Merge branch 'for-2.6.35' of git://git.kernel.dk/linux-2.6-block

* 'for-2.6.35' of git://git.kernel.dk/linux-2.6-block: (86 commits)
  pipe: set lower and upper limit on max pages in the pipe page array
  pipe: add support for shrinking and growing pipes
  drbd: This is now equivalent to drbd release 8.3.8rc1
  drbd: Do not free p_uuid early, this is done in the exit code of the receiver
  drbd: Null pointer deref fix to the large "multi bio rewrite"
  drbd: Fix: Do not detach, if a bio with a barrier fails
  drbd: Ensure to not trigger late-new-UUID creation multiple times
  drbd: Do not Oops when C_STANDALONE when uuid gets generated
  writeback: fix mixed up arguments to bdi_start_writeback()
  writeback: fix problem with !CONFIG_BLOCK compilation
  block: improve automatic native capacity unlocking
  block: use struct parsed_partitions *state universally in partition check code
  block,ide: simplify bdops->set_capacity() to ->unlock_native_capacity()
  block: restart partition scan after resizing a device
  buffer: make invalidate_bdev() drain all percpu LRU add caches
  block: remove all rcu head initializations
  writeback: fixups for !dirty_writeback_centisecs
  writeback: bdi_writeback_task() must set task state before calling schedule()
  writeback: ensure that WB_SYNC_NONE writeback with sb pinned is sync
  drivers/block/drbd: Use kzalloc
  ...

Documentation/cgroups/blkio-controller.txt

@@ -17,6 +17,9 @@ HOWTO
 You can do a very simple testing of running two dd threads in two different
 cgroups. Here is what you can do.
 
+- Enable Block IO controller
+	CONFIG_BLK_CGROUP=y
+
 - Enable group scheduling in CFQ
 	CONFIG_CFQ_GROUP_IOSCHED=y
 
@@ -54,32 +57,52 @@ cgroups. Here is what you can do.
 
 Various user visible config options
 ===================================
-CONFIG_CFQ_GROUP_IOSCHED
-	- Enables group scheduling in CFQ. Currently only 1 level of group
-	  creation is allowed.
-
-CONFIG_DEBUG_CFQ_IOSCHED
-	- Enables some debugging messages in blktrace. Also creates extra
-	  cgroup file blkio.dequeue.
-
-Config options selected automatically
-=====================================
-These config options are not user visible and are selected/deselected
-automatically based on IO scheduler configuration.
-
 CONFIG_BLK_CGROUP
-	- Block IO controller. Selected by CONFIG_CFQ_GROUP_IOSCHED.
+	- Block IO controller.
 
 CONFIG_DEBUG_BLK_CGROUP
-	- Debug help. Selected by CONFIG_DEBUG_CFQ_IOSCHED.
+	- Debug help. Right now some additional stats file show up in cgroup
+	  if this option is enabled.
+
+CONFIG_CFQ_GROUP_IOSCHED
+	- Enables group scheduling in CFQ. Currently only 1 level of group
+	  creation is allowed.
 
 Details of cgroup files
 =======================
 - blkio.weight
-	- Specifies per cgroup weight.
-
+	- Specifies per cgroup weight. This is default weight of the group
+	  on all the devices until and unless overridden by per device rule.
+	  (See blkio.weight_device).
 	  Currently allowed range of weights is from 100 to 1000.
 
+- blkio.weight_device
+	- One can specify per cgroup per device rules using this interface.
+	  These rules override the default value of group weight as specified
+	  by blkio.weight.
+
+	  Following is the format.
+
+	  #echo dev_maj:dev_minor weight > /path/to/cgroup/blkio.weight_device
+	  Configure weight=300 on /dev/sdb (8:16) in this cgroup
+	  # echo 8:16 300 > blkio.weight_device
+	  # cat blkio.weight_device
+	  dev     weight
+	  8:16    300
+
+	  Configure weight=500 on /dev/sda (8:0) in this cgroup
+	  # echo 8:0 500 > blkio.weight_device
+	  # cat blkio.weight_device
+	  dev     weight
+	  8:0     500
+	  8:16    300
+
+	  Remove specific weight for /dev/sda in this cgroup
+	  # echo 8:0 0 > blkio.weight_device
+	  # cat blkio.weight_device
+	  dev     weight
+	  8:16    300
+
 - blkio.time
 	- disk time allocated to cgroup per device in milliseconds. First
 	  two fields specify the major and minor number of the device and
@@ -92,13 +115,105 @@ Details of cgroup files
 	  third field specifies the number of sectors transferred by the
 	  group to/from the device.
 
+- blkio.io_service_bytes
+	- Number of bytes transferred to/from the disk by the group. These
+	  are further divided by the type of operation - read or write, sync
+	  or async. First two fields specify the major and minor number of the
+	  device, third field specifies the operation type and the fourth field
+	  specifies the number of bytes.
+
+- blkio.io_serviced
+	- Number of IOs completed to/from the disk by the group. These
+	  are further divided by the type of operation - read or write, sync
+	  or async. First two fields specify the major and minor number of the
+	  device, third field specifies the operation type and the fourth field
+	  specifies the number of IOs.
+
+- blkio.io_service_time
+	- Total amount of time between request dispatch and request completion
+	  for the IOs done by this cgroup. This is in nanoseconds to make it
+	  meaningful for flash devices too. For devices with queue depth of 1,
+	  this time represents the actual service time. When queue_depth > 1,
+	  that is no longer true as requests may be served out of order. This
+	  may cause the service time for a given IO to include the service time
+	  of multiple IOs when served out of order which may result in total
+	  io_service_time > actual time elapsed. This time is further divided by
+	  the type of operation - read or write, sync or async. First two fields
+	  specify the major and minor number of the device, third field
+	  specifies the operation type and the fourth field specifies the
+	  io_service_time in ns.
+
+- blkio.io_wait_time
+	- Total amount of time the IOs for this cgroup spent waiting in the
+	  scheduler queues for service. This can be greater than the total time
+	  elapsed since it is cumulative io_wait_time for all IOs. It is not a
+	  measure of total time the cgroup spent waiting but rather a measure of
+	  the wait_time for its individual IOs. For devices with queue_depth > 1
+	  this metric does not include the time spent waiting for service once
+	  the IO is dispatched to the device but till it actually gets serviced
+	  (there might be a time lag here due to re-ordering of requests by the
+	  device). This is in nanoseconds to make it meaningful for flash
+	  devices too. This time is further divided by the type of operation -
+	  read or write, sync or async. First two fields specify the major and
+	  minor number of the device, third field specifies the operation type
+	  and the fourth field specifies the io_wait_time in ns.
+
+- blkio.io_merged
+	- Total number of bios/requests merged into requests belonging to this
+	  cgroup. This is further divided by the type of operation - read or
+	  write, sync or async.
+
+- blkio.io_queued
+	- Total number of requests queued up at any given instant for this
+	  cgroup. This is further divided by the type of operation - read or
+	  write, sync or async.
+
+- blkio.avg_queue_size
+	- Debugging aid only enabled if CONFIG_DEBUG_BLK_CGROUP=y.
+	  The average queue size for this cgroup over the entire time of this
+	  cgroup's existence. Queue size samples are taken each time one of the
+	  queues of this cgroup gets a timeslice.
+
+- blkio.group_wait_time
+	- Debugging aid only enabled if CONFIG_DEBUG_BLK_CGROUP=y.
+	  This is the amount of time the cgroup had to wait since it became busy
+	  (i.e., went from 0 to 1 request queued) to get a timeslice for one of
+	  its queues. This is different from the io_wait_time which is the
+	  cumulative total of the amount of time spent by each IO in that cgroup
+	  waiting in the scheduler queue. This is in nanoseconds. If this is
+	  read when the cgroup is in a waiting (for timeslice) state, the stat
+	  will only report the group_wait_time accumulated till the last time it
+	  got a timeslice and will not include the current delta.
+
+- blkio.empty_time
+	- Debugging aid only enabled if CONFIG_DEBUG_BLK_CGROUP=y.
+	  This is the amount of time a cgroup spends without any pending
+	  requests when not being served, i.e., it does not include any time
+	  spent idling for one of the queues of the cgroup. This is in
+	  nanoseconds. If this is read when the cgroup is in an empty state,
+	  the stat will only report the empty_time accumulated till the last
+	  time it had a pending request and will not include the current delta.
+
+- blkio.idle_time
+	- Debugging aid only enabled if CONFIG_DEBUG_BLK_CGROUP=y.
+	  This is the amount of time spent by the IO scheduler idling for a
+	  given cgroup in anticipation of a better request than the exising ones
+	  from other queues/cgroups. This is in nanoseconds. If this is read
+	  when the cgroup is in an idling state, the stat will only report the
+	  idle_time accumulated till the last idle period and will not include
+	  the current delta.
+
 - blkio.dequeue
-	- Debugging aid only enabled if CONFIG_DEBUG_CFQ_IOSCHED=y. This
+	- Debugging aid only enabled if CONFIG_DEBUG_BLK_CGROUP=y. This
 	  gives the statistics about how many a times a group was dequeued
 	  from service tree of the device. First two fields specify the major
 	  and minor number of the device and third field specifies the number
 	  of times a group was dequeued from a particular device.
 
+- blkio.reset_stats
+	- Writing an int to this file will result in resetting all the stats
+	  for that cgroup.
+
 CFQ sysfs tunable
 =================
 /sys/block/<disk>/queue/iosched/group_isolation

block/Kconfig

@@ -77,29 +77,6 @@ config BLK_DEV_INTEGRITY
 	T10/SCSI Data Integrity Field or the T13/ATA External Path
 	Protection.  If in doubt, say N.
 
-config BLK_CGROUP
-	tristate "Block cgroup support"
-	depends on CGROUPS
-	depends on CFQ_GROUP_IOSCHED
-	default n
-	---help---
-	Generic block IO controller cgroup interface. This is the common
-	cgroup interface which should be used by various IO controlling
-	policies.
-
-	Currently, CFQ IO scheduler uses it to recognize task groups and
-	control disk bandwidth allocation (proportional time slice allocation)
-	to such task groups.
-
-config DEBUG_BLK_CGROUP
-	bool
-	depends on BLK_CGROUP
-	default n
-	---help---
-	Enable some debugging help. Currently it stores the cgroup path
-	in the blk group which can be used by cfq for tracing various
-	group related activity.
-
 endif # BLOCK
 
 config BLOCK_COMPAT

block/Kconfig.iosched

@@ -23,7 +23,8 @@ config IOSCHED_DEADLINE
 
 config IOSCHED_CFQ
 	tristate "CFQ I/O scheduler"
-	select BLK_CGROUP if CFQ_GROUP_IOSCHED
+	# If BLK_CGROUP is a module, CFQ has to be built as module.
+	depends on (BLK_CGROUP=m && m) || !BLK_CGROUP || BLK_CGROUP=y
 	default y
 	---help---
 	  The CFQ I/O scheduler tries to distribute bandwidth equally
@@ -33,22 +34,15 @@ config IOSCHED_CFQ
 
 	  This is the default I/O scheduler.
 
+	  Note: If BLK_CGROUP=m, then CFQ can be built only as module.
+
 config CFQ_GROUP_IOSCHED
 	bool "CFQ Group Scheduling support"
-	depends on IOSCHED_CFQ && CGROUPS
+	depends on IOSCHED_CFQ && BLK_CGROUP
 	default n
 	---help---
 	  Enable group IO scheduling in CFQ.
 
-config DEBUG_CFQ_IOSCHED
-	bool "Debug CFQ Scheduling"
-	depends on CFQ_GROUP_IOSCHED
-	select DEBUG_BLK_CGROUP
-	default n
-	---help---
-	  Enable CFQ IO scheduling debugging in CFQ. Currently it makes
-	  blktrace output more verbose.
-
 choice
 	prompt "Default I/O scheduler"
 	default DEFAULT_CFQ

block/Makefile

@@ -5,7 +5,7 @@
 obj-$(CONFIG_BLOCK) := elevator.o blk-core.o blk-tag.o blk-sysfs.o \
 			blk-barrier.o blk-settings.o blk-ioc.o blk-map.o \
 			blk-exec.o blk-merge.o blk-softirq.o blk-timeout.o \
-			blk-iopoll.o ioctl.o genhd.o scsi_ioctl.o
+			blk-iopoll.o blk-lib.o ioctl.o genhd.o scsi_ioctl.o
 
 obj-$(CONFIG_BLK_DEV_BSG)	+= bsg.o
 obj-$(CONFIG_BLK_CGROUP)	+= blk-cgroup.o

block/blk-barrier.c

@@ -286,26 +286,31 @@ static void bio_end_empty_barrier(struct bio *bio, int err)
 			set_bit(BIO_EOPNOTSUPP, &bio->bi_flags);
 		clear_bit(BIO_UPTODATE, &bio->bi_flags);
 	}
-
+	if (bio->bi_private)
 		complete(bio->bi_private);
+	bio_put(bio);
 }
 
