int kblockd_schedule_work(struct request_queue *q, struct work_struct *work)
 }
 EXPORT_SYMBOL(kblockd_schedule_work);
 
+int kblockd_schedule_delayed_work(struct request_queue *q,
+				  struct delayed_work *work,
+				  unsigned long delay)
+{
+	return queue_delayed_work(kblockd_workqueue, work, delay);
+}
+EXPORT_SYMBOL(kblockd_schedule_delayed_work);
+
 int __init blk_dev_init(void)
 {
 	BUILD_BUG_ON(__REQ_NR_BITS > 8 *

block/blk-merge.c

@@ -351,7 +351,7 @@ static void blk_account_io_merge(struct request *req)
 		part = disk_map_sector_rcu(req->rq_disk, blk_rq_pos(req));
 
 		part_round_stats(cpu, part);
-		part_dec_in_flight(part, rq_data_dir(req));
+		part_dec_in_flight(part);
 
 		part_stat_unlock();
 	}

block/blk-settings.c

@@ -33,23 +33,6 @@ void blk_queue_prep_rq(struct request_queue *q, prep_rq_fn *pfn)
 }
 EXPORT_SYMBOL(blk_queue_prep_rq);
 
-/**
- * blk_queue_set_discard - set a discard_sectors function for queue
- * @q:		queue
- * @dfn:	prepare_discard function
- *
- * It's possible for a queue to register a discard callback which is used
- * to transform a discard request into the appropriate type for the
- * hardware. If none is registered, then discard requests are failed
- * with %EOPNOTSUPP.
- *
- */
-void blk_queue_set_discard(struct request_queue *q, prepare_discard_fn *dfn)
-{
-	q->prepare_discard_fn = dfn;
-}
-EXPORT_SYMBOL(blk_queue_set_discard);
-
 /**
  * blk_queue_merge_bvec - set a merge_bvec function for queue
  * @q:		queue
@@ -111,7 +94,9 @@ void blk_set_default_limits(struct queue_limits *lim)
 	lim->max_hw_segments = MAX_HW_SEGMENTS;
 	lim->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
 	lim->max_segment_size = MAX_SEGMENT_SIZE;
-	lim->max_sectors = lim->max_hw_sectors = SAFE_MAX_SECTORS;
+	lim->max_sectors = BLK_DEF_MAX_SECTORS;
+	lim->max_hw_sectors = INT_MAX;
+	lim->max_discard_sectors = SAFE_MAX_SECTORS;
 	lim->logical_block_size = lim->physical_block_size = lim->io_min = 512;
 	lim->bounce_pfn = (unsigned long)(BLK_BOUNCE_ANY >> PAGE_SHIFT);
 	lim->alignment_offset = 0;
@@ -164,6 +149,7 @@ void blk_queue_make_request(struct request_queue *q, make_request_fn *mfn)
 	q->unplug_timer.data = (unsigned long)q;
 
 	blk_set_default_limits(&q->limits);
+	blk_queue_max_sectors(q, SAFE_MAX_SECTORS);
 
 	/*
 	 * If the caller didn't supply a lock, fall back to our embedded
@@ -253,6 +239,18 @@ void blk_queue_max_hw_sectors(struct request_queue *q, unsigned int max_sectors)
 }
 EXPORT_SYMBOL(blk_queue_max_hw_sectors);
 
+/**
+ * blk_queue_max_discard_sectors - set max sectors for a single discard
+ * @q:  the request queue for the device
+ * @max_discard: maximum number of sectors to discard
+ **/
+void blk_queue_max_discard_sectors(struct request_queue *q,
+		unsigned int max_discard_sectors)
+{
+	q->limits.max_discard_sectors = max_discard_sectors;
+}
+EXPORT_SYMBOL(blk_queue_max_discard_sectors);
+
 /**
  * blk_queue_max_phys_segments - set max phys segments for a request for this queue
  * @q:  the request queue for the device

block/blk-sysfs.c

@@ -452,6 +452,7 @@ int blk_register_queue(struct gendisk *disk)
 	if (ret) {
 		kobject_uevent(&q->kobj, KOBJ_REMOVE);
 		kobject_del(&q->kobj);
+		blk_trace_remove_sysfs(disk_to_dev(disk));
 		return ret;
 	}
 
@@ -465,11 +466,11 @@ void blk_unregister_queue(struct gendisk *disk)
 	if (WARN_ON(!q))
 		return;
 
-	if (q->request_fn) {
+	if (q->request_fn)
 		elv_unregister_queue(q);
 
-		kobject_uevent(&q->kobj, KOBJ_REMOVE);
-		kobject_del(&q->kobj);
-		kobject_put(&disk_to_dev(disk)->kobj);
-	}
+	kobject_uevent(&q->kobj, KOBJ_REMOVE);
+	kobject_del(&q->kobj);
+	blk_trace_remove_sysfs(disk_to_dev(disk));
+	kobject_put(&disk_to_dev(disk)->kobj);
 }