From 9744197c3d7b329590c2be33ad7b17409bd798fe Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Fri, 19 Sep 2008 11:49:54 +1000
Subject: md: Don't wait UNINTERRUPTIBLE for other resync to finish

When two md arrays share some block device (e.g each uses different
partitions on the one device), a resync of one array will wait for
the resync on the other to finish.

This can be a long time and as it currently waits TASK_UNINTERRUPTIBLE,
the softlockup code notices and complains.

So use TASK_INTERRUPTIBLE instead and make sure to flush signals
before calling schedule.

Signed-off-by: NeilBrown <neilb@suse.de>
---
 drivers/md/md.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

(limited to 'drivers/md/md.c')

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 4790c83d78d0..deeac4b44173 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -5761,7 +5761,11 @@ void md_do_sync(mddev_t *mddev)
 					 * time 'round when curr_resync == 2
 					 */
 					continue;
-				prepare_to_wait(&resync_wait, &wq, TASK_UNINTERRUPTIBLE);
+				/* We need to wait 'interruptible' so as not to
+				 * contribute to the load average, and not to
+				 * be caught by 'softlockup'
+				 */
+				prepare_to_wait(&resync_wait, &wq, TASK_INTERRUPTIBLE);
 				if (!kthread_should_stop() &&
 				    mddev2->curr_resync >= mddev->curr_resync) {
 					printk(KERN_INFO "md: delaying %s of %s"
@@ -5769,6 +5773,8 @@ void md_do_sync(mddev_t *mddev)
 					       " share one or more physical units)\n",
 					       desc, mdname(mddev), mdname(mddev2));
 					mddev_put(mddev2);
+					if (signal_pending(current))
+						flush_signals(current);
 					schedule();
 					finish_wait(&resync_wait, &wq);
 					goto try_again;
-- 
cgit v1.2.3


From ed9e1982347b36573cd622ee5f4e2a7ccd79b3fd Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Mon, 25 Aug 2008 19:56:05 +0900
Subject: block: implement and use {disk|part}_to_dev()

Implement {disk|part}_to_dev() and use them to access generic device
instead of directly dereferencing {disk|part}->dev.  To make sure no
user is left behind, rename generic devices fields to __dev.

This is in preparation of unifying partition 0 handling with other
partitions.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/blk-integrity.c      |  5 +--
 block/blk-sysfs.c          |  4 +--
 block/genhd.c              | 27 ++++++++--------
 drivers/block/aoe/aoeblk.c |  4 +--
 drivers/block/nbd.c        |  4 +--
 drivers/ide/ide-probe.c    |  2 +-
 drivers/md/dm.c            |  4 +--
 drivers/md/md.c            | 10 +++---
 fs/block_dev.c             |  4 +--
 fs/partitions/check.c      | 79 ++++++++++++++++++++++++----------------------
 include/linux/genhd.h      | 20 ++++++------
 11 files changed, 86 insertions(+), 77 deletions(-)

(limited to 'drivers/md/md.c')

diff --git a/block/blk-integrity.c b/block/blk-integrity.c
index d87606eaca1d..69023da63151 100644
--- a/block/blk-integrity.c
+++ b/block/blk-integrity.c
@@ -331,7 +331,8 @@ int blk_integrity_register(struct gendisk *disk, struct blk_integrity *template)
 			return -1;
 
 		if (kobject_init_and_add(&bi->kobj, &integrity_ktype,
-					 &disk->dev.kobj, "%s", "integrity")) {
+					 &disk_to_dev(disk)->kobj,
+					 "%s", "integrity")) {
 			kmem_cache_free(integrity_cachep, bi);
 			return -1;
 		}
@@ -375,7 +376,7 @@ void blk_integrity_unregister(struct gendisk *disk)
 
 	kobject_uevent(&bi->kobj, KOBJ_REMOVE);
 	kobject_del(&bi->kobj);
-	kobject_put(&disk->dev.kobj);
+	kobject_put(&disk_to_dev(disk)->kobj);
 	kmem_cache_free(integrity_cachep, bi);
 }
 EXPORT_SYMBOL(blk_integrity_unregister);
diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index 304ec73ab821..b9a6ed166649 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -310,7 +310,7 @@ int blk_register_queue(struct gendisk *disk)
 	if (!q->request_fn)
 		return 0;
 
-	ret = kobject_add(&q->kobj, kobject_get(&disk->dev.kobj),
+	ret = kobject_add(&q->kobj, kobject_get(&disk_to_dev(disk)->kobj),
 			  "%s", "queue");
 	if (ret < 0)
 		return ret;
@@ -339,6 +339,6 @@ void blk_unregister_queue(struct gendisk *disk)
 
 		kobject_uevent(&q->kobj, KOBJ_REMOVE);
 		kobject_del(&q->kobj);
-		kobject_put(&disk->dev.kobj);
+		kobject_put(&disk_to_dev(disk)->kobj);
 	}
 }
diff --git a/block/genhd.c b/block/genhd.c
index 67e5a59ced2a..0a2f16bd54b7 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -59,7 +59,7 @@ struct hd_struct *disk_get_part(struct gendisk *disk, int partno)
 	rcu_read_lock();
 	part = rcu_dereference(disk->__part[partno - 1]);
 	if (part)
-		get_device(&part->dev);
+		get_device(part_to_dev(part));
 	rcu_read_unlock();
 
 	return part;
@@ -130,7 +130,7 @@ struct hd_struct *disk_part_iter_next(struct disk_part_iter *piter)
 		if (!(piter->flags & DISK_PITER_INCL_EMPTY) && !part->nr_sects)
 			continue;
 
-		get_device(&part->dev);
+		get_device(part_to_dev(part));
 		piter->part = part;
 		piter->idx += inc;
 		break;
@@ -435,7 +435,7 @@ static struct kobject *exact_match(dev_t devt, int *partno, void *data)
 {
 	struct gendisk *p = data;
 