 /**
  * blkdev_issue_flush - queue a flush
  * @bdev:	blockdev to issue flush for
+ * @gfp_mask:	memory allocation flags (for bio_alloc)
  * @error_sector:	error sector
+ * @flags:	BLKDEV_IFL_* flags to control behaviour
  *
  * Description:
  *    Issue a flush for the block device in question. Caller can supply
  *    room for storing the error offset in case of a flush error, if they
- *    wish to.
+ *    wish to. If WAIT flag is not passed then caller may check only what
+ *    request was pushed in some internal queue for later handling.
  */
-int blkdev_issue_flush(struct block_device *bdev, sector_t *error_sector)
+int blkdev_issue_flush(struct block_device *bdev, gfp_t gfp_mask,
+		sector_t *error_sector, unsigned long flags)
 {
 	DECLARE_COMPLETION_ONSTACK(wait);
 	struct request_queue *q;
 	struct bio *bio;
-	int ret;
+	int ret = 0;
 
 	if (bdev->bd_disk == NULL)
 		return -ENXIO;
@@ -314,23 +319,25 @@ int blkdev_issue_flush(struct block_device *bdev, sector_t *error_sector)
 	if (!q)
 		return -ENXIO;
 
-	bio = bio_alloc(GFP_KERNEL, 0);
+	bio = bio_alloc(gfp_mask, 0);
 	bio->bi_end_io = bio_end_empty_barrier;
-	bio->bi_private = &wait;
 	bio->bi_bdev = bdev;
-	submit_bio(WRITE_BARRIER, bio);
+	if (test_bit(BLKDEV_WAIT, &flags))
+		bio->bi_private = &wait;
 
+	bio_get(bio);
+	submit_bio(WRITE_BARRIER, bio);
+	if (test_bit(BLKDEV_WAIT, &flags)) {
 		wait_for_completion(&wait);
-
 		/*
 		 * The driver must store the error location in ->bi_sector, if
-	 * it supports it. For non-stacked drivers, this should be copied
-	 * from blk_rq_pos(rq).
+		 * it supports it. For non-stacked drivers, this should be
+		 * copied from blk_rq_pos(rq).
 		 */
 		if (error_sector)
 			*error_sector = bio->bi_sector;
+	}
 
-	ret = 0;
 	if (bio_flagged(bio, BIO_EOPNOTSUPP))
 		ret = -EOPNOTSUPP;
 	else if (!bio_flagged(bio, BIO_UPTODATE))
@@ -340,107 +347,3 @@ int blkdev_issue_flush(struct block_device *bdev, sector_t *error_sector)
 	return ret;
 }
 EXPORT_SYMBOL(blkdev_issue_flush);
-
-static void blkdev_discard_end_io(struct bio *bio, int err)
-{
-	if (err) {
-		if (err == -EOPNOTSUPP)
-			set_bit(BIO_EOPNOTSUPP, &bio->bi_flags);
-		clear_bit(BIO_UPTODATE, &bio->bi_flags);
-	}
-
-	if (bio->bi_private)
-		complete(bio->bi_private);
-	__free_page(bio_page(bio));
-
-	bio_put(bio);
-}
-
-/**
- * blkdev_issue_discard - queue a discard
- * @bdev:	blockdev to issue discard for
- * @sector:	start sector
- * @nr_sects:	number of sectors to discard
- * @gfp_mask:	memory allocation flags (for bio_alloc)
- * @flags:	DISCARD_FL_* flags to control behaviour
- *
- * Description:
- *    Issue a discard request for the sectors in question.
- */
-int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
-		sector_t nr_sects, gfp_t gfp_mask, int flags)
-{
-	DECLARE_COMPLETION_ONSTACK(wait);
-	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 (!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);
-
-		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;
-
-		/*
-		 * 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;
-			sector += max_discard_sectors;
-		} else {
-			bio->bi_size = nr_sects << 9;
-			nr_sects = 0;
-		}
-
-		bio_get(bio);
-		submit_bio(type, bio);
-
-		if (flags & DISCARD_FL_WAIT)
-			wait_for_completion(&wait);
-
-		if (bio_flagged(bio, BIO_EOPNOTSUPP))
-			ret = -EOPNOTSUPP;
-		else if (!bio_flagged(bio, BIO_UPTODATE))
-			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);

block/blk-cgroup.h

@@ -23,11 +23,84 @@ extern struct cgroup_subsys blkio_subsys;
 #define blkio_subsys_id blkio_subsys.subsys_id
 #endif
 
+enum stat_type {
+	/* Total time spent (in ns) between request dispatch to the driver and
+	 * request completion for IOs doen by this cgroup. This may not be
+	 * accurate when NCQ is turned on. */
+	BLKIO_STAT_SERVICE_TIME = 0,
+	/* Total bytes transferred */
+	BLKIO_STAT_SERVICE_BYTES,
+	/* Total IOs serviced, post merge */
+	BLKIO_STAT_SERVICED,
+	/* Total time spent waiting in scheduler queue in ns */
+	BLKIO_STAT_WAIT_TIME,
+	/* Number of IOs merged */
+	BLKIO_STAT_MERGED,
+	/* Number of IOs queued up */
+	BLKIO_STAT_QUEUED,
+	/* All the single valued stats go below this */
+	BLKIO_STAT_TIME,
+	BLKIO_STAT_SECTORS,
+#ifdef CONFIG_DEBUG_BLK_CGROUP
+	BLKIO_STAT_AVG_QUEUE_SIZE,
+	BLKIO_STAT_IDLE_TIME,
+	BLKIO_STAT_EMPTY_TIME,
+	BLKIO_STAT_GROUP_WAIT_TIME,
+	BLKIO_STAT_DEQUEUE
+#endif
+};
+
+enum stat_sub_type {
+	BLKIO_STAT_READ = 0,
+	BLKIO_STAT_WRITE,
+	BLKIO_STAT_SYNC,
+	BLKIO_STAT_ASYNC,
+	BLKIO_STAT_TOTAL
+};
+
+/* blkg state flags */
+enum blkg_state_flags {
+	BLKG_waiting = 0,
+	BLKG_idling,
+	BLKG_empty,
+};
+
 struct blkio_cgroup {
 	struct cgroup_subsys_state css;
 	unsigned int weight;
 	spinlock_t lock;
 	struct hlist_head blkg_list;
+	struct list_head policy_list; /* list of blkio_policy_node */
+};
+
+struct blkio_group_stats {
+	/* total disk time and nr sectors dispatched by this group */
+	uint64_t time;
+	uint64_t sectors;
+	uint64_t stat_arr[BLKIO_STAT_QUEUED + 1][BLKIO_STAT_TOTAL];
+#ifdef CONFIG_DEBUG_BLK_CGROUP
+	/* Sum of number of IOs queued across all samples */
+	uint64_t avg_queue_size_sum;
+	/* Count of samples taken for average */
+	uint64_t avg_queue_size_samples;
+	/* How many times this group has been removed from service tree */
+	unsigned long dequeue;
+
+	/* Total time spent waiting for it to be assigned a timeslice. */
+	uint64_t group_wait_time;
+	uint64_t start_group_wait_time;
+
+	/* Time spent idling for this blkio_group */
+	uint64_t idle_time;
+	uint64_t start_idle_time;
+	/*
+	 * Total time when we have requests queued and do not contain the
+	 * current active queue.
+	 */
+	uint64_t empty_time;
+	uint64_t start_empty_time;
+	uint16_t flags;
+#endif
 };
 
 struct blkio_group {
@@ -35,20 +108,25 @@ struct blkio_group {
 	void *key;
 	struct hlist_node blkcg_node;
 	unsigned short blkcg_id;
-#ifdef CONFIG_DEBUG_BLK_CGROUP
 	/* Store cgroup path */
 	char path[128];
-	/* How many times this group has been removed from service tree */
-	unsigned long dequeue;
-#endif
 	/* The device MKDEV(major, minor), this group has been created for */
 	dev_t dev;
 
-	/* total disk time and nr sectors dispatched by this group */
-	unsigned long time;
-	unsigned long sectors;
+	/* Need to serialize the stats in the case of reset/update */
+	spinlock_t stats_lock;
+	struct blkio_group_stats stats;
 };
 
+struct blkio_policy_node {
+	struct list_head node;
+	dev_t dev;
+	unsigned int weight;
+};
+
+extern unsigned int blkcg_get_weight(struct blkio_cgroup *blkcg,
+				     dev_t dev);
+
 typedef void (blkio_unlink_group_fn) (void *key, struct blkio_group *blkg);
 typedef void (blkio_update_group_weight_fn) (struct blkio_group *blkg,
 						unsigned int weight);
@@ -67,6 +145,11 @@ struct blkio_policy_type {
 extern void blkio_policy_register(struct blkio_policy_type *);
 extern void blkio_policy_unregister(struct blkio_policy_type *);
 
+static inline char *blkg_path(struct blkio_group *blkg)
+{
+	return blkg->path;
+}
+
 #else
 
 struct blkio_group {
@@ -78,6 +161,8 @@ struct blkio_policy_type {
 static inline void blkio_policy_register(struct blkio_policy_type *blkiop) { }
 static inline void blkio_policy_unregister(struct blkio_policy_type *blkiop) { }
 
+static inline char *blkg_path(struct blkio_group *blkg) { return NULL; }
+
 #endif
 
 #define BLKIO_WEIGHT_MIN	100
@@ -85,16 +170,42 @@ static inline void blkio_policy_unregister(struct blkio_policy_type *blkiop) { }
 #define BLKIO_WEIGHT_DEFAULT	500
 
 #ifdef CONFIG_DEBUG_BLK_CGROUP
-static inline char *blkg_path(struct blkio_group *blkg)
-{
-	return blkg->path;
-}
-void blkiocg_update_blkio_group_dequeue_stats(struct blkio_group *blkg,
+void blkiocg_update_avg_queue_size_stats(struct blkio_group *blkg);
+void blkiocg_update_dequeue_stats(struct blkio_group *blkg,
 				unsigned long dequeue);
+void blkiocg_update_set_idle_time_stats(struct blkio_group *blkg);
+void blkiocg_update_idle_time_stats(struct blkio_group *blkg);
+void blkiocg_set_start_empty_time(struct blkio_group *blkg);
+
+#define BLKG_FLAG_FNS(name)						\
+static inline void blkio_mark_blkg_##name(				\
+		struct blkio_group_stats *stats)			\
+{									\
+	stats->flags |= (1 << BLKG_##name);				\
+}									\
+static inline void blkio_clear_blkg_##name(				\
+		struct blkio_group_stats *stats)			\
+{									\
+	stats->flags &= ~(1 << BLKG_##name);				\
+}									\
+static inline int blkio_blkg_##name(struct blkio_group_stats *stats)	\
+{									\
+	return (stats->flags & (1 << BLKG_##name)) != 0;		\
+}									\
+
+BLKG_FLAG_FNS(waiting)
+BLKG_FLAG_FNS(idling)
+BLKG_FLAG_FNS(empty)
+#undef BLKG_FLAG_FNS
 #else
-static inline char *blkg_path(struct blkio_group *blkg) { return NULL; }
-static inline void blkiocg_update_blkio_group_dequeue_stats(
-			struct blkio_group *blkg, unsigned long dequeue) {}
+static inline void blkiocg_update_avg_queue_size_stats(
+						struct blkio_group *blkg) {}
+static inline void blkiocg_update_dequeue_stats(struct blkio_group *blkg,
+						unsigned long dequeue) {}
+static inline void blkiocg_update_set_idle_time_stats(struct blkio_group *blkg)
+{}
+static inline void blkiocg_update_idle_time_stats(struct blkio_group *blkg) {}
+static inline void blkiocg_set_start_empty_time(struct blkio_group *blkg) {}
 #endif
 
 #if defined(CONFIG_BLK_CGROUP) || defined(CONFIG_BLK_CGROUP_MODULE)
@@ -105,26 +216,43 @@ extern void blkiocg_add_blkio_group(struct blkio_cgroup *blkcg,
 extern int blkiocg_del_blkio_group(struct blkio_group *blkg);
 extern struct blkio_group *blkiocg_lookup_group(struct blkio_cgroup *blkcg,
 						void *key);
-void blkiocg_update_blkio_group_stats(struct blkio_group *blkg,
-			unsigned long time, unsigned long sectors);
+void blkiocg_update_timeslice_used(struct blkio_group *blkg,
+					unsigned long time);
+void blkiocg_update_dispatch_stats(struct blkio_group *blkg, uint64_t bytes,
+						bool direction, bool sync);
+void blkiocg_update_completion_stats(struct blkio_group *blkg,
+	uint64_t start_time, uint64_t io_start_time, bool direction, bool sync);
+void blkiocg_update_io_merged_stats(struct blkio_group *blkg, bool direction,
+					bool sync);
+void blkiocg_update_io_add_stats(struct blkio_group *blkg,
+		struct blkio_group *curr_blkg, bool direction, bool sync);
+void blkiocg_update_io_remove_stats(struct blkio_group *blkg,
+					bool direction, bool sync);
 #else
 struct cgroup;
 static inline struct blkio_cgroup *
 cgroup_to_blkio_cgroup(struct cgroup *cgroup) { return NULL; }
 
 static inline void blkiocg_add_blkio_group(struct blkio_cgroup *blkcg,
-			struct blkio_group *blkg, void *key, dev_t dev)
-{
-}
+			struct blkio_group *blkg, void *key, dev_t dev) {}
 
 static inline int
 blkiocg_del_blkio_group(struct blkio_group *blkg) { return 0; }
 
 static inline struct blkio_group *
 blkiocg_lookup_group(struct blkio_cgroup *blkcg, void *key) { return NULL; }
-static inline void blkiocg_update_blkio_group_stats(struct blkio_group *blkg,
-			unsigned long time, unsigned long sectors)
-{
-}
+static inline void blkiocg_update_timeslice_used(struct blkio_group *blkg,
+						unsigned long time) {}
+static inline void blkiocg_update_dispatch_stats(struct blkio_group *blkg,
+				uint64_t bytes, bool direction, bool sync) {}
+static inline void blkiocg_update_completion_stats(struct blkio_group *blkg,
+		uint64_t start_time, uint64_t io_start_time, bool direction,
+		bool sync) {}
+static inline void blkiocg_update_io_merged_stats(struct blkio_group *blkg,
+						bool direction, bool sync) {}
+static inline void blkiocg_update_io_add_stats(struct blkio_group *blkg,
+		struct blkio_group *curr_blkg, bool direction, bool sync) {}
+static inline void blkiocg_update_io_remove_stats(struct blkio_group *blkg,
+						bool direction, bool sync) {}
 #endif
 #endif /* _BLK_CGROUP_H */

block/blk-core.c

@@ -127,6 +127,7 @@ void blk_rq_init(struct request_queue *q, struct request *rq)
 	rq->tag = -1;
 	rq->ref_count = 1;
 	rq->start_time = jiffies;
+	set_start_time_ns(rq);
 }
 EXPORT_SYMBOL(blk_rq_init);
 