-	return &p->dev.kobj;
+	return &disk_to_dev(p)->kobj;
 }
 
 static int exact_lock(dev_t devt, void *data)
@@ -460,7 +460,7 @@ void add_disk(struct gendisk *disk)
 	int retval;
 
 	disk->flags |= GENHD_FL_UP;
-	disk->dev.devt = MKDEV(disk->major, disk->first_minor);
+	disk_to_dev(disk)->devt = MKDEV(disk->major, disk->first_minor);
 	blk_register_region(disk_devt(disk), disk->minors, NULL,
 			    exact_match, exact_lock, disk);
 	register_disk(disk);
@@ -468,7 +468,8 @@ void add_disk(struct gendisk *disk)
 
 	bdi = &disk->queue->backing_dev_info;
 	bdi_register_dev(bdi, disk_devt(disk));
-	retval = sysfs_create_link(&disk->dev.kobj, &bdi->dev->kobj, "bdi");
+	retval = sysfs_create_link(&disk_to_dev(disk)->kobj, &bdi->dev->kobj,
+				   "bdi");
 	WARN_ON(retval);
 }
 
@@ -477,7 +478,7 @@ EXPORT_SYMBOL(del_gendisk);	/* in partitions/check.c */
 
 void unlink_gendisk(struct gendisk *disk)
 {
-	sysfs_remove_link(&disk->dev.kobj, "bdi");
+	sysfs_remove_link(&disk_to_dev(disk)->kobj, "bdi");
 	bdi_unregister(&disk->queue->backing_dev_info);
 	blk_unregister_queue(disk);
 	blk_unregister_region(disk_devt(disk), disk->minors);
@@ -903,7 +904,7 @@ static int diskstats_show(struct seq_file *seqf, void *v)
 	int cpu;
 
 	/*
-	if (&gp->dev.kobj.entry == block_class.devices.next)
+	if (&disk_to_dev(gp)->kobj.entry == block_class.devices.next)
 		seq_puts(seqf,	"major minor name"
 				"     rio rmerge rsect ruse wio wmerge "
 				"wsect wuse running use aveq"
@@ -972,7 +973,7 @@ static void media_change_notify_thread(struct work_struct *work)
 	 * set enviroment vars to indicate which event this is for
 	 * so that user space will know to go check the media status.
 	 */
-	kobject_uevent_env(&gd->dev.kobj, KOBJ_CHANGE, envp);
+	kobject_uevent_env(&disk_to_dev(gd)->kobj, KOBJ_CHANGE, envp);
 	put_device(gd->driverfs_dev);
 }
 
@@ -1062,9 +1063,9 @@ struct gendisk *alloc_disk_ext_node(int minors, int ext_minors, int node_id)
 		disk->minors = minors;
 		disk->ext_minors = ext_minors;
 		rand_initialize_disk(disk);
-		disk->dev.class = &block_class;
-		disk->dev.type = &disk_type;
-		device_initialize(&disk->dev);
+		disk_to_dev(disk)->class = &block_class;
+		disk_to_dev(disk)->type = &disk_type;
+		device_initialize(disk_to_dev(disk));
 		INIT_WORK(&disk->async_notify,
 			media_change_notify_thread);
 	}
@@ -1086,7 +1087,7 @@ struct kobject *get_disk(struct gendisk *disk)
 	owner = disk->fops->owner;
 	if (owner && !try_module_get(owner))
 		return NULL;
-	kobj = kobject_get(&disk->dev.kobj);
+	kobj = kobject_get(&disk_to_dev(disk)->kobj);
 	if (kobj == NULL) {
 		module_put(owner);
 		return NULL;
@@ -1100,7 +1101,7 @@ EXPORT_SYMBOL(get_disk);
 void put_disk(struct gendisk *disk)
 {
 	if (disk)
-		kobject_put(&disk->dev.kobj);
+		kobject_put(&disk_to_dev(disk)->kobj);
 }
 
 EXPORT_SYMBOL(put_disk);
diff --git a/drivers/block/aoe/aoeblk.c b/drivers/block/aoe/aoeblk.c
index 0c39782b2660..3edb6cb7d68f 100644
--- a/drivers/block/aoe/aoeblk.c
+++ b/drivers/block/aoe/aoeblk.c
@@ -109,12 +109,12 @@ static const struct attribute_group attr_group = {
 static int
 aoedisk_add_sysfs(struct aoedev *d)
 {
-	return sysfs_create_group(&d->gd->dev.kobj, &attr_group);
+	return sysfs_create_group(&disk_to_dev(d->gd)->kobj, &attr_group);
 }
 void
 aoedisk_rm_sysfs(struct aoedev *d)
 {
-	sysfs_remove_group(&d->gd->dev.kobj, &attr_group);
+	sysfs_remove_group(&disk_to_dev(d->gd)->kobj, &attr_group);
 }
 
 static int
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index 1778e4a2c672..7b3351260d56 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -403,7 +403,7 @@ static int nbd_do_it(struct nbd_device *lo)
 	BUG_ON(lo->magic != LO_MAGIC);
 
 	lo->pid = current->pid;
-	ret = sysfs_create_file(&lo->disk->dev.kobj, &pid_attr.attr);
+	ret = sysfs_create_file(&disk_to_dev(lo->disk)->kobj, &pid_attr.attr);
 	if (ret) {
 		printk(KERN_ERR "nbd: sysfs_create_file failed!");
 		return ret;
@@ -412,7 +412,7 @@ static int nbd_do_it(struct nbd_device *lo)
 	while ((req = nbd_read_stat(lo)) != NULL)
 		nbd_end_request(req);
 
-	sysfs_remove_file(&lo->disk->dev.kobj, &pid_attr.attr);
+	sysfs_remove_file(&disk_to_dev(lo->disk)->kobj, &pid_attr.attr);
 	return 0;
 }
 
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index a51a30e9eab3..70aa86c8807e 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1188,7 +1188,7 @@ static struct kobject *exact_match(dev_t dev, int *part, void *data)
 {
 	struct gendisk *p = data;
 	*part &= (1 << PARTN_BITS) - 1;
-	return &p->dev.kobj;
+	return &disk_to_dev(p)->kobj;
 }
 
 static int exact_lock(dev_t dev, void *data)
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 653624792eaf..637806695bb9 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1186,7 +1186,7 @@ static void event_callback(void *context)
 	list_splice_init(&md->uevent_list, &uevents);
 	spin_unlock_irqrestore(&md->uevent_lock, flags);
 
-	dm_send_uevents(&uevents, &md->disk->dev.kobj);
+	dm_send_uevents(&uevents, &disk_to_dev(md->disk)->kobj);
 
 	atomic_inc(&md->event_nr);
 	wake_up(&md->eventq);
@@ -1643,7 +1643,7 @@ out:
  *---------------------------------------------------------------*/
 void dm_kobject_uevent(struct mapped_device *md)
 {
-	kobject_uevent(&md->disk->dev.kobj, KOBJ_CHANGE);
+	kobject_uevent(&disk_to_dev(md->disk)->kobj, KOBJ_CHANGE);
 }
 
 uint32_t dm_next_uevent_seq(struct mapped_device *md)
diff --git a/drivers/md/md.c b/drivers/md/md.c
index deeac4b44173..96e9fccd2eab 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -1465,9 +1465,9 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
 		goto fail;
 
 	if (rdev->bdev->bd_part)
-		ko = &rdev->bdev->bd_part->dev.kobj;
+		ko = &part_to_dev(rdev->bdev->bd_part)->kobj;
 	else
-		ko = &rdev->bdev->bd_disk->dev.kobj;
+		ko = &disk_to_dev(rdev->bdev->bd_disk)->kobj;
 	if ((err = sysfs_create_link(&rdev->kobj, ko, "block"))) {
 		kobject_del(&rdev->kobj);
 		goto fail;
@@ -3470,8 +3470,8 @@ static struct kobject *md_probe(dev_t dev, int *part, void *data)
 	disk->queue = mddev->queue;
 	add_disk(disk);
 	mddev->gendisk = disk;
-	error = kobject_init_and_add(&mddev->kobj, &md_ktype, &disk->dev.kobj,
-				     "%s", "md");
+	error = kobject_init_and_add(&mddev->kobj, &md_ktype,
+				     &disk_to_dev(disk)->kobj, "%s", "md");
 	mutex_unlock(&disks_mutex);
 	if (error)
 		printk(KERN_WARNING "md: cannot register %s/md - name in use\n",
@@ -3761,7 +3761,7 @@ static int do_md_run(mddev_t * mddev)
 	sysfs_notify(&mddev->kobj, NULL, "array_state");
 	sysfs_notify(&mddev->kobj, NULL, "sync_action");
 	sysfs_notify(&mddev->kobj, NULL, "degraded");
-	kobject_uevent(&mddev->gendisk->dev.kobj, KOBJ_CHANGE);
+	kobject_uevent(&disk_to_dev(mddev->gendisk)->kobj, KOBJ_CHANGE);
 	return 0;
 }
 
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 2f2873b9a041..a02df22f37c3 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -543,9 +543,9 @@ EXPORT_SYMBOL(bd_release);
 static struct kobject *bdev_get_kobj(struct block_device *bdev)
 {
 	if (bdev->bd_contains != bdev)
-		return kobject_get(&bdev->bd_part->dev.kobj);
+		return kobject_get(&part_to_dev(bdev->bd_part)->kobj);
 	else
-		return kobject_get(&bdev->bd_disk->dev.kobj);
+		return kobject_get(&disk_to_dev(bdev->bd_disk)->kobj);
 }
 
 static struct kobject *bdev_get_holder(struct block_device *bdev)
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index 0d4b7f28f13f..ac0df3acdcda 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -309,7 +309,7 @@ static inline void disk_sysfs_add_subdirs(struct gendisk *disk)
 {
 	struct kobject *k;
 
-	k = kobject_get(&disk->dev.kobj);
+	k = kobject_get(&disk_to_dev(disk)->kobj);
 	disk->holder_dir = kobject_create_and_add("holders", k);
 	disk->slave_dir = kobject_create_and_add("slaves", k);
 	kobject_put(k);
@@ -322,7 +322,7 @@ static void delete_partition_rcu_cb(struct rcu_head *head)
 	part->start_sect = 0;
 	part->nr_sects = 0;
 	part_stat_set_all(part, 0);
-	put_device(&part->dev);
+	put_device(part_to_dev(part));
 }
 
 void delete_partition(struct gendisk *disk, int partno)
@@ -336,7 +336,7 @@ void delete_partition(struct gendisk *disk, int partno)
 	blk_free_devt(part_devt(part));
 	rcu_assign_pointer(disk->__part[partno-1], NULL);
 	kobject_put(part->holder_dir);
-	device_del(&part->dev);
+	device_del(part_to_dev(part));
 
 	call_rcu(&part->rcu_head, delete_partition_rcu_cb);
 }
@@ -354,6 +354,9 @@ int add_partition(struct gendisk *disk, int partno,
 {
 	struct hd_struct *p;
 	dev_t devt = MKDEV(0, 0);
+	struct device *ddev = disk_to_dev(disk);
+	struct device *pdev;
+	const char *dname;
 	int err;
 
 	if (disk->__part[partno - 1])
@@ -367,42 +370,43 @@ int add_partition(struct gendisk *disk, int partno,
 		err = -ENOMEM;
 		goto out_free;
 	}
+	pdev = part_to_dev(p);
+
 	p->start_sect = start;
 	p->nr_sects = len;
 	p->partno = partno;
 	p->policy = disk->policy;
 
-	if (isdigit(disk->dev.bus_id[strlen(disk->dev.bus_id)-1]))
-		snprintf(p->dev.bus_id, BUS_ID_SIZE,
-		"%sp%d", disk->dev.bus_id, partno);
+	dname = dev_name(ddev);
+	if (isdigit(dname[strlen(dname) - 1]))
+		snprintf(pdev->bus_id, BUS_ID_SIZE, "%sp%d", dname, partno);
 	else
-		snprintf(p->dev.bus_id, BUS_ID_SIZE,
-			 "%s%d", disk->dev.bus_id, partno);
+		snprintf(pdev->bus_id, BUS_ID_SIZE, "%s%d", dname, partno);
 