@@ -450,6 +451,7 @@ void blk_cleanup_queue(struct request_queue *q)
 	 */
 	blk_sync_queue(q);
 
+	del_timer_sync(&q->backing_dev_info.laptop_mode_wb_timer);
 	mutex_lock(&q->sysfs_lock);
 	queue_flag_set_unlocked(QUEUE_FLAG_DEAD, q);
 	mutex_unlock(&q->sysfs_lock);
@@ -510,6 +512,8 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id)
 		return NULL;
 	}
 
+	setup_timer(&q->backing_dev_info.laptop_mode_wb_timer,
+		    laptop_mode_timer_fn, (unsigned long) q);
 	init_timer(&q->unplug_timer);
 	setup_timer(&q->timeout, blk_rq_timed_out_timer, (unsigned long) q);
 	INIT_LIST_HEAD(&q->timeout_list);
@@ -568,6 +572,22 @@ blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id)
 {
 	struct request_queue *q = blk_alloc_queue_node(GFP_KERNEL, node_id);
 
+	return blk_init_allocated_queue_node(q, rfn, lock, node_id);
+}
+EXPORT_SYMBOL(blk_init_queue_node);
+
+struct request_queue *
+blk_init_allocated_queue(struct request_queue *q, request_fn_proc *rfn,
+			 spinlock_t *lock)
+{
+	return blk_init_allocated_queue_node(q, rfn, lock, -1);
+}
+EXPORT_SYMBOL(blk_init_allocated_queue);
+
+struct request_queue *
+blk_init_allocated_queue_node(struct request_queue *q, request_fn_proc *rfn,
+			      spinlock_t *lock, int node_id)
+{
 	if (!q)
 		return NULL;
 
@@ -601,7 +621,7 @@ blk_init_queue_node(request_fn_proc *rfn, spinlock_t *lock, int node_id)
 	blk_put_queue(q);
 	return NULL;
 }
-EXPORT_SYMBOL(blk_init_queue_node);
+EXPORT_SYMBOL(blk_init_allocated_queue_node);
 
 int blk_get_queue(struct request_queue *q)
 {
@@ -1198,6 +1218,7 @@ static int __make_request(struct request_queue *q, struct bio *bio)
 		if (!blk_rq_cpu_valid(req))
 			req->cpu = bio->bi_comp_cpu;
 		drive_stat_acct(req, 0);
+		elv_bio_merged(q, req, bio);
 		if (!attempt_back_merge(q, req))
 			elv_merged_request(q, req, el_ret);
 		goto out;
@@ -1231,6 +1252,7 @@ static int __make_request(struct request_queue *q, struct bio *bio)
 		if (!blk_rq_cpu_valid(req))
 			req->cpu = bio->bi_comp_cpu;
 		drive_stat_acct(req, 0);
+		elv_bio_merged(q, req, bio);
 		if (!attempt_front_merge(q, req))
 			elv_merged_request(q, req, el_ret);
 		goto out;
@@ -1855,8 +1877,10 @@ void blk_dequeue_request(struct request *rq)
 	 * and to it is freed is accounted as io that is in progress at
 	 * the driver side.
 	 */
-	if (blk_account_rq(rq))
+	if (blk_account_rq(rq)) {
 		q->in_flight[rq_is_sync(rq)]++;
+		set_io_start_time_ns(rq);
+	}
 }
 
 /**
@@ -2098,7 +2122,7 @@ static void blk_finish_request(struct request *req, int error)
 	BUG_ON(blk_queued_rq(req));
 
 	if (unlikely(laptop_mode) && blk_fs_request(req))
-		laptop_io_completion();
+		laptop_io_completion(&req->q->backing_dev_info);
 
 	blk_delete_timer(req);
 
@@ -2517,4 +2541,3 @@ int __init blk_dev_init(void)
 
 	return 0;
 }
-

block/blk-lib.c

+/*
+ * Functions related to generic helpers functions
+ */
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/bio.h>
+#include <linux/blkdev.h>
+#include <linux/scatterlist.h>
+
+#include "blk.h"
+
+static void blkdev_discard_end_io(struct bio *bio, int err)
+{
+	if (err) {
+		if (err == -EOPNOTSUPP)
+			set_bit(BIO_EOPNOTSUPP, &bio->bi_flags);
+		clear_bit(BIO_UPTODATE, &bio->bi_flags);
+	}
+
+	if (bio->bi_private)
+		complete(bio->bi_private);
+	__free_page(bio_page(bio));
+
+	bio_put(bio);
+}
+
+/**
+ * blkdev_issue_discard - queue a discard
+ * @bdev:	blockdev to issue discard for
+ * @sector:	start sector
+ * @nr_sects:	number of sectors to discard
+ * @gfp_mask:	memory allocation flags (for bio_alloc)
+ * @flags:	BLKDEV_IFL_* flags to control behaviour
+ *
+ * Description:
+ *    Issue a discard request for the sectors in question.
+ */
+int blkdev_issue_discard(struct block_device *bdev, sector_t sector,
+		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags)
+{
+	DECLARE_COMPLETION_ONSTACK(wait);
+	struct request_queue *q = bdev_get_queue(bdev);
+	int type = flags & BLKDEV_IFL_BARRIER ?
+		DISCARD_BARRIER : DISCARD_NOBARRIER;
+	struct bio *bio;
+	struct page *page;
+	int ret = 0;
+
+	if (!q)
+		return -ENXIO;
+
+	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);
+
+		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 & 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;
+			sector += max_discard_sectors;
+		} else {
+			bio->bi_size = nr_sects << 9;
+			nr_sects = 0;
+		}
+
+		bio_get(bio);
+		submit_bio(type, bio);
+
+		if (flags & BLKDEV_IFL_WAIT)
+			wait_for_completion(&wait);
+
+		if (bio_flagged(bio, BIO_EOPNOTSUPP))
+			ret = -EOPNOTSUPP;
+		else if (!bio_flagged(bio, BIO_UPTODATE))
+			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);
+
+struct bio_batch
+{
+	atomic_t 		done;
+	unsigned long 		flags;
+	struct completion 	*wait;
+	bio_end_io_t		*end_io;
+};
+
+static void bio_batch_end_io(struct bio *bio, int err)
+{
+	struct bio_batch *bb = bio->bi_private;
+
+	if (err) {
+		if (err == -EOPNOTSUPP)
+			set_bit(BIO_EOPNOTSUPP, &bb->flags);
+		else
+			clear_bit(BIO_UPTODATE, &bb->flags);
+	}
+	if (bb) {
+		if (bb->end_io)
+			bb->end_io(bio, err);
+		atomic_inc(&bb->done);
+		complete(bb->wait);
+	}
+	bio_put(bio);
+}
+
+/**
+ * blkdev_issue_zeroout generate number of zero filed write bios
+ * @bdev:	blockdev to issue
+ * @sector:	start sector
+ * @nr_sects:	number of sectors to write
+ * @gfp_mask:	memory allocation flags (for bio_alloc)
+ * @flags:	BLKDEV_IFL_* flags to control behaviour
+ *
+ * Description:
+ *  Generate and issue number of bios with zerofiled pages.
+ *  Send barrier at the beginning and at the end if requested. This guarantie
+ *  correct request ordering. Empty barrier allow us to avoid post queue flush.
+ */
+
+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;
+	struct bio *bio;
+	struct bio_batch bb;
+	unsigned int sz, issued = 0;
+	DECLARE_COMPLETION_ONSTACK(wait);
+
+	atomic_set(&bb.done, 0);
+	bb.flags = 1 << BIO_UPTODATE;
+	bb.wait = &wait;
+	bb.end_io = NULL;
+
+	if (flags & BLKDEV_IFL_BARRIER) {
+		/* issue async barrier before the data */
+		ret = blkdev_issue_flush(bdev, gfp_mask, NULL, 0);
+		if (ret)
+			return ret;
+	}
+submit:
+	while (nr_sects != 0) {
+		bio = bio_alloc(gfp_mask,
+				min(nr_sects, (sector_t)BIO_MAX_PAGES));
+		if (!bio)
+			break;
+
+		bio->bi_sector = sector;
+		bio->bi_bdev   = bdev;
+		bio->bi_end_io = bio_batch_end_io;
+		if (flags & BLKDEV_IFL_WAIT)
+			bio->bi_private = &bb;
+
+		while (nr_sects != 0) {
+			sz = min((sector_t) PAGE_SIZE >> 9 , nr_sects);
+			if (sz == 0)
+				/* bio has maximum size possible */
+				break;
+			ret = bio_add_page(bio, ZERO_PAGE(0), sz << 9, 0);
+			nr_sects -= ret >> 9;
+			sector += ret >> 9;
+			if (ret < (sz << 9))
+				break;
+		}
+		issued++;
+		submit_bio(WRITE, bio);
+	}
+	/*
+	 * When all data bios are in flight. Send final barrier if requeted.
+	 */
+	if (nr_sects == 0 && flags & BLKDEV_IFL_BARRIER)
+		ret = blkdev_issue_flush(bdev, gfp_mask, NULL,
+					flags & BLKDEV_IFL_WAIT);
+
+
+	if (flags & BLKDEV_IFL_WAIT)
+		/* Wait for bios in-flight */
+		while ( issued != atomic_read(&bb.done))
+			wait_for_completion(&wait);
+
+	if (!test_bit(BIO_UPTODATE, &bb.flags))
+		/* One of bios in the batch was completed with error.*/
+		ret = -EIO;
+
+	if (ret)
+		goto out;
+
+	if (test_bit(BIO_EOPNOTSUPP, &bb.flags)) {
+		ret = -EOPNOTSUPP;
+		goto out;
+	}
+	if (nr_sects != 0)
+		goto submit;
+out:
+	return ret;
+}
+EXPORT_SYMBOL(blkdev_issue_zeroout);

block/cfq-iosched.c

@@ -55,6 +55,7 @@ static const int cfq_hist_divisor = 4;
 #define RQ_CIC(rq)		\
 	((struct cfq_io_context *) (rq)->elevator_private)
 #define RQ_CFQQ(rq)		(struct cfq_queue *) ((rq)->elevator_private2)
+#define RQ_CFQG(rq)		(struct cfq_group *) ((rq)->elevator_private3)
 
 static struct kmem_cache *cfq_pool;
 static struct kmem_cache *cfq_ioc_pool;
@@ -143,8 +144,6 @@ struct cfq_queue {
 	struct cfq_queue *new_cfqq;
 	struct cfq_group *cfqg;
 	struct cfq_group *orig_cfqg;
-	/* Sectors dispatched in current dispatch round */
-	unsigned long nr_sectors;
 };
 
 /*
@@ -346,7 +345,7 @@ CFQ_CFQQ_FNS(deep);
 CFQ_CFQQ_FNS(wait_busy);
 #undef CFQ_CFQQ_FNS
 
-#ifdef CONFIG_DEBUG_CFQ_IOSCHED
+#ifdef CONFIG_CFQ_GROUP_IOSCHED
 #define cfq_log_cfqq(cfqd, cfqq, fmt, args...)	\
 	blk_add_trace_msg((cfqd)->queue, "cfq%d%c %s " fmt, (cfqq)->pid, \
 			cfq_cfqq_sync((cfqq)) ? 'S' : 'A', \
@@ -858,7 +857,7 @@ cfq_group_service_tree_del(struct cfq_data *cfqd, struct cfq_group *cfqg)
 	if (!RB_EMPTY_NODE(&cfqg->rb_node))
 		cfq_rb_erase(&cfqg->rb_node, st);
 	cfqg->saved_workload_slice = 0;
-	blkiocg_update_blkio_group_dequeue_stats(&cfqg->blkg, 1);
+	blkiocg_update_dequeue_stats(&cfqg->blkg, 1);
 }
 
 static inline unsigned int cfq_cfqq_slice_usage(struct cfq_queue *cfqq)
@@ -884,8 +883,7 @@ static inline unsigned int cfq_cfqq_slice_usage(struct cfq_queue *cfqq)
 			slice_used = cfqq->allocated_slice;
 	}
 
-	cfq_log_cfqq(cfqq->cfqd, cfqq, "sl_used=%u sect=%lu", slice_used,
-				cfqq->nr_sectors);
+	cfq_log_cfqq(cfqq->cfqd, cfqq, "sl_used=%u", slice_used);
 	return slice_used;
 }
 