-	device_initialize(&p->dev);
-	p->dev.class = &block_class;
-	p->dev.type = &part_type;
-	p->dev.parent = &disk->dev;
+	device_initialize(pdev);
+	pdev->class = &block_class;
+	pdev->type = &part_type;
+	pdev->parent = ddev;
 
 	err = blk_alloc_devt(p, &devt);
 	if (err)
-		goto out_put;
-	p->dev.devt = devt;
+		goto out_free;
+	pdev->devt = devt;
 
 	/* delay uevent until 'holders' subdir is created */
-	p->dev.uevent_suppress = 1;
-	err = device_add(&p->dev);
+	pdev->uevent_suppress = 1;
+	err = device_add(pdev);
 	if (err)
 		goto out_put;
 
 	err = -ENOMEM;
-	p->holder_dir = kobject_create_and_add("holders", &p->dev.kobj);
+	p->holder_dir = kobject_create_and_add("holders", &pdev->kobj);
 	if (!p->holder_dir)
 		goto out_del;
 
-	p->dev.uevent_suppress = 0;
+	pdev->uevent_suppress = 0;
 	if (flags & ADDPART_FLAG_WHOLEDISK) {
-		err = device_create_file(&p->dev, &dev_attr_whole_disk);
+		err = device_create_file(pdev, &dev_attr_whole_disk);
 		if (err)
 			goto out_del;
 	}
@@ -412,8 +416,8 @@ int add_partition(struct gendisk *disk, int partno,
 	rcu_assign_pointer(disk->__part[partno - 1], p);
 
 	/* suppress uevent if the disk supresses it */
-	if (!disk->dev.uevent_suppress)
-		kobject_uevent(&p->dev.kobj, KOBJ_ADD);
+	if (!ddev->uevent_suppress)
+		kobject_uevent(&pdev->kobj, KOBJ_ADD);
 
 	return 0;
 
@@ -422,9 +426,9 @@ out_free:
 	return err;
 out_del:
 	kobject_put(p->holder_dir);
-	device_del(&p->dev);
+	device_del(pdev);
 out_put:
-	put_device(&p->dev);
+	put_device(pdev);
 	blk_free_devt(devt);
 	return err;
 }
@@ -432,30 +436,31 @@ out_put:
 /* Not exported, helper to add_disk(). */
 void register_disk(struct gendisk *disk)
 {
+	struct device *ddev = disk_to_dev(disk);
 	struct block_device *bdev;
 	struct disk_part_iter piter;
 	struct hd_struct *part;
 	char *s;
 	int err;
 
-	disk->dev.parent = disk->driverfs_dev;
+	ddev->parent = disk->driverfs_dev;
 
-	strlcpy(disk->dev.bus_id, disk->disk_name, BUS_ID_SIZE);
+	strlcpy(ddev->bus_id, disk->disk_name, BUS_ID_SIZE);
 	/* ewww... some of these buggers have / in the name... */
-	s = strchr(disk->dev.bus_id, '/');
+	s = strchr(ddev->bus_id, '/');
 	if (s)
 		*s = '!';
 
 	/* delay uevents, until we scanned partition table */
-	disk->dev.uevent_suppress = 1;
+	ddev->uevent_suppress = 1;
 
-	if (device_add(&disk->dev))
+	if (device_add(ddev))
 		return;
 #ifndef CONFIG_SYSFS_DEPRECATED
-	err = sysfs_create_link(block_depr, &disk->dev.kobj,
-				kobject_name(&disk->dev.kobj));
+	err = sysfs_create_link(block_depr, &ddev->kobj,
+				kobject_name(&ddev->kobj));
 	if (err) {
-		device_del(&disk->dev);
+		device_del(ddev);
 		return;
 	}
 #endif
@@ -481,13 +486,13 @@ void register_disk(struct gendisk *disk)
 
 exit:
 	/* announce disk after possible partitions are created */
-	disk->dev.uevent_suppress = 0;
-	kobject_uevent(&disk->dev.kobj, KOBJ_ADD);
+	ddev->uevent_suppress = 0;
+	kobject_uevent(&ddev->kobj, KOBJ_ADD);
 
 	/* announce possible partitions */
 	disk_part_iter_init(&piter, disk, 0);
 	while ((part = disk_part_iter_next(&piter)))
-		kobject_uevent(&part->dev.kobj, KOBJ_ADD);
+		kobject_uevent(&part_to_dev(part)->kobj, KOBJ_ADD);
 	disk_part_iter_exit(&piter);
 }
 
@@ -518,7 +523,7 @@ int rescan_partitions(struct gendisk *disk, struct block_device *bdev)
 		return -EIO;
 
 	/* tell userspace that the media / partition table may have changed */
-	kobject_uevent(&disk->dev.kobj, KOBJ_CHANGE);
+	kobject_uevent(&disk_to_dev(disk)->kobj, KOBJ_CHANGE);
 
 	for (p = 1; p < state->limit; p++) {
 		sector_t size = state->parts[p].size;
@@ -591,7 +596,7 @@ void del_gendisk(struct gendisk *disk)
 	kobject_put(disk->slave_dir);
 	disk->driverfs_dev = NULL;
 #ifndef CONFIG_SYSFS_DEPRECATED
-	sysfs_remove_link(block_depr, disk->dev.bus_id);
+	sysfs_remove_link(block_depr, dev_name(disk_to_dev(disk)));
 #endif
-	device_del(&disk->dev);
+	device_del(disk_to_dev(disk));
 }
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 6fc532424062..e4e18c509ac5 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -15,9 +15,11 @@
 
 #ifdef CONFIG_BLOCK
 
-#define kobj_to_dev(k) container_of(k, struct device, kobj)
-#define dev_to_disk(device) container_of(device, struct gendisk, dev)
-#define dev_to_part(device) container_of(device, struct hd_struct, dev)
+#define kobj_to_dev(k)		container_of((k), struct device, kobj)
+#define dev_to_disk(device)	container_of((device), struct gendisk, __dev)
+#define dev_to_part(device)	container_of((device), struct hd_struct, __dev)
+#define disk_to_dev(disk)	(&((disk)->__dev))
+#define part_to_dev(part)	(&((part)->__dev))
 
 extern struct device_type part_type;
 extern struct kobject *block_depr;
@@ -88,7 +90,7 @@ struct disk_stats {
 struct hd_struct {
 	sector_t start_sect;
 	sector_t nr_sects;
-	struct device dev;
+	struct device __dev;
 	struct kobject *holder_dir;
 	int policy, partno;
 #ifdef CONFIG_FAIL_MAKE_REQUEST
@@ -139,7 +141,7 @@ struct gendisk {
 
 	int flags;
 	struct device *driverfs_dev;  // FIXME: remove
-	struct device dev;
+	struct device __dev;
 	struct kobject *holder_dir;
 	struct kobject *slave_dir;
 