@@ -919,8 +917,8 @@ static void cfq_group_served(struct cfq_data *cfqd, struct cfq_group *cfqg,
 
 	cfq_log_cfqg(cfqd, cfqg, "served: vt=%llu min_vt=%llu", cfqg->vdisktime,
 					st->min_vdisktime);
-	blkiocg_update_blkio_group_stats(&cfqg->blkg, used_sl,
-						cfqq->nr_sectors);
+	blkiocg_update_timeslice_used(&cfqg->blkg, used_sl);
+	blkiocg_set_start_empty_time(&cfqg->blkg);
 }
 
 #ifdef CONFIG_CFQ_GROUP_IOSCHED
@@ -961,7 +959,6 @@ cfq_find_alloc_cfqg(struct cfq_data *cfqd, struct cgroup *cgroup, int create)
 	if (!cfqg)
 		goto done;
 
-	cfqg->weight = blkcg->weight;
 	for_each_cfqg_st(cfqg, i, j, st)
 		*st = CFQ_RB_ROOT;
 	RB_CLEAR_NODE(&cfqg->rb_node);
@@ -978,6 +975,7 @@ cfq_find_alloc_cfqg(struct cfq_data *cfqd, struct cgroup *cgroup, int create)
 	sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor);
 	blkiocg_add_blkio_group(blkcg, &cfqg->blkg, (void *)cfqd,
 					MKDEV(major, minor));
+	cfqg->weight = blkcg_get_weight(blkcg, cfqg->blkg.dev);
 
 	/* Add group on cfqd list */
 	hlist_add_head(&cfqg->cfqd_node, &cfqd->cfqg_list);
@@ -1004,6 +1002,12 @@ static struct cfq_group *cfq_get_cfqg(struct cfq_data *cfqd, int create)
 	return cfqg;
 }
 
+static inline struct cfq_group *cfq_ref_get_cfqg(struct cfq_group *cfqg)
+{
+	atomic_inc(&cfqg->ref);
+	return cfqg;
+}
+
 static void cfq_link_cfqq_cfqg(struct cfq_queue *cfqq, struct cfq_group *cfqg)
 {
 	/* Currently, all async queues are mapped to root group */
@@ -1087,6 +1091,12 @@ static struct cfq_group *cfq_get_cfqg(struct cfq_data *cfqd, int create)
 {
 	return &cfqd->root_group;
 }
+
+static inline struct cfq_group *cfq_ref_get_cfqg(struct cfq_group *cfqg)
+{
+	return cfqg;
+}
+
 static inline void
 cfq_link_cfqq_cfqg(struct cfq_queue *cfqq, struct cfq_group *cfqg) {
 	cfqq->cfqg = cfqg;
@@ -1389,7 +1399,12 @@ static void cfq_reposition_rq_rb(struct cfq_queue *cfqq, struct request *rq)
 {
 	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_add_rq_rb(rq);
+	blkiocg_update_io_add_stats(&(RQ_CFQG(rq))->blkg,
+			&cfqq->cfqd->serving_group->blkg, rq_data_dir(rq),
+			rq_is_sync(rq));
 }
 
 static struct request *
@@ -1445,6 +1460,8 @@ static void cfq_remove_request(struct request *rq)
 	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));
 	if (rq_is_meta(rq)) {
 		WARN_ON(!cfqq->meta_pending);
 		cfqq->meta_pending--;
@@ -1476,6 +1493,13 @@ static void cfq_merged_request(struct request_queue *q, struct request *req,
 	}
 }
 
+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));
+}
+
 static void
 cfq_merged_requests(struct request_queue *q, struct request *rq,
 		    struct request *next)
@@ -1493,6 +1517,8 @@ cfq_merged_requests(struct request_queue *q, struct request *rq,
 	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));
 }
 
 static int cfq_allow_merge(struct request_queue *q, struct request *rq,
@@ -1520,18 +1546,24 @@ static int cfq_allow_merge(struct request_queue *q, struct request *rq,
 	return cfqq == RQ_CFQQ(rq);
 }
 
+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);
+}
+
 static void __cfq_set_active_queue(struct cfq_data *cfqd,
 				   struct cfq_queue *cfqq)
 {
 	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);
 		cfqq->slice_start = 0;
 		cfqq->dispatch_start = jiffies;
 		cfqq->allocated_slice = 0;
 		cfqq->slice_end = 0;
 		cfqq->slice_dispatch = 0;
-		cfqq->nr_sectors = 0;
 
 		cfq_clear_cfqq_wait_request(cfqq);
 		cfq_clear_cfqq_must_dispatch(cfqq);
@@ -1539,7 +1571,7 @@ static void __cfq_set_active_queue(struct cfq_data *cfqd,
 		cfq_clear_cfqq_fifo_expire(cfqq);
 		cfq_mark_cfqq_slice_new(cfqq);
 
-		del_timer(&cfqd->idle_slice_timer);
+		cfq_del_timer(cfqd, cfqq);
 	}
 
 	cfqd->active_queue = cfqq;
@@ -1555,7 +1587,7 @@ __cfq_slice_expired(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 	cfq_log_cfqq(cfqd, cfqq, "slice expired t=%d", timed_out);
 
 	if (cfq_cfqq_wait_request(cfqq))
-		del_timer(&cfqd->idle_slice_timer);
+		cfq_del_timer(cfqd, cfqq);
 
 	cfq_clear_cfqq_wait_request(cfqq);
 	cfq_clear_cfqq_wait_busy(cfqq);
@@ -1857,6 +1889,7 @@ static void cfq_arm_slice_timer(struct cfq_data *cfqd)
 	sl = cfqd->cfq_slice_idle;
 
 	mod_timer(&cfqd->idle_slice_timer, jiffies + sl);
+	blkiocg_update_set_idle_time_stats(&cfqq->cfqg->blkg);
 	cfq_log_cfqq(cfqd, cfqq, "arm_idle: %lu", sl);
 }
 
@@ -1876,7 +1909,8 @@ static void cfq_dispatch_insert(struct request_queue *q, struct request *rq)
 	elv_dispatch_sort(q, rq);
 
 	cfqd->rq_in_flight[cfq_cfqq_sync(cfqq)]++;
-	cfqq->nr_sectors += blk_rq_sectors(rq);
+	blkiocg_update_dispatch_stats(&cfqq->cfqg->blkg, blk_rq_bytes(rq),
+					rq_data_dir(rq), rq_is_sync(rq));
 }
 
 /*
@@ -3185,12 +3219,15 @@ cfq_rq_enqueued(struct cfq_data *cfqd, struct cfq_queue *cfqq,
 		if (cfq_cfqq_wait_request(cfqq)) {
 			if (blk_rq_bytes(rq) > PAGE_CACHE_SIZE ||
 			    cfqd->busy_queues > 1) {
-				del_timer(&cfqd->idle_slice_timer);
+				cfq_del_timer(cfqd, cfqq);
 				cfq_clear_cfqq_wait_request(cfqq);
 				__blk_run_queue(cfqd->queue);
-			} else
+			} else {
+				blkiocg_update_idle_time_stats(
+						&cfqq->cfqg->blkg);
 				cfq_mark_cfqq_must_dispatch(cfqq);
 			}
+		}
 	} else if (cfq_should_preempt(cfqd, cfqq, rq)) {
 		/*
 		 * not the active queue - expire current slice if it is
@@ -3214,7 +3251,9 @@ static void cfq_insert_request(struct request_queue *q, struct request *rq)
 	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,
+			&cfqd->serving_group->blkg, rq_data_dir(rq),
+			rq_is_sync(rq));
 	cfq_rq_enqueued(cfqd, cfqq, rq);
 }
 
@@ -3300,6 +3339,9 @@ static void cfq_completed_request(struct request_queue *q, struct request *rq)
 	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));
 
 	cfqd->rq_in_flight[cfq_cfqq_sync(cfqq)]--;
 
@@ -3440,6 +3482,10 @@ static void cfq_put_request(struct request *rq)
 		rq->elevator_private = NULL;
 		rq->elevator_private2 = NULL;
 
+		/* Put down rq reference on cfqg */
+		cfq_put_cfqg(RQ_CFQG(rq));
+		rq->elevator_private3 = NULL;
+
 		cfq_put_queue(cfqq);
 	}
 }
@@ -3528,6 +3574,7 @@ cfq_set_request(struct request_queue *q, struct request *rq, gfp_t gfp_mask)
 
 	rq->elevator_private = cic;
 	rq->elevator_private2 = cfqq;
+	rq->elevator_private3 = cfq_ref_get_cfqg(cfqq->cfqg);
 	return 0;
 
 queue_fail:
@@ -3743,7 +3790,6 @@ static void *cfq_init_queue(struct request_queue *q)
 	 * second, in order to have larger depth for async operations.
 	 */
 	cfqd->last_delayed_sync = jiffies - HZ;
-	INIT_RCU_HEAD(&cfqd->rcu);
 	return cfqd;
 }
 
@@ -3872,6 +3918,7 @@ static struct elevator_type iosched_cfq = {
 		.elevator_merged_fn =		cfq_merged_request,
 		.elevator_merge_req_fn =	cfq_merged_requests,
 		.elevator_allow_merge_fn =	cfq_allow_merge,
+		.elevator_bio_merged_fn =	cfq_bio_merged,
 		.elevator_dispatch_fn =		cfq_dispatch_requests,
 		.elevator_add_req_fn =		cfq_insert_request,
 		.elevator_activate_req_fn =	cfq_activate_request,

block/elevator.c

@@ -539,6 +539,15 @@ void elv_merge_requests(struct request_queue *q, struct request *rq,
 	q->last_merge = rq;
 }
 
+void elv_bio_merged(struct request_queue *q, struct request *rq,
+			struct bio *bio)
+{
+	struct elevator_queue *e = q->elevator;
+
+	if (e->ops->elevator_bio_merged_fn)
+		e->ops->elevator_bio_merged_fn(q, rq, bio);
+}
+
 void elv_requeue_request(struct request_queue *q, struct request *rq)
 {
 	/*
@@ -921,6 +930,7 @@ int elv_register_queue(struct request_queue *q)
 	}
 	return error;
 }
+EXPORT_SYMBOL(elv_register_queue);
 
 static void __elv_unregister_queue(struct elevator_queue *e)
 {
@@ -933,6 +943,7 @@ void elv_unregister_queue(struct request_queue *q)
 	if (q)
 		__elv_unregister_queue(q->elevator);
 }
+EXPORT_SYMBOL(elv_unregister_queue);
 
 void elv_register(struct elevator_type *e)
 {

block/genhd.c

@@ -596,6 +596,7 @@ struct gendisk *get_gendisk(dev_t devt, int *partno)
 
 	return disk;
 }
+EXPORT_SYMBOL(get_gendisk);
 
 /**
  * bdget_disk - do bdget() by gendisk and partition number
@@ -987,7 +988,6 @@ int disk_expand_part_tbl(struct gendisk *disk, int partno)
 	if (!new_ptbl)
 		return -ENOMEM;
 
-	INIT_RCU_HEAD(&new_ptbl->rcu_head);
 	new_ptbl->len = target;
 
 	for (i = 0; i < len; i++)

block/ioctl.c

@@ -126,7 +126,7 @@ static int blk_ioctl_discard(struct block_device *bdev, uint64_t start,
 	if (start + len > (bdev->bd_inode->i_size >> 9))
 		return -EINVAL;
 	return blkdev_issue_discard(bdev, start, len, GFP_KERNEL,
-				    DISCARD_FL_WAIT);
+				    BLKDEV_IFL_WAIT);
 }
 
 static int put_ushort(unsigned long arg, unsigned short val)

drivers/block/Kconfig

@@ -76,6 +76,17 @@ config BLK_DEV_XD
 
 	  It's pretty unlikely that you have one of these: say N.
 
+config GDROM
+	tristate "SEGA Dreamcast GD-ROM drive"
+	depends on SH_DREAMCAST
+	help
+	  A standard SEGA Dreamcast comes with a modified CD ROM drive called a
+	  "GD-ROM" by SEGA to signify it is capable of reading special disks
+	  with up to 1 GB of data. This drive will also read standard CD ROM
+	  disks. Select this option to access any disks in your GD ROM drive.
+	  Most users will want to say "Y" here.
+	  You can also build this as a module which will be called gdrom.
+
 config PARIDE
 	tristate "Parallel port IDE device support"
 	depends on PARPORT_PC
@@ -103,17 +114,6 @@ config PARIDE
 	  "MicroSolutions backpack protocol", "DataStor Commuter protocol"
 	  etc.).
 