@@ -163,7 +165,7 @@ struct gendisk {
 static inline struct gendisk *part_to_disk(struct hd_struct *part)
 {
 	if (likely(part))
-		return dev_to_disk((part)->dev.parent);
+		return dev_to_disk(part_to_dev(part)->parent);
 	return NULL;
 }
 
@@ -174,12 +176,12 @@ static inline int disk_max_parts(struct gendisk *disk)
 
 static inline dev_t disk_devt(struct gendisk *disk)
 {
-	return disk->dev.devt;
+	return disk_to_dev(disk)->devt;
 }
 
 static inline dev_t part_devt(struct hd_struct *part)
 {
-	return part->dev.devt;
+	return part_to_dev(part)->devt;
 }
 
 extern struct hd_struct *disk_get_part(struct gendisk *disk, int partno);
@@ -187,7 +189,7 @@ extern struct hd_struct *disk_get_part(struct gendisk *disk, int partno);
 static inline void disk_put_part(struct hd_struct *part)
 {
 	if (likely(part))
-		put_device(&part->dev);
+		put_device(part_to_dev(part));
 }
 
 /*
-- 
cgit v1.2.3


From 0762b8bde9729f10f8e6249809660ff2ec3ad735 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Mon, 25 Aug 2008 19:56:12 +0900
Subject: block: always set bdev->bd_part

Till now, bdev->bd_part is set only if the bdev was for parts other
than part0.  This patch makes bdev->bd_part always set so that code
paths don't have to differenciate common handling.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/blk-core.c      |  2 +-
 drivers/md/md.c       |  5 +---
 fs/block_dev.c        | 67 ++++++++++++++++++++++++---------------------------
 fs/partitions/check.c |  7 +-----
 include/linux/genhd.h |  2 +-
 5 files changed, 35 insertions(+), 48 deletions(-)

(limited to 'drivers/md/md.c')

diff --git a/block/blk-core.c b/block/blk-core.c
index e0a5ee36849c..a4a7c08d2f20 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -1274,7 +1274,7 @@ __setup("fail_make_request=", setup_fail_make_request);
 static int should_fail_request(struct bio *bio)
 {
 	if ((bio->bi_bdev->bd_disk->flags & GENHD_FL_FAIL) ||
-	    (bio->bi_bdev->bd_part && bio->bi_bdev->bd_part->make_it_fail))
+	    bio->bi_bdev->bd_part->make_it_fail)
 		return should_fail(&fail_make_request, bio->bi_size);
 
 	return 0;
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 96e9fccd2eab..2bd9cf416123 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -1464,10 +1464,7 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
 	if ((err = kobject_add(&rdev->kobj, &mddev->kobj, "dev-%s", b)))
 		goto fail;
 
-	if (rdev->bdev->bd_part)
-		ko = &part_to_dev(rdev->bdev->bd_part)->kobj;
-	else
-		ko = &disk_to_dev(rdev->bdev->bd_disk)->kobj;
+	ko = &part_to_dev(rdev->bdev->bd_part)->kobj;
 	if ((err = sysfs_create_link(&rdev->kobj, ko, "block"))) {
 		kobject_del(&rdev->kobj);
 		goto fail;
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 57d572642854..c3fa19bd64df 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -540,14 +540,6 @@ EXPORT_SYMBOL(bd_release);
  *           /sys/block/sda/holders/dm-0 --> /sys/block/dm-0
  */
 
-static struct kobject *bdev_get_kobj(struct block_device *bdev)
-{
-	if (bdev->bd_contains != bdev)
-		return kobject_get(&part_to_dev(bdev->bd_part)->kobj);
-	else
-		return kobject_get(&disk_to_dev(bdev->bd_disk)->kobj);
-}
-
 static int add_symlink(struct kobject *from, struct kobject *to)
 {
 	if (!from || !to)
@@ -596,7 +588,7 @@ static int bd_holder_grab_dirs(struct block_device *bdev,
 	if (!bo->hdev)
 		goto fail_put_sdir;
 
-	bo->sdev = bdev_get_kobj(bdev);
+	bo->sdev = kobject_get(&part_to_dev(bdev->bd_part)->kobj);
 	if (!bo->sdev)
 		goto fail_put_hdev;
 
@@ -919,7 +911,6 @@ static int __blkdev_put(struct block_device *bdev, int for_part);
 
 static int do_open(struct block_device *bdev, struct file *file, int for_part)
 {
-	struct module *owner = NULL;
 	struct gendisk *disk;
 	struct hd_struct *part = NULL;
 	int ret;
@@ -941,25 +932,27 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part)
 
 	ret = -ENXIO;
 	file->f_mapping = bdev->bd_inode->i_mapping;
+
 	lock_kernel();
+
 	disk = get_gendisk(bdev->bd_dev, &partno);
-	if (!disk) {
-		unlock_kernel();
-		bdput(bdev);
-		return ret;
-	}
-	owner = disk->fops->owner;
+	if (!disk)
+		goto out_unlock_kernel;
+	part = disk_get_part(disk, partno);
+	if (!part)
+		goto out_unlock_kernel;
 
 	mutex_lock_nested(&bdev->bd_mutex, for_part);
 	if (!bdev->bd_openers) {
 		bdev->bd_disk = disk;
+		bdev->bd_part = part;
 		bdev->bd_contains = bdev;
 		if (!partno) {
 			struct backing_dev_info *bdi;
 			if (disk->fops->open) {
 				ret = disk->fops->open(bdev->bd_inode, file);
 				if (ret)
-					goto out_first;
+					goto out_clear;
 			}
 			if (!bdev->bd_openers) {
 				bd_set_size(bdev,(loff_t)get_capacity(disk)<<9);
@@ -975,31 +968,32 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part)
 			whole = bdget_disk(disk, 0);
 			ret = -ENOMEM;
 			if (!whole)
-				goto out_first;
+				goto out_clear;
 			BUG_ON(for_part);
 			ret = __blkdev_get(whole, file->f_mode, file->f_flags, 1);
 			if (ret)
-				goto out_first;
+				goto out_clear;
 			bdev->bd_contains = whole;
-			part = disk_get_part(disk, partno);
 			bdev->bd_inode->i_data.backing_dev_info =
 			   whole->bd_inode->i_data.backing_dev_info;
 			if (!(disk->flags & GENHD_FL_UP) ||
 			    !part || !part->nr_sects) {
 				ret = -ENXIO;
-				goto out_first;
+				goto out_clear;
 			}
-			bdev->bd_part = part;
 			bd_set_size(bdev, (loff_t)part->nr_sects << 9);
 		}
 	} else {
+		disk_put_part(part);
 		put_disk(disk);
-		module_put(owner);
+		module_put(disk->fops->owner);
+		part = NULL;
+		disk = NULL;
 		if (bdev->bd_contains == bdev) {
 			if (bdev->bd_disk->fops->open) {
 				ret = bdev->bd_disk->fops->open(bdev->bd_inode, file);
 				if (ret)
-					goto out;
+					goto out_unlock_bdev;
 			}
 			if (bdev->bd_invalidated)
 				rescan_partitions(bdev->bd_disk, bdev);
@@ -1012,20 +1006,24 @@ static int do_open(struct block_device *bdev, struct file *file, int for_part)
 	unlock_kernel();
 	return 0;
 
-out_first:
+ out_clear:
 	bdev->bd_disk = NULL;
+	bdev->bd_part = NULL;
 	bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info;
 	if (bdev != bdev->bd_contains)
 		__blkdev_put(bdev->bd_contains, 1);
 	bdev->bd_contains = NULL;
-	put_disk(disk);
-	disk_put_part(part);
-	module_put(owner);
-out:
+ out_unlock_bdev:
 	mutex_unlock(&bdev->bd_mutex);
+ out_unlock_kernel:
 	unlock_kernel();
-	if (ret)
-		bdput(bdev);
+
+	disk_put_part(part);
+	if (disk)
+		module_put(disk->fops->owner);
+	put_disk(disk);
+	bdput(bdev);
+
 	return ret;
 }
 
@@ -1110,11 +1108,8 @@ static int __blkdev_put(struct block_device *bdev, int for_part)
 
 		put_disk(disk);
 		module_put(owner);
-
-		if (bdev->bd_contains != bdev) {
-			disk_put_part(bdev->bd_part);
-			bdev->bd_part = NULL;
-		}
+		disk_put_part(bdev->bd_part);
+		bdev->bd_part = NULL;
 		bdev->bd_disk = NULL;
 		bdev->bd_inode->i_data.backing_dev_info = &default_backing_dev_info;
 		if (bdev != bdev->bd_contains)
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index f0f604950ff4..87298c0fc8ce 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -134,12 +134,7 @@ char *disk_name(struct gendisk *hd, int partno, char *buf)
 
 const char *bdevname(struct block_device *bdev, char *buf)
 {
-	int partno = 0;
-
-	if (bdev->bd_part)
-		partno = bdev->bd_part->partno;
-
-	return disk_name(bdev->bd_disk, partno, buf);
+	return disk_name(bdev->bd_disk, bdev->bd_part->partno, buf);
 }
 
 EXPORT_SYMBOL(bdevname);
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 2c0e1b597ab4..45a3682b5d87 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -412,7 +412,7 @@ extern void rand_initialize_disk(struct gendisk *disk);
 
 static inline sector_t get_start_sect(struct block_device *bdev)
 {
-	return bdev->bd_contains == bdev ? 0 : bdev->bd_part->start_sect;
+	return bdev->bd_part->start_sect;
 }
 static inline sector_t get_capacity(struct gendisk *disk)
 {
-- 
cgit v1.2.3


From 074a7aca7afa6f230104e8e65eba3420263714a5 Mon Sep 17 00:00:00 2001
From: Tejun Heo <tj@kernel.org>
Date: Mon, 25 Aug 2008 19:56:14 +0900
Subject: block: move stats from disk to part0

Move stats related fields - stamp, in_flight, dkstats - from disk to
part0 and unify stat handling such that...

* part_stat_*() now updates part0 together if the specified partition
  is not part0.  ie. part_stat_*() are now essentially all_stat_*().

* {disk|all}_stat_*() are gone.

* part_round_stats() is updated similary.  It handles part0 stats
  automatically and disk_round_stats() is killed.

* part_{inc|dec}_in_fligh() is implemented which automatically updates
  part0 stats for parts other than part0.

* disk_map_sector_rcu() is updated to return part0 if no part matches.
  Combined with the above changes, this makes NULL special case
  handling in callers unnecessary.

* Separate stats show code paths for disk are collapsed into part
  stats show code paths.

* Rename disk_stat_lock/unlock() to part_stat_lock/unlock()

While at it, reposition stat handling macros a bit and add missing
parentheses around macro parameters.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 block/blk-core.c           |  84 ++++++++++--------------
 block/blk-merge.c          |  12 ++--
 block/genhd.c              |  97 +++++++--------------------
 drivers/block/aoe/aoecmd.c |  12 ++--
 drivers/md/dm.c            |  27 ++++----
 drivers/md/linear.c        |   9 +--
 drivers/md/md.c            |   4 +-
 drivers/md/multipath.c     |   9 +--
 drivers/md/raid0.c         |   9 +--
 drivers/md/raid1.c         |   9 +--
 drivers/md/raid10.c        |   9 +--
 drivers/md/raid5.c         |   9 +--
 fs/partitions/check.c      |  12 ++--
 include/linux/genhd.h      | 159 +++++++++++++--------------------------------
 14 files changed, 165 insertions(+), 296 deletions(-)

(limited to 'drivers/md/md.c')

diff --git a/block/blk-core.c b/block/blk-core.c
index 505ec61067df..98138f002524 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -61,21 +61,17 @@ static void drive_stat_acct(struct request *rq, int new_io)
 	if (!blk_fs_request(rq) || !rq->rq_disk)
 		return;
 