-config GDROM
-	tristate "SEGA Dreamcast GD-ROM drive"
-	depends on SH_DREAMCAST
-	help
-	  A standard SEGA Dreamcast comes with a modified CD ROM drive called a
-	  "GD-ROM" by SEGA to signify it is capable of reading special disks
-	  with up to 1 GB of data. This drive will also read standard CD ROM
-	  disks. Select this option to access any disks in your GD ROM drive.
-	  Most users will want to say "Y" here.
-	  You can also build this as a module which will be called gdrom.
-
 source "drivers/block/paride/Kconfig"
 
 config BLK_CPQ_DA

drivers/block/drbd/drbd_bitmap.c

@@ -84,6 +84,9 @@ struct drbd_bitmap {
 #define BM_MD_IO_ERROR  1
 #define BM_P_VMALLOCED  2
 
+static int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
+			       unsigned long e, int val, const enum km_type km);
+
 static int bm_is_locked(struct drbd_bitmap *b)
 {
 	return test_bit(BM_LOCKED, &b->bm_flags);
@@ -441,7 +444,7 @@ static void bm_memset(struct drbd_bitmap *b, size_t offset, int c, size_t len)
  * In case this is actually a resize, we copy the old bitmap into the new one.
  * Otherwise, the bitmap is initialized to all bits set.
  */
-int drbd_bm_resize(struct drbd_conf *mdev, sector_t capacity)
+int drbd_bm_resize(struct drbd_conf *mdev, sector_t capacity, int set_new_bits)
 {
 	struct drbd_bitmap *b = mdev->bitmap;
 	unsigned long bits, words, owords, obits, *p_addr, *bm;
@@ -516,7 +519,7 @@ int drbd_bm_resize(struct drbd_conf *mdev, sector_t capacity)
 	obits  = b->bm_bits;
 
 	growing = bits > obits;
-	if (opages)
+	if (opages && growing && set_new_bits)
 		bm_set_surplus(b);
 
 	b->bm_pages = npages;
@@ -526,8 +529,12 @@ int drbd_bm_resize(struct drbd_conf *mdev, sector_t capacity)
 	b->bm_dev_capacity = capacity;
 
 	if (growing) {
+		if (set_new_bits) {
 			bm_memset(b, owords, 0xff, words-owords);
 			b->bm_set += bits - obits;
+		} else
+			bm_memset(b, owords, 0x00, words-owords);
+
 	}
 
 	if (want < have) {
@@ -773,7 +780,7 @@ static void bm_page_io_async(struct drbd_conf *mdev, struct drbd_bitmap *b, int
 	/* nothing to do, on disk == in memory */
 # define bm_cpu_to_lel(x) ((void)0)
 # else
-void bm_cpu_to_lel(struct drbd_bitmap *b)
+static void bm_cpu_to_lel(struct drbd_bitmap *b)
 {
 	/* need to cpu_to_lel all the pages ...
 	 * this may be optimized by using
@@ -1015,7 +1022,7 @@ unsigned long _drbd_bm_find_next_zero(struct drbd_conf *mdev, unsigned long bm_f
  * wants bitnr, not sector.
  * expected to be called for only a few bits (e - s about BITS_PER_LONG).
  * Must hold bitmap lock already. */
-int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
+static int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
 	unsigned long e, int val, const enum km_type km)
 {
 	struct drbd_bitmap *b = mdev->bitmap;
@@ -1053,7 +1060,7 @@ int __bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
  * for val != 0, we change 0 -> 1, return code positive
  * for val == 0, we change 1 -> 0, return code negative
  * wants bitnr, not sector */
-int bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
+static int bm_change_bits_to(struct drbd_conf *mdev, const unsigned long s,
 	const unsigned long e, int val)
 {
 	unsigned long flags;

drivers/block/drbd/drbd_main.c

@@ -684,6 +684,9 @@ static int is_valid_state(struct drbd_conf *mdev, union drbd_state ns)
 	else if (ns.conn > C_CONNECTED && ns.pdsk < D_INCONSISTENT)
 		rv = SS_NO_REMOTE_DISK;
 
+	else if (ns.conn > C_CONNECTED && ns.disk < D_UP_TO_DATE && ns.pdsk < D_UP_TO_DATE)
+		rv = SS_NO_UP_TO_DATE_DISK;
+
 	else if ((ns.conn == C_CONNECTED ||
 		  ns.conn == C_WF_BITMAP_S ||
 		  ns.conn == C_SYNC_SOURCE ||
@@ -840,7 +843,12 @@ static union drbd_state sanitize_state(struct drbd_conf *mdev, union drbd_state
 			break;
 		case C_WF_BITMAP_S:
 		case C_PAUSED_SYNC_S:
-			ns.pdsk = D_OUTDATED;
+			/* remap any consistent state to D_OUTDATED,
+			 * but disallow "upgrade" of not even consistent states.
+			 */
+			ns.pdsk =
+				(D_DISKLESS < os.pdsk && os.pdsk < D_OUTDATED)
+				? os.pdsk : D_OUTDATED;
 			break;
 		case C_SYNC_SOURCE:
 			ns.pdsk = D_INCONSISTENT;
@@ -1205,21 +1213,20 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
 	&&  (ns.pdsk < D_INCONSISTENT ||
 	     ns.pdsk == D_UNKNOWN ||
 	     ns.pdsk == D_OUTDATED)) {
-		kfree(mdev->p_uuid);
-		mdev->p_uuid = NULL;
 		if (get_ldev(mdev)) {
 			if ((ns.role == R_PRIMARY || ns.peer == R_PRIMARY) &&
-			    mdev->ldev->md.uuid[UI_BITMAP] == 0 && ns.disk >= D_UP_TO_DATE) {
-				drbd_uuid_new_current(mdev);
-				drbd_send_uuids(mdev);
-			}
+			    mdev->ldev->md.uuid[UI_BITMAP] == 0 && ns.disk >= D_UP_TO_DATE &&
+			    !atomic_read(&mdev->new_c_uuid))
+				atomic_set(&mdev->new_c_uuid, 2);
 			put_ldev(mdev);
 		}
 	}
 
 	if (ns.pdsk < D_INCONSISTENT && get_ldev(mdev)) {
-		if (ns.peer == R_PRIMARY && mdev->ldev->md.uuid[UI_BITMAP] == 0)
-			drbd_uuid_new_current(mdev);
+		/* Diskless peer becomes primary or got connected do diskless, primary peer. */
+		if (ns.peer == R_PRIMARY && mdev->ldev->md.uuid[UI_BITMAP] == 0 &&
+		    !atomic_read(&mdev->new_c_uuid))
+			atomic_set(&mdev->new_c_uuid, 2);
 
 		/* D_DISKLESS Peer becomes secondary */
 		if (os.peer == R_PRIMARY && ns.peer == R_SECONDARY)
@@ -1232,7 +1239,7 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
 	    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);  /* to start sync... */
+		drbd_send_sizes(mdev, 0, 0);  /* to start sync... */
 		drbd_send_uuids(mdev);
 		drbd_send_state(mdev);
 	}
@@ -1343,6 +1350,24 @@ static void after_state_ch(struct drbd_conf *mdev, union drbd_state os,
 	drbd_md_sync(mdev);
 }
 
+static int w_new_current_uuid(struct drbd_conf *mdev, struct drbd_work *w, int cancel)
+{
+	if (get_ldev(mdev)) {
+		if (mdev->ldev->md.uuid[UI_BITMAP] == 0) {
+			drbd_uuid_new_current(mdev);
+			if (get_net_conf(mdev)) {
+				drbd_send_uuids(mdev);
+				put_net_conf(mdev);
+			}
+			drbd_md_sync(mdev);
+		}
+		put_ldev(mdev);
+	}
+	atomic_dec(&mdev->new_c_uuid);
+	wake_up(&mdev->misc_wait);
+
+	return 1;
+}
 
 static int drbd_thread_setup(void *arg)
 {
@@ -1755,7 +1780,7 @@ int drbd_send_sync_uuid(struct drbd_conf *mdev, u64 val)
 			     (struct p_header *)&p, sizeof(p));
 }
 
-int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply)
+int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply, enum dds_flags flags)
 {
 	struct p_sizes p;
 	sector_t d_size, u_size;
@@ -1767,7 +1792,6 @@ int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply)
 		d_size = drbd_get_max_capacity(mdev->ldev);
 		u_size = mdev->ldev->dc.disk_size;
 		q_order_type = drbd_queue_order_type(mdev);
-		p.queue_order_type = cpu_to_be32(drbd_queue_order_type(mdev));
 		put_ldev(mdev);
 	} else {
 		d_size = 0;
@@ -1779,7 +1803,8 @@ int drbd_send_sizes(struct drbd_conf *mdev, int trigger_reply)
 	p.u_size = cpu_to_be64(u_size);
 	p.c_size = cpu_to_be64(trigger_reply ? 0 : drbd_get_capacity(mdev->this_bdev));
 	p.max_segment_size = cpu_to_be32(queue_max_segment_size(mdev->rq_queue));
-	p.queue_order_type = cpu_to_be32(q_order_type);
+	p.queue_order_type = cpu_to_be16(q_order_type);
+	p.dds_flags = cpu_to_be16(flags);
 
 	ok = drbd_send_cmd(mdev, USE_DATA_SOCKET, P_SIZES,
 			   (struct p_header *)&p, sizeof(p));
@@ -2180,6 +2205,43 @@ int drbd_send_ov_request(struct drbd_conf *mdev, sector_t sector, int size)
 	return ok;
 }
 
+static int drbd_send_delay_probe(struct drbd_conf *mdev, struct drbd_socket *ds)
+{
+	struct p_delay_probe dp;
+	int offset, ok = 0;
+	struct timeval now;
+
+	mutex_lock(&ds->mutex);
+	if (likely(ds->socket)) {
+		do_gettimeofday(&now);
+		offset = now.tv_usec - mdev->dps_time.tv_usec +
+			 (now.tv_sec - mdev->dps_time.tv_sec) * 1000000;
+		dp.seq_num  = cpu_to_be32(mdev->delay_seq);
+		dp.offset   = cpu_to_be32(offset);
+
+		ok = _drbd_send_cmd(mdev, ds->socket, P_DELAY_PROBE,
+				    (struct p_header *)&dp, sizeof(dp), 0);
+	}
+	mutex_unlock(&ds->mutex);
+
+	return ok;
+}
+
+static int drbd_send_delay_probes(struct drbd_conf *mdev)
+{
+	int ok;
+
+	mdev->delay_seq++;
+	do_gettimeofday(&mdev->dps_time);
+	ok = drbd_send_delay_probe(mdev, &mdev->meta);
+	ok = ok && drbd_send_delay_probe(mdev, &mdev->data);
+
+	mdev->dp_volume_last = mdev->send_cnt;
+	mod_timer(&mdev->delay_probe_timer, jiffies + mdev->sync_conf.dp_interval * HZ / 10);
+
+	return ok;
+}
+
 /* called on sndtimeo
  * returns FALSE if we should retry,
  * TRUE if we think connection is dead
@@ -2309,6 +2371,44 @@ static int _drbd_send_zc_bio(struct drbd_conf *mdev, struct bio *bio)
 	return 1;
 }
 
+static int _drbd_send_zc_ee(struct drbd_conf *mdev, struct drbd_epoch_entry *e)
+{
+	struct page *page = e->pages;
+	unsigned len = e->size;
+	page_chain_for_each(page) {
+		unsigned l = min_t(unsigned, len, PAGE_SIZE);
+		if (!_drbd_send_page(mdev, page, 0, l))
+			return 0;
+		len -= l;
+	}
+	return 1;
+}
+
+static void consider_delay_probes(struct drbd_conf *mdev)
+{
+	if (mdev->state.conn != C_SYNC_SOURCE || mdev->agreed_pro_version < 93)
+		return;
+
+	if (mdev->dp_volume_last + mdev->sync_conf.dp_volume * 2 < mdev->send_cnt)
+		drbd_send_delay_probes(mdev);
+}
+
+static int w_delay_probes(struct drbd_conf *mdev, struct drbd_work *w, int cancel)
+{
+	if (!cancel && mdev->state.conn == C_SYNC_SOURCE)
+		drbd_send_delay_probes(mdev);
+
+	return 1;
+}
+
+static void delay_probe_timer_fn(unsigned long data)
+{
+	struct drbd_conf *mdev = (struct drbd_conf *) data;
+
+	if (list_empty(&mdev->delay_probe_work.list))
+		drbd_queue_work(&mdev->data.work, &mdev->delay_probe_work);
+}
+
 /* Used to send write requests
  * R_PRIMARY -> Peer	(P_DATA)
  */
@@ -2360,7 +2460,7 @@ int drbd_send_dblock(struct drbd_conf *mdev, struct drbd_request *req)
 		drbd_send(mdev, mdev->data.socket, &p, sizeof(p), MSG_MORE));
 	if (ok && dgs) {
 		dgb = mdev->int_dig_out;
-		drbd_csum(mdev, mdev->integrity_w_tfm, req->master_bio, dgb);
+		drbd_csum_bio(mdev, mdev->integrity_w_tfm, req->master_bio, dgb);
 		ok = drbd_send(mdev, mdev->data.socket, dgb, dgs, MSG_MORE);
 	}
 	if (ok) {
@@ -2371,6 +2471,10 @@ int drbd_send_dblock(struct drbd_conf *mdev, struct drbd_request *req)
 	}
 
 	drbd_put_data_sock(mdev);
+
+	if (ok)
+		consider_delay_probes(mdev);
+
 	return ok;
 }
 
@@ -2409,13 +2513,17 @@ int drbd_send_block(struct drbd_conf *mdev, enum drbd_packets cmd,
 					sizeof(p), MSG_MORE);
 	if (ok && dgs) {
 		dgb = mdev->int_dig_out;
-		drbd_csum(mdev, mdev->integrity_w_tfm, e->private_bio, dgb);
+		drbd_csum_ee(mdev, mdev->integrity_w_tfm, e, dgb);
 		ok = drbd_send(mdev, mdev->data.socket, dgb, dgs, MSG_MORE);
 	}
 	if (ok)
-		ok = _drbd_send_zc_bio(mdev, e->private_bio);
+		ok = _drbd_send_zc_ee(mdev, e);
 
 	drbd_put_data_sock(mdev);
+
+	if (ok)
+		consider_delay_probes(mdev);
+
 	return ok;
 }
 