-	cpu = disk_stat_lock();
+	cpu = part_stat_lock();
 	part = disk_map_sector_rcu(rq->rq_disk, rq->sector);
 
 	if (!new_io)
-		all_stat_inc(cpu, rq->rq_disk, part, merges[rw], rq->sector);
+		part_stat_inc(cpu, part, merges[rw]);
 	else {
-		disk_round_stats(cpu, rq->rq_disk);
-		rq->rq_disk->in_flight++;
-		if (part) {
-			part_round_stats(cpu, part);
-			part->in_flight++;
-		}
+		part_round_stats(cpu, part);
+		part_inc_in_flight(part);
 	}
 
-	disk_stat_unlock();
+	part_stat_unlock();
 }
 
 void blk_queue_congestion_threshold(struct request_queue *q)
@@ -983,8 +979,22 @@ static inline void add_request(struct request_queue *q, struct request *req)
 	__elv_add_request(q, req, ELEVATOR_INSERT_SORT, 0);
 }
 
-/*
- * disk_round_stats()	- Round off the performance stats on a struct
+static void part_round_stats_single(int cpu, struct hd_struct *part,
+				    unsigned long now)
+{
+	if (now == part->stamp)
+		return;
+
+	if (part->in_flight) {
+		__part_stat_add(cpu, part, time_in_queue,
+				part->in_flight * (now - part->stamp));
+		__part_stat_add(cpu, part, io_ticks, (now - part->stamp));
+	}
+	part->stamp = now;
+}
+
+/**
+ * part_round_stats()	- Round off the performance stats on a struct
  * disk_stats.
  *
  * The average IO queue length and utilisation statistics are maintained
@@ -998,36 +1008,15 @@ static inline void add_request(struct request_queue *q, struct request *req)
  * /proc/diskstats.  This accounts immediately for all queue usage up to
  * the current jiffies and restarts the counters again.
  */
-void disk_round_stats(int cpu, struct gendisk *disk)
-{
-	unsigned long now = jiffies;
-
-	if (now == disk->stamp)
-		return;
-
-	if (disk->in_flight) {
-		disk_stat_add(cpu, disk, time_in_queue,
-			      disk->in_flight * (now - disk->stamp));
-		disk_stat_add(cpu, disk, io_ticks, (now - disk->stamp));
-	}
-	disk->stamp = now;
-}
-EXPORT_SYMBOL_GPL(disk_round_stats);
-
 void part_round_stats(int cpu, struct hd_struct *part)
 {
 	unsigned long now = jiffies;
 
-	if (now == part->stamp)
-		return;
-
-	if (part->in_flight) {
-		part_stat_add(cpu, part, time_in_queue,
-			      part->in_flight * (now - part->stamp));
-		part_stat_add(cpu, part, io_ticks, (now - part->stamp));
-	}
-	part->stamp = now;
+	if (part->partno)
+		part_round_stats_single(cpu, &part_to_disk(part)->part0, now);
+	part_round_stats_single(cpu, part, now);
 }
+EXPORT_SYMBOL_GPL(part_round_stats);
 
 /*
  * queue lock must be held
@@ -1567,11 +1556,10 @@ static int __end_that_request_first(struct request *req, int error,
 		struct hd_struct *part;
 		int cpu;
 
-		cpu = disk_stat_lock();
+		cpu = part_stat_lock();
 		part = disk_map_sector_rcu(req->rq_disk, req->sector);
-		all_stat_add(cpu, req->rq_disk, part, sectors[rw],
-			     nr_bytes >> 9, req->sector);
-		disk_stat_unlock();
+		part_stat_add(cpu, part, sectors[rw], nr_bytes >> 9);
+		part_stat_unlock();
 	}
 
 	total_bytes = bio_nbytes = 0;
@@ -1758,19 +1746,15 @@ static void end_that_request_last(struct request *req, int error)
 		struct hd_struct *part;
 		int cpu;
 
-		cpu = disk_stat_lock();
+		cpu = part_stat_lock();
 		part = disk_map_sector_rcu(disk, req->sector);
 
-		all_stat_inc(cpu, disk, part, ios[rw], req->sector);
-		all_stat_add(cpu, disk, part, ticks[rw], duration, req->sector);
-		disk_round_stats(cpu, disk);
-		disk->in_flight--;
-		if (part) {
-			part_round_stats(cpu, part);
-			part->in_flight--;
-		}
+		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);
 
-		disk_stat_unlock();
+		part_stat_unlock();
 	}
 
 	if (req->end_io)
diff --git a/block/blk-merge.c b/block/blk-merge.c
index d926a24bf1fd..c77196d55899 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -390,17 +390,13 @@ static int attempt_merge(struct request_queue *q, struct request *req,
 		struct hd_struct *part;
 		int cpu;
 
-		cpu = disk_stat_lock();
+		cpu = part_stat_lock();
 		part = disk_map_sector_rcu(req->rq_disk, req->sector);
 
-		disk_round_stats(cpu, req->rq_disk);
-		req->rq_disk->in_flight--;
-		if (part) {
-			part_round_stats(cpu, part);
-			part->in_flight--;
-		}
+		part_round_stats(cpu, part);
+		part_dec_in_flight(part);
 
-		disk_stat_unlock();
+		part_stat_unlock();
 	}
 
 	req->ioprio = ioprio_best(req->ioprio, next->ioprio);
diff --git a/block/genhd.c b/block/genhd.c
index 06a252f2b967..e1cb96fb883e 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -176,7 +176,7 @@ EXPORT_SYMBOL_GPL(disk_part_iter_exit);
  * while preemption is disabled.
  *
  * RETURNS:
- * Found partition on success, NULL if there's no matching partition.
+ * Found partition on success, part0 is returned if no partition matches
  */
 struct hd_struct *disk_map_sector_rcu(struct gendisk *disk, sector_t sector)
 {
@@ -189,7 +189,7 @@ struct hd_struct *disk_map_sector_rcu(struct gendisk *disk, sector_t sector)
 		    sector < part->start_sect + part->nr_sects)
 			return part;
 	}
-	return NULL;
+	return &disk->part0;
 }
 EXPORT_SYMBOL_GPL(disk_map_sector_rcu);
 