@@ -2600,6 +2708,7 @@ void drbd_init_set_defaults(struct drbd_conf *mdev)
 	atomic_set(&mdev->net_cnt, 0);
 	atomic_set(&mdev->packet_seq, 0);
 	atomic_set(&mdev->pp_in_use, 0);
+	atomic_set(&mdev->new_c_uuid, 0);
 
 	mutex_init(&mdev->md_io_mutex);
 	mutex_init(&mdev->data.mutex);
@@ -2628,16 +2737,26 @@ void drbd_init_set_defaults(struct drbd_conf *mdev)
 	INIT_LIST_HEAD(&mdev->unplug_work.list);
 	INIT_LIST_HEAD(&mdev->md_sync_work.list);
 	INIT_LIST_HEAD(&mdev->bm_io_work.w.list);
+	INIT_LIST_HEAD(&mdev->delay_probes);
+	INIT_LIST_HEAD(&mdev->delay_probe_work.list);
+	INIT_LIST_HEAD(&mdev->uuid_work.list);
+
 	mdev->resync_work.cb  = w_resync_inactive;
 	mdev->unplug_work.cb  = w_send_write_hint;
 	mdev->md_sync_work.cb = w_md_sync;
 	mdev->bm_io_work.w.cb = w_bitmap_io;
+	mdev->delay_probe_work.cb = w_delay_probes;
+	mdev->uuid_work.cb = w_new_current_uuid;
 	init_timer(&mdev->resync_timer);
 	init_timer(&mdev->md_sync_timer);
+	init_timer(&mdev->delay_probe_timer);
 	mdev->resync_timer.function = resync_timer_fn;
 	mdev->resync_timer.data = (unsigned long) mdev;
 	mdev->md_sync_timer.function = md_sync_timer_fn;
 	mdev->md_sync_timer.data = (unsigned long) mdev;
+	mdev->delay_probe_timer.function = delay_probe_timer_fn;
+	mdev->delay_probe_timer.data = (unsigned long) mdev;
+
 
 	init_waitqueue_head(&mdev->misc_wait);
 	init_waitqueue_head(&mdev->state_wait);
@@ -2680,7 +2799,7 @@ void drbd_mdev_cleanup(struct drbd_conf *mdev)
 	drbd_set_my_capacity(mdev, 0);
 	if (mdev->bitmap) {
 		/* maybe never allocated. */
-		drbd_bm_resize(mdev, 0);
+		drbd_bm_resize(mdev, 0, 1);
 		drbd_bm_cleanup(mdev);
 	}
 
@@ -3129,7 +3248,7 @@ int __init drbd_init(void)
 	if (err)
 		goto Enomem;
 
-	drbd_proc = proc_create("drbd", S_IFREG | S_IRUGO , NULL, &drbd_proc_fops);
+	drbd_proc = proc_create_data("drbd", S_IFREG | S_IRUGO , NULL, &drbd_proc_fops, NULL);
 	if (!drbd_proc)	{
 		printk(KERN_ERR "drbd: unable to register proc file\n");
 		goto Enomem;
@@ -3660,7 +3779,8 @@ _drbd_fault_str(unsigned int type) {
 		[DRBD_FAULT_DT_RD] = "Data read",
 		[DRBD_FAULT_DT_RA] = "Data read ahead",
 		[DRBD_FAULT_BM_ALLOC] = "BM allocation",
-		[DRBD_FAULT_AL_EE] = "EE allocation"
+		[DRBD_FAULT_AL_EE] = "EE allocation",
+		[DRBD_FAULT_RECEIVE] = "receive data corruption",
 	};
 
 	return (type < DRBD_FAULT_MAX) ? _faults[type] : "**Unknown**";

drivers/block/drbd/drbd_nl.c

@@ -510,7 +510,7 @@ void drbd_resume_io(struct drbd_conf *mdev)
  * Returns 0 on success, negative return values indicate errors.
  * You should call drbd_md_sync() after calling this function.
  */
-enum determine_dev_size drbd_determin_dev_size(struct drbd_conf *mdev, int force) __must_hold(local)
+enum determine_dev_size drbd_determin_dev_size(struct drbd_conf *mdev, enum dds_flags flags) __must_hold(local)
 {
 	sector_t prev_first_sect, prev_size; /* previous meta location */
 	sector_t la_size;
@@ -541,12 +541,12 @@ enum determine_dev_size drbd_determin_dev_size(struct drbd_conf *mdev, int force
 	/* TODO: should only be some assert here, not (re)init... */
 	drbd_md_set_sector_offsets(mdev, mdev->ldev);
 
-	size = drbd_new_dev_size(mdev, mdev->ldev, force);
+	size = drbd_new_dev_size(mdev, mdev->ldev, flags & DDSF_FORCED);
 
 	if (drbd_get_capacity(mdev->this_bdev) != size ||
 	    drbd_bm_capacity(mdev) != size) {
 		int err;
-		err = drbd_bm_resize(mdev, size);
+		err = drbd_bm_resize(mdev, size, !(flags & DDSF_NO_RESYNC));
 		if (unlikely(err)) {
 			/* currently there is only one error: ENOMEM! */
 			size = drbd_bm_capacity(mdev)>>1;
@@ -704,9 +704,6 @@ void drbd_setup_queue_param(struct drbd_conf *mdev, unsigned int max_seg_s) __mu
 	struct request_queue * const b = mdev->ldev->backing_bdev->bd_disk->queue;
 	int max_segments = mdev->ldev->dc.max_bio_bvecs;
 
-	if (b->merge_bvec_fn && !mdev->ldev->dc.use_bmbv)
-		max_seg_s = PAGE_SIZE;
-
 	max_seg_s = min(queue_max_sectors(b) * queue_logical_block_size(b), max_seg_s);
 
 	blk_queue_max_hw_sectors(q, max_seg_s >> 9);
@@ -1199,13 +1196,12 @@ static int drbd_nl_net_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
 	}
 
 	/* allocation not in the IO path, cqueue thread context */
-	new_conf = kmalloc(sizeof(struct net_conf), GFP_KERNEL);
+	new_conf = kzalloc(sizeof(struct net_conf), GFP_KERNEL);
 	if (!new_conf) {
 		retcode = ERR_NOMEM;
 		goto fail;
 	}
 
-	memset(new_conf, 0, sizeof(struct net_conf));
 	new_conf->timeout	   = DRBD_TIMEOUT_DEF;
 	new_conf->try_connect_int  = DRBD_CONNECT_INT_DEF;
 	new_conf->ping_int	   = DRBD_PING_INT_DEF;
@@ -1477,8 +1473,8 @@ static int drbd_nl_resize(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
 {
 	struct resize rs;
 	int retcode = NO_ERROR;
-	int ldsc = 0; /* local disk size changed */
 	enum determine_dev_size dd;
+	enum dds_flags ddsf;
 
 	memset(&rs, 0, sizeof(struct resize));
 	if (!resize_from_tags(mdev, nlp->tag_list, &rs)) {
@@ -1502,13 +1498,17 @@ static int drbd_nl_resize(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
 		goto fail;
 	}
 
-	if (mdev->ldev->known_size != drbd_get_capacity(mdev->ldev->backing_bdev)) {
-		mdev->ldev->known_size = drbd_get_capacity(mdev->ldev->backing_bdev);
-		ldsc = 1;
+	if (rs.no_resync && mdev->agreed_pro_version < 93) {
+		retcode = ERR_NEED_APV_93;
+		goto fail;
 	}
 
+	if (mdev->ldev->known_size != drbd_get_capacity(mdev->ldev->backing_bdev))
+		mdev->ldev->known_size = drbd_get_capacity(mdev->ldev->backing_bdev);
+
 	mdev->ldev->dc.disk_size = (sector_t)rs.resize_size;
-	dd = drbd_determin_dev_size(mdev, rs.resize_force);
+	ddsf = (rs.resize_force ? DDSF_FORCED : 0) | (rs.no_resync ? DDSF_NO_RESYNC : 0);
+	dd = drbd_determin_dev_size(mdev, ddsf);
 	drbd_md_sync(mdev);
 	put_ldev(mdev);
 	if (dd == dev_size_error) {
@@ -1516,12 +1516,12 @@ static int drbd_nl_resize(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp,
 		goto fail;
 	}
 
-	if (mdev->state.conn == C_CONNECTED && (dd != unchanged || ldsc)) {
+	if (mdev->state.conn == C_CONNECTED) {
 		if (dd == grew)
 			set_bit(RESIZE_PENDING, &mdev->flags);
 
 		drbd_send_uuids(mdev);
-		drbd_send_sizes(mdev, 1);
+		drbd_send_sizes(mdev, 1, ddsf);
 	}
 
  fail:
@@ -1551,6 +1551,10 @@ static int drbd_nl_syncer_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *n
 		sc.rate       = DRBD_RATE_DEF;
 		sc.after      = DRBD_AFTER_DEF;
 		sc.al_extents = DRBD_AL_EXTENTS_DEF;
+		sc.dp_volume  = DRBD_DP_VOLUME_DEF;
+		sc.dp_interval = DRBD_DP_INTERVAL_DEF;
+		sc.throttle_th = DRBD_RS_THROTTLE_TH_DEF;
+		sc.hold_off_th = DRBD_RS_HOLD_OFF_TH_DEF;
 	} else
 		memcpy(&sc, &mdev->sync_conf, sizeof(struct syncer_conf));
 
@@ -2207,9 +2211,9 @@ void drbd_bcast_ee(struct drbd_conf *mdev,
 {
 	struct cn_msg *cn_reply;
 	struct drbd_nl_cfg_reply *reply;
-	struct bio_vec *bvec;
 	unsigned short *tl;
-	int i;
+	struct page *page;
+	unsigned len;
 
 	if (!e)
 		return;
@@ -2247,11 +2251,15 @@ void drbd_bcast_ee(struct drbd_conf *mdev,
 	put_unaligned(T_ee_data, tl++);
 	put_unaligned(e->size, tl++);
 
-	__bio_for_each_segment(bvec, e->private_bio, i, 0) {
-		void *d = kmap(bvec->bv_page);
-		memcpy(tl, d + bvec->bv_offset, bvec->bv_len);
-		kunmap(bvec->bv_page);
-		tl=(unsigned short*)((char*)tl + bvec->bv_len);
+	len = e->size;
+	page = e->pages;
+	page_chain_for_each(page) {
+		void *d = kmap_atomic(page, KM_USER0);
+		unsigned l = min_t(unsigned, len, PAGE_SIZE);
+		memcpy(tl, d, l);
+		kunmap_atomic(d, KM_USER0);
+		tl = (unsigned short*)((char*)tl + l);
+		len -= l;
 	}
 	put_unaligned(TT_END, tl++); /* Close the tag list */
 

drivers/block/drbd/drbd_proc.c

@@ -73,14 +73,21 @@ static void drbd_syncer_progress(struct drbd_conf *mdev, struct seq_file *seq)
 	seq_printf(seq, "sync'ed:%3u.%u%% ", res / 10, res % 10);
 	/* if more than 1 GB display in MB */
 	if (mdev->rs_total > 0x100000L)
-		seq_printf(seq, "(%lu/%lu)M\n\t",
+		seq_printf(seq, "(%lu/%lu)M",
 			    (unsigned long) Bit2KB(rs_left >> 10),
 			    (unsigned long) Bit2KB(mdev->rs_total >> 10));
 	else
-		seq_printf(seq, "(%lu/%lu)K\n\t",
+		seq_printf(seq, "(%lu/%lu)K",
 			    (unsigned long) Bit2KB(rs_left),
 			    (unsigned long) Bit2KB(mdev->rs_total));
 
+	if (mdev->state.conn == C_SYNC_TARGET)
+		seq_printf(seq, " queue_delay: %d.%d ms\n\t",
+			   mdev->data_delay / 1000,
+			   (mdev->data_delay % 1000) / 100);
+	else if (mdev->state.conn == C_SYNC_SOURCE)
+		seq_printf(seq, " delay_probe: %u\n\t", mdev->delay_seq);
+
 	/* see drivers/md/md.c
 	 * We do not want to overflow, so the order of operands and
 	 * the * 100 / 100 trick are important. We do a +1 to be
@@ -128,6 +135,14 @@ static void drbd_syncer_progress(struct drbd_conf *mdev, struct seq_file *seq)
 	else
 		seq_printf(seq, " (%ld)", dbdt);
 
+	if (mdev->state.conn == C_SYNC_TARGET) {
+		if (mdev->c_sync_rate > 1000)
+			seq_printf(seq, " want: %d,%03d",
+				   mdev->c_sync_rate / 1000, mdev->c_sync_rate % 1000);
+		else
+			seq_printf(seq, " want: %d", mdev->c_sync_rate);
+	}
+
 	seq_printf(seq, " K/sec\n");
 }
 

drivers/block/drbd/drbd_req.c

@@ -722,6 +722,7 @@ static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio)
 	struct drbd_request *req;
 	int local, remote;
 	int err = -EIO;
+	int ret = 0;
 
 	/* allocate outside of all locks; */
 	req = drbd_req_new(mdev, bio);
@@ -784,7 +785,7 @@ static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio)
 			    (mdev->state.pdsk == D_INCONSISTENT &&
 			     mdev->state.conn >= C_CONNECTED));
 
-	if (!(local || remote)) {
+	if (!(local || remote) && !mdev->state.susp) {
 		dev_err(DEV, "IO ERROR: neither local nor remote disk\n");
 		goto fail_free_complete;
 	}
@@ -810,6 +811,16 @@ static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio)
 	/* GOOD, everything prepared, grab the spin_lock */
 	spin_lock_irq(&mdev->req_lock);
 
+	if (mdev->state.susp) {
+		/* If we got suspended, use the retry mechanism of
+		   generic_make_request() to restart processing of this
+		   bio. In the next call to drbd_make_request_26
+		   we sleep in inc_ap_bio() */
+		ret = 1;
+		spin_unlock_irq(&mdev->req_lock);
+		goto fail_free_complete;
+	}
+
 	if (remote) {
 		remote = (mdev->state.pdsk == D_UP_TO_DATE ||
 			    (mdev->state.pdsk == D_INCONSISTENT &&
@@ -947,12 +958,14 @@ static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio)
 		req->private_bio = NULL;
 		put_ldev(mdev);
 	}
+	if (!ret)
 		bio_endio(bio, err);
+
 	drbd_req_free(req);
 	dec_ap_bio(mdev);
 	kfree(b);
 
-	return 0;
+	return ret;
 }
 
 /* helper function for drbd_make_request
@@ -962,11 +975,6 @@ static int drbd_make_request_common(struct drbd_conf *mdev, struct bio *bio)
  */
 static int drbd_fail_request_early(struct drbd_conf *mdev, int is_write)
 {
-	/* Unconfigured */
-	if (mdev->state.conn == C_DISCONNECTING &&
-	    mdev->state.disk == D_DISKLESS)
-		return 1;
-
 	if (mdev->state.role != R_PRIMARY &&
 		(!allow_oos || is_write)) {
 		if (__ratelimit(&drbd_ratelimit_state)) {
@@ -1070,15 +1078,21 @@ int drbd_make_request_26(struct request_queue *q, struct bio *bio)
 
 		/* we need to get a "reference count" (ap_bio_cnt)
 		 * to avoid races with the disconnect/reconnect/suspend code.
-		 * In case we need to split the bio here, we need to get two references
+		 * In case we need to split the bio here, we need to get three references
 		 * atomically, otherwise we might deadlock when trying to submit the
 		 * second one! */
-		inc_ap_bio(mdev, 2);
+		inc_ap_bio(mdev, 3);
 
 		D_ASSERT(e_enr == s_enr + 1);
 
-		drbd_make_request_common(mdev, &bp->bio1);
-		drbd_make_request_common(mdev, &bp->bio2);
+		while (drbd_make_request_common(mdev, &bp->bio1))
+			inc_ap_bio(mdev, 1);
+
+		while (drbd_make_request_common(mdev, &bp->bio2))
+			inc_ap_bio(mdev, 1);
+
+		dec_ap_bio(mdev);
+
 		bio_pair_release(bp);
 	}
 	return 0;
@@ -1115,7 +1129,7 @@ int drbd_merge_bvec(struct request_queue *q, struct bvec_merge_data *bvm, struct
 	} else if (limit && get_ldev(mdev)) {
 		struct request_queue * const b =
 			mdev->ldev->backing_bdev->bd_disk->queue;
-		if (b->merge_bvec_fn && mdev->ldev->dc.use_bmbv) {
+		if (b->merge_bvec_fn) {
 			backing_limit = b->merge_bvec_fn(b, bvm, bvec);
 			limit = min(limit, backing_limit);
 		}

drivers/block/drbd/drbd_strings.c

@@ -70,7 +70,7 @@ static const char *drbd_disk_s_names[] = {
 
 static const char *drbd_state_sw_errors[] = {
 	[-SS_TWO_PRIMARIES] = "Multiple primaries not allowed by config",
-	[-SS_NO_UP_TO_DATE_DISK] = "Refusing to be Primary without at least one UpToDate disk",
+	[-SS_NO_UP_TO_DATE_DISK] = "Need access to UpToDate data",
 	[-SS_NO_LOCAL_DISK] = "Can not resync without local disk",
 	[-SS_NO_REMOTE_DISK] = "Can not resync without remote disk",
 	[-SS_CONNECTED_OUTDATES] = "Refusing to be Outdated while Connected",

drivers/block/drbd/drbd_wrappers.h

@@ -18,23 +18,9 @@ static inline void drbd_set_my_capacity(struct drbd_conf *mdev,
 
 #define drbd_bio_uptodate(bio) bio_flagged(bio, BIO_UPTODATE)
 
-static inline int drbd_bio_has_active_page(struct bio *bio)
-{
-	struct bio_vec *bvec;
-	int i;
-
-	__bio_for_each_segment(bvec, bio, i, 0) {
-		if (page_count(bvec->bv_page) > 1)
-			return 1;
-	}
-
-	return 0;
-}
-
 /* bi_end_io handlers */
 extern void drbd_md_io_complete(struct bio *bio, int error);
-extern void drbd_endio_read_sec(struct bio *bio, int error);
-extern void drbd_endio_write_sec(struct bio *bio, int error);
+extern void drbd_endio_sec(struct bio *bio, int error);
 extern void drbd_endio_pri(struct bio *bio, int error);
 
 /*

drivers/ide/ide-disk.c

@@ -407,32 +407,24 @@ static int ide_disk_get_capacity(ide_drive_t *drive)
 	return 0;
 }
 
-static u64 ide_disk_set_capacity(ide_drive_t *drive, u64 capacity)
+static void ide_disk_unlock_native_capacity(ide_drive_t *drive)
 {
-	u64 set = min(capacity, drive->probed_capacity);
 	u16 *id = drive->id;
 	int lba48 = ata_id_lba48_enabled(id);
 
 	if ((drive->dev_flags & IDE_DFLAG_LBA) == 0 ||
 	    ata_id_hpa_enabled(id) == 0)
-		goto out;
+		return;
 
 	/*
 	 * according to the spec the SET MAX ADDRESS command shall be
 	 * immediately preceded by a READ NATIVE MAX ADDRESS command
 	 */
-	capacity = ide_disk_hpa_get_native_capacity(drive, lba48);
-	if (capacity == 0)
-		goto out;
-
-	set = ide_disk_hpa_set_capacity(drive, set, lba48);
-	if (set) {
-		/* needed for ->resume to disable HPA */
-		drive->dev_flags |= IDE_DFLAG_NOHPA;
-		return set;
-	}
-out:
-	return drive->capacity64;
+	if (!ide_disk_hpa_get_native_capacity(drive, lba48))
+		return;
+
+	if (ide_disk_hpa_set_capacity(drive, drive->probed_capacity, lba48))
+		drive->dev_flags |= IDE_DFLAG_NOHPA; /* disable HPA on resume */
 }
 
 static void idedisk_prepare_flush(struct request_queue *q, struct request *rq)
@@ -784,7 +776,7 @@ static int ide_disk_set_doorlock(ide_drive_t *drive, struct gendisk *disk,
 
 const struct ide_disk_ops ide_ata_disk_ops = {
 	.check			= ide_disk_check,
-	.set_capacity	= ide_disk_set_capacity,
+	.unlock_native_capacity	= ide_disk_unlock_native_capacity,
 	.get_capacity		= ide_disk_get_capacity,
 	.setup			= ide_disk_setup,
 	.flush			= ide_disk_flush,

drivers/ide/ide-gd.c

@@ -288,17 +288,14 @@ static int ide_gd_media_changed(struct gendisk *disk)
 	return ret;
 }
 
-static unsigned long long ide_gd_set_capacity(struct gendisk *disk,
-					      unsigned long long capacity)
+static void ide_gd_unlock_native_capacity(struct gendisk *disk)
 {
 	struct ide_disk_obj *idkp = ide_drv_g(disk, ide_disk_obj);
 	ide_drive_t *drive = idkp->drive;
 	const struct ide_disk_ops *disk_ops = drive->disk_ops;
 
-	if (disk_ops->set_capacity)
-		return disk_ops->set_capacity(drive, capacity);
-
-	return drive->capacity64;
+	if (disk_ops->unlock_native_capacity)
+		disk_ops->unlock_native_capacity(drive);
 }
 
 static int ide_gd_revalidate_disk(struct gendisk *disk)
@@ -329,7 +326,7 @@ static const struct block_device_operations ide_gd_ops = {
 	.locked_ioctl		= ide_gd_ioctl,
 	.getgeo			= ide_gd_getgeo,
 	.media_changed		= ide_gd_media_changed,
-	.set_capacity		= ide_gd_set_capacity,
+	.unlock_native_capacity	= ide_gd_unlock_native_capacity,
 	.revalidate_disk	= ide_gd_revalidate_disk
 };
 

fs/block_dev.c

@@ -417,7 +417,7 @@ int blkdev_fsync(struct file *filp, struct dentry *dentry, int datasync)
 	 */
 	mutex_unlock(&bd_inode->i_mutex);
 
-	error = blkdev_issue_flush(bdev, NULL);
+	error = blkdev_issue_flush(bdev, GFP_KERNEL, NULL, BLKDEV_IFL_WAIT);
 	if (error == -EOPNOTSUPP)
 		error = 0;
 
@@ -668,41 +668,209 @@ void bd_forget(struct inode *inode)
 		iput(bdev->bd_inode);
 }
 
-int bd_claim(struct block_device *bdev, void *holder)
+/**
+ * bd_may_claim - test whether a block device can be claimed
+ * @bdev: block device of interest
+ * @whole: whole block device containing @bdev, may equal @bdev
+ * @holder: holder trying to claim @bdev
+ *
+ * Test whther @bdev can be claimed by @holder.
+ *
+ * CONTEXT:
+ * spin_lock(&bdev_lock).
+ *
+ * RETURNS:
+ * %true if @bdev can be claimed, %false otherwise.
+ */
+static bool bd_may_claim(struct block_device *bdev, struct block_device *whole,
+			 void *holder)
 {
-	int res;
-	spin_lock(&bdev_lock);
-
-	/* first decide result */
 	if (bdev->bd_holder == holder)
-		res = 0;	 /* already a holder */
+		return true;	 /* already a holder */
 	else if (bdev->bd_holder != NULL)
-		res = -EBUSY; 	 /* held by someone else */
+		return false; 	 /* held by someone else */
 	else if (bdev->bd_contains == bdev)
-		res = 0;  	 /* is a whole device which isn't held */
+		return true;  	 /* is a whole device which isn't held */
 
-	else if (bdev->bd_contains->bd_holder == bd_claim)
-		res = 0; 	 /* is a partition of a device that is being partitioned */
-	else if (bdev->bd_contains->bd_holder != NULL)
-		res = -EBUSY;	 /* is a partition of a held device */
+	else if (whole->bd_holder == bd_claim)
+		return true; 	 /* is a partition of a device that is being partitioned */
+	else if (whole->bd_holder != NULL)
+		return false;	 /* is a partition of a held device */
 	else
-		res = 0;	 /* is a partition of an un-held device */
+		return true;	 /* is a partition of an un-held device */
+}
+
+/**
+ * bd_prepare_to_claim - prepare to claim a block device
+ * @bdev: block device of interest
+ * @whole: the whole device containing @bdev, may equal @bdev
+ * @holder: holder trying to claim @bdev
+ *
+ * Prepare to claim @bdev.  This function fails if @bdev is already
+ * claimed by another holder and waits if another claiming is in
+ * progress.  This function doesn't actually claim.  On successful
+ * return, the caller has ownership of bd_claiming and bd_holder[s].
+ *
+ * CONTEXT:
+ * spin_lock(&bdev_lock).  Might release bdev_lock, sleep and regrab
+ * it multiple times.
+ *
+ * RETURNS:
+ * 0 if @bdev can be claimed, -EBUSY otherwise.
+ */
+static int bd_prepare_to_claim(struct block_device *bdev,
+			       struct block_device *whole, void *holder)
+{
+retry:
+	/* if someone else claimed, fail */
+	if (!bd_may_claim(bdev, whole, holder))
+		return -EBUSY;
+
+	/* if someone else is claiming, wait for it to finish */
+	if (whole->bd_claiming && whole->bd_claiming != holder) {
+		wait_queue_head_t *wq = bit_waitqueue(&whole->bd_claiming, 0);
+		DEFINE_WAIT(wait);
+
+		prepare_to_wait(wq, &wait, TASK_UNINTERRUPTIBLE);
+		spin_unlock(&bdev_lock);
+		schedule();
+		finish_wait(wq, &wait);
+		spin_lock(&bdev_lock);
+		goto retry;
+	}
+
+	/* yay, all mine */
+	return 0;
+}
 
-	/* now impose change */
-	if (res==0) {
+/**
+ * bd_start_claiming - start claiming a block device
+ * @bdev: block device of interest
+ * @holder: holder trying to claim @bdev
+ *
+ * @bdev is about to be opened exclusively.  Check @bdev can be opened
+ * exclusively and mark that an exclusive open is in progress.  Each
+ * successful call to this function must be matched with a call to
+ * either bd_claim() or bd_abort_claiming().  If this function
+ * succeeds, the matching bd_claim() is guaranteed to succeed.
+ *
+ * CONTEXT:
+ * Might sleep.
+ *
+ * RETURNS:
+ * Pointer to the block device containing @bdev on success, ERR_PTR()
+ * value on failure.
+ */
+static struct block_device *bd_start_claiming(struct block_device *bdev,
+					      void *holder)
+{
+	struct gendisk *disk;
+	struct block_device *whole;
+	int partno, err;
+
+	might_sleep();
+
+	/*
+	 * @bdev might not have been initialized properly yet, look up
+	 * and grab the outer block device the hard way.
+	 */
+	disk = get_gendisk(bdev->bd_dev, &partno);
+	if (!disk)
+		return ERR_PTR(-ENXIO);
+
+	whole = bdget_disk(disk, 0);
+	put_disk(disk);
+	if (!whole)
+		return ERR_PTR(-ENOMEM);
+
+	/* prepare to claim, if successful, mark claiming in progress */
+	spin_lock(&bdev_lock);
+
+	err = bd_prepare_to_claim(bdev, whole, holder);
+	if (err == 0) {
+		whole->bd_claiming = holder;
+		spin_unlock(&bdev_lock);
+		return whole;
+	} else {
+		spin_unlock(&bdev_lock);
+		bdput(whole);
+		return ERR_PTR(err);
+	}
+}
+
+/* releases bdev_lock */
+static void __bd_abort_claiming(struct block_device *whole, void *holder)
+{
+	BUG_ON(whole->bd_claiming != holder);
+	whole->bd_claiming = NULL;
+	wake_up_bit(&whole->bd_claiming, 0);
+
+	spin_unlock(&bdev_lock);
+	bdput(whole);
+}
+
+/**
+ * bd_abort_claiming - abort claiming a block device
+ * @whole: whole block device returned by bd_start_claiming()
+ * @holder: holder trying to claim @bdev
+ *
+ * Abort a claiming block started by bd_start_claiming().  Note that
+ * @whole is not the block device to be claimed but the whole device
+ * returned by bd_start_claiming().
+ *
+ * CONTEXT:
+ * Grabs and releases bdev_lock.
+ */
+static void bd_abort_claiming(struct block_device *whole, void *holder)
+{
+	spin_lock(&bdev_lock);
+	__bd_abort_claiming(whole, holder);		/* releases bdev_lock */
+}
+
+/**
+ * bd_claim - claim a block device
+ * @bdev: block device to claim
+ * @holder: holder trying to claim @bdev
+ *
+ * Try to claim @bdev which must have been opened successfully.  This
+ * function may be called with or without preceding
+ * blk_start_claiming().  In the former case, this function is always
+ * successful and terminates the claiming block.
+ *
+ * CONTEXT:
+ * Might sleep.
+ *
+ * RETURNS:
+ * 0 if successful, -EBUSY if @bdev is already claimed.
+ */
+int bd_claim(struct block_device *bdev, void *holder)
+{
+	struct block_device *whole = bdev->bd_contains;
+	int res;
+
+	might_sleep();
+
+	spin_lock(&bdev_lock);
+
+	res = bd_prepare_to_claim(bdev, whole, holder);
+	if (res == 0) {
 		/* note that for a whole device bd_holders
 		 * will be incremented twice, and bd_holder will
 		 * be set to bd_claim before being set to holder
 		 */
-		bdev->bd_contains->bd_holders ++;
-		bdev->bd_contains->bd_holder = bd_claim;
+		whole->bd_holders++;
+		whole->bd_holder = bd_claim;
 		bdev->bd_holders++;
 		bdev->bd_holder = holder;
 	}
+
+	if (whole->bd_claiming)
+		__bd_abort_claiming(whole, holder);	/* releases bdev_lock */
+	else
 		spin_unlock(&bdev_lock);
+
 	return res;
 }
-
 EXPORT_SYMBOL(bd_claim);
 
 void bd_release(struct block_device *bdev)
@@ -1316,6 +1484,7 @@ EXPORT_SYMBOL(blkdev_get);
 
 static int blkdev_open(struct inode * inode, struct file * filp)
 {
+	struct block_device *whole = NULL;
 	struct block_device *bdev;
 	int res;
 
@@ -1338,22 +1507,25 @@ static int blkdev_open(struct inode * inode, struct file * filp)
 	if (bdev == NULL)
 		return -ENOMEM;
 
+	if (filp->f_mode & FMODE_EXCL) {
+		whole = bd_start_claiming(bdev, filp);
+		if (IS_ERR(whole)) {
+			bdput(bdev);
+			return PTR_ERR(whole);
+		}
+	}
+
 	filp->f_mapping = bdev->bd_inode->i_mapping;
 
 	res = blkdev_get(bdev, filp->f_mode);
-	if (res)
-		return res;
 
-	if (filp->f_mode & FMODE_EXCL) {
-		res = bd_claim(bdev, filp);
-		if (res)
-			goto out_blkdev_put;
+	if (whole) {
+		if (res == 0)
+			BUG_ON(bd_claim(bdev, filp) != 0);
+		else
+			bd_abort_claiming(whole, filp);
 	}
 
-	return 0;
-
- out_blkdev_put:
-	blkdev_put(bdev, filp->f_mode);
 	return res;
 }
 
@@ -1564,27 +1736,34 @@ EXPORT_SYMBOL(lookup_bdev);
  */
 struct block_device *open_bdev_exclusive(const char *path, fmode_t mode, void *holder)
 {
-	struct block_device *bdev;
-	int error = 0;
+	struct block_device *bdev, *whole;
+	int error;
 
 	bdev = lookup_bdev(path);
 	if (IS_ERR(bdev))
 		return bdev;
 
+	whole = bd_start_claiming(bdev, holder);
+	if (IS_ERR(whole)) {
+		bdput(bdev);
+		return whole;
+	}
+
 	error = blkdev_get(bdev, mode);
 	if (error)
-		return ERR_PTR(error);
+		goto out_abort_claiming;
+
 	error = -EACCES;
 	if ((mode & FMODE_WRITE) && bdev_read_only(bdev))
-		goto blkdev_put;
-	error = bd_claim(bdev, holder);
-	if (error)
-		goto blkdev_put;
+		goto out_blkdev_put;
 
+	BUG_ON(bd_claim(bdev, holder) != 0);
 	return bdev;
 
-blkdev_put:
+out_blkdev_put:
 	blkdev_put(bdev, mode);
+out_abort_claiming:
+	bd_abort_claiming(whole, holder);
 	return ERR_PTR(error);
 }
 

fs/btrfs/extent-tree.c

@@ -1589,7 +1589,7 @@ static void btrfs_issue_discard(struct block_device *bdev,
 				u64 start, u64 len)
 {
 	blkdev_issue_discard(bdev, start >> 9, len >> 9, GFP_KERNEL,
-			     DISCARD_FL_BARRIER);
+			BLKDEV_IFL_WAIT | BLKDEV_IFL_BARRIER);
 }
 
 static int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr,

fs/buffer.c

@@ -275,6 +275,7 @@ void invalidate_bdev(struct block_device *bdev)
 		return;
 
 	invalidate_bh_lrus();
+	lru_add_drain_all();	/* make sure all lru add caches are flushed */
 	invalidate_mapping_pages(mapping, 0, -1);
 }
 EXPORT_SYMBOL(invalidate_bdev);

fs/ext3/fsync.c

@@ -90,6 +90,7 @@ int ext3_sync_file(struct file * file, struct dentry *dentry, int datasync)
 	 * storage
 	 */
 	if (needs_barrier)
-		blkdev_issue_flush(inode->i_sb->s_bdev, NULL);
+		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL,
+				BLKDEV_IFL_WAIT);
 	return ret;
 }

fs/ext4/fsync.c

@@ -100,9 +100,11 @@ int ext4_sync_file(struct file *file, struct dentry *dentry, int datasync)
 		if (ext4_should_writeback_data(inode) &&
 		    (journal->j_fs_dev != journal->j_dev) &&
 		    (journal->j_flags & JBD2_BARRIER))
-			blkdev_issue_flush(inode->i_sb->s_bdev, NULL);
+			blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL,
+					NULL, BLKDEV_IFL_WAIT);
 		jbd2_log_wait_commit(journal, commit_tid);
 	} else if (journal->j_flags & JBD2_BARRIER)
-		blkdev_issue_flush(inode->i_sb->s_bdev, NULL);
+		blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL,
+			BLKDEV_IFL_WAIT);
 	return ret;
 }

fs/fcntl.c

@@ -14,6 +14,7 @@
 #include <linux/dnotify.h>
 #include <linux/slab.h>
 #include <linux/module.h>
+#include <linux/pipe_fs_i.h>
 #include <linux/security.h>
 #include <linux/ptrace.h>
 #include <linux/signal.h>
@@ -412,6 +413,10 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
 	case F_NOTIFY:
 		err = fcntl_dirnotify(fd, filp, arg);
 		break;
+	case F_SETPIPE_SZ:
+	case F_GETPIPE_SZ:
+		err = pipe_fcntl(filp, cmd, arg);
+		break;
 	default:
 		break;
 	}

fs/gfs2/rgrp.c

@@ -854,7 +854,8 @@ static void gfs2_rgrp_send_discards(struct gfs2_sbd *sdp, u64 offset,
 				if ((start + nr_sects) != blk) {
 					rv = blkdev_issue_discard(bdev, start,
 							    nr_sects, GFP_NOFS,
-							    DISCARD_FL_BARRIER);
+							    BLKDEV_IFL_WAIT |
+							    BLKDEV_IFL_BARRIER);
 					if (rv)
 						goto fail;
 					nr_sects = 0;
@@ -869,7 +870,7 @@ static void gfs2_rgrp_send_discards(struct gfs2_sbd *sdp, u64 offset,
 	}
 	if (nr_sects) {
 		rv = blkdev_issue_discard(bdev, start, nr_sects, GFP_NOFS,
-					 DISCARD_FL_BARRIER);
+					 BLKDEV_IFL_WAIT | BLKDEV_IFL_BARRIER);
 		if (rv)
 			goto fail;
 	}

fs/jbd2/checkpoint.c

@@ -530,7 +530,8 @@ int jbd2_cleanup_journal_tail(journal_t *journal)
 	 */
 	if ((journal->j_fs_dev != journal->j_dev) &&
 	    (journal->j_flags & JBD2_BARRIER))
-		blkdev_issue_flush(journal->j_fs_dev, NULL);
+		blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL,
+			BLKDEV_IFL_WAIT);
 	if (!(journal->j_flags & JBD2_ABORT))
 		jbd2_journal_update_superblock(journal, 1);
 	return 0;

fs/jbd2/commit.c

@@ -717,7 +717,8 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 	if (commit_transaction->t_flushed_data_blocks &&
 	    (journal->j_fs_dev != journal->j_dev) &&
 	    (journal->j_flags & JBD2_BARRIER))
-		blkdev_issue_flush(journal->j_fs_dev, NULL);
+		blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL,
+			BLKDEV_IFL_WAIT);
 
 	/* Done it all: now write the commit record asynchronously. */
 	if (JBD2_HAS_INCOMPAT_FEATURE(journal,
@@ -727,7 +728,8 @@ void jbd2_journal_commit_transaction(journal_t *journal)
 		if (err)
 			__jbd2_journal_abort_hard(journal);
 		if (journal->j_flags & JBD2_BARRIER)
-			blkdev_issue_flush(journal->j_dev, NULL);
+			blkdev_issue_flush(journal->j_dev, GFP_KERNEL, NULL,
+				BLKDEV_IFL_WAIT);
 	}
 
 	err = journal_finish_inode_data_buffers(journal, commit_transaction);

fs/nilfs2/the_nilfs.c

@@ -674,7 +674,7 @@ int nilfs_discard_segments(struct the_nilfs *nilfs, __u64 *segnump,
 						   start * sects_per_block,
 						   nblocks * sects_per_block,
 						   GFP_NOFS,
-						   DISCARD_FL_BARRIER);
+						   BLKDEV_IFL_BARRIER);
 			if (ret < 0)
 				return ret;
 			nblocks = 0;
@@ -684,7 +684,7 @@ int nilfs_discard_segments(struct the_nilfs *nilfs, __u64 *segnump,
 		ret = blkdev_issue_discard(nilfs->ns_bdev,
 					   start * sects_per_block,
 					   nblocks * sects_per_block,
-					   GFP_NOFS, DISCARD_FL_BARRIER);
+					   GFP_NOFS, BLKDEV_IFL_BARRIER);
 	return ret;
 }
 

fs/partitions/acorn.h

@@ -7,8 +7,8 @@
  *  format, and everyone stick to it?
  */
 
-int adfspart_check_CUMANA(struct parsed_partitions *state, struct block_device *bdev);
-int adfspart_check_ADFS(struct parsed_partitions *state, struct block_device *bdev);
-int adfspart_check_ICS(struct parsed_partitions *state, struct block_device *bdev);
-int adfspart_check_POWERTEC(struct parsed_partitions *state, struct block_device *bdev);
-int adfspart_check_EESOX(struct parsed_partitions *state, struct block_device *bdev);
+int adfspart_check_CUMANA(struct parsed_partitions *state);
+int adfspart_check_ADFS(struct parsed_partitions *state);
+int adfspart_check_ICS(struct parsed_partitions *state);
+int adfspart_check_POWERTEC(struct parsed_partitions *state);
+int adfspart_check_EESOX(struct parsed_partitions *state);

fs/partitions/amiga.h

@@ -2,5 +2,5 @@
  *  fs/partitions/amiga.h
  */
 
-int amiga_partition(struct parsed_partitions *state, struct block_device *bdev);
+int amiga_partition(struct parsed_partitions *state);
 

fs/partitions/atari.h

@@ -31,4 +31,4 @@ struct rootsector
   u16 checksum;			/* checksum for bootable disks */
 } __attribute__((__packed__));
 
-int atari_partition(struct parsed_partitions *state, struct block_device *bdev);
+int atari_partition(struct parsed_partitions *state);

fs/partitions/ibm.h

-int ibm_partition(struct parsed_partitions *, struct block_device *);
+int ibm_partition(struct parsed_partitions *);