@@ -580,24 +580,24 @@ void __init printk_all_partitions(void)
 		 * numbers in hex - the same format as the root=
 		 * option takes.
 		 */
-		printk("%s %10llu %s",
-		       bdevt_str(disk_devt(disk), devt_buf),
-		       (unsigned long long)get_capacity(disk) >> 1,
-		       disk_name(disk, 0, name_buf));
-		if (disk->driverfs_dev != NULL &&
-		    disk->driverfs_dev->driver != NULL)
-			printk(" driver: %s\n",
-			       disk->driverfs_dev->driver->name);
-		else
-			printk(" (driver?)\n");
+		disk_part_iter_init(&piter, disk, DISK_PITER_INCL_PART0);
+		while ((part = disk_part_iter_next(&piter))) {
+			bool is_part0 = part == &disk->part0;
 
-		/* now show the partitions */
-		disk_part_iter_init(&piter, disk, 0);
-		while ((part = disk_part_iter_next(&piter)))
-			printk("  %s %10llu %s\n",
+			printk("%s%s %10llu %s", is_part0 ? "" : "  ",
 			       bdevt_str(part_devt(part), devt_buf),
 			       (unsigned long long)part->nr_sects >> 1,
 			       disk_name(disk, part->partno, name_buf));
+			if (is_part0) {
+				if (disk->driverfs_dev != NULL &&
+				    disk->driverfs_dev->driver != NULL)
+					printk(" driver: %s\n",
+					      disk->driverfs_dev->driver->name);
+				else
+					printk(" (driver?)\n");
+			} else
+				printk("\n");
+		}
 		disk_part_iter_exit(&piter);
 	}
 	class_dev_iter_exit(&iter);
@@ -674,12 +674,7 @@ static int show_partition(struct seq_file *seqf, void *v)
 		return 0;
 
 	/* show the full disk and all non-0 size partitions of it */
-	seq_printf(seqf, "%4d  %7d %10llu %s\n",
-		MAJOR(disk_devt(sgp)), MINOR(disk_devt(sgp)),
-		(unsigned long long)get_capacity(sgp) >> 1,
-		disk_name(sgp, 0, buf));
-
-	disk_part_iter_init(&piter, sgp, 0);
+	disk_part_iter_init(&piter, sgp, DISK_PITER_INCL_PART0);
 	while ((part = disk_part_iter_next(&piter)))
 		seq_printf(seqf, "%4d  %7d %10llu %s\n",
 			   MAJOR(part_devt(part)), MINOR(part_devt(part)),
@@ -768,40 +763,13 @@ static ssize_t disk_capability_show(struct device *dev,
 	return sprintf(buf, "%x\n", disk->flags);
 }
 
-static ssize_t disk_stat_show(struct device *dev,
-			      struct device_attribute *attr, char *buf)
-{
-	struct gendisk *disk = dev_to_disk(dev);
-	int cpu;
-
-	cpu = disk_stat_lock();
-	disk_round_stats(cpu, disk);
-	disk_stat_unlock();
-	return sprintf(buf,
-		"%8lu %8lu %8llu %8u "
-		"%8lu %8lu %8llu %8u "
-		"%8u %8u %8u"
-		"\n",
-		disk_stat_read(disk, ios[READ]),
-		disk_stat_read(disk, merges[READ]),
-		(unsigned long long)disk_stat_read(disk, sectors[READ]),
-		jiffies_to_msecs(disk_stat_read(disk, ticks[READ])),
-		disk_stat_read(disk, ios[WRITE]),
-		disk_stat_read(disk, merges[WRITE]),
-		(unsigned long long)disk_stat_read(disk, sectors[WRITE]),
-		jiffies_to_msecs(disk_stat_read(disk, ticks[WRITE])),
-		disk->in_flight,
-		jiffies_to_msecs(disk_stat_read(disk, io_ticks)),
-		jiffies_to_msecs(disk_stat_read(disk, time_in_queue)));
-}
-
 static DEVICE_ATTR(range, S_IRUGO, disk_range_show, NULL);
 static DEVICE_ATTR(ext_range, S_IRUGO, disk_ext_range_show, NULL);
 static DEVICE_ATTR(removable, S_IRUGO, disk_removable_show, NULL);
 static DEVICE_ATTR(ro, S_IRUGO, disk_ro_show, NULL);
 static DEVICE_ATTR(size, S_IRUGO, part_size_show, NULL);
 static DEVICE_ATTR(capability, S_IRUGO, disk_capability_show, NULL);
-static DEVICE_ATTR(stat, S_IRUGO, disk_stat_show, NULL);
+static DEVICE_ATTR(stat, S_IRUGO, part_stat_show, NULL);
 #ifdef CONFIG_FAIL_MAKE_REQUEST
 static struct device_attribute dev_attr_fail =
 	__ATTR(make-it-fail, S_IRUGO|S_IWUSR, part_fail_show, part_fail_store);
@@ -836,7 +804,7 @@ static void disk_release(struct device *dev)
 
 	kfree(disk->random);
 	kfree(disk->__part);
-	free_disk_stats(disk);
+	free_part_stats(&disk->part0);
 	kfree(disk);
 }
 struct class block_class = {
@@ -873,28 +841,11 @@ static int diskstats_show(struct seq_file *seqf, void *v)
 				"\n\n");
 	*/
  
-	cpu = disk_stat_lock();
-	disk_round_stats(cpu, gp);
-	disk_stat_unlock();
-	seq_printf(seqf, "%4d %7d %s %lu %lu %llu %u %lu %lu %llu %u %u %u %u\n",
-		MAJOR(disk_devt(gp)), MINOR(disk_devt(gp)),
-		disk_name(gp, 0, buf),
-		disk_stat_read(gp, ios[0]), disk_stat_read(gp, merges[0]),
-		(unsigned long long)disk_stat_read(gp, sectors[0]),
-		jiffies_to_msecs(disk_stat_read(gp, ticks[0])),
-		disk_stat_read(gp, ios[1]), disk_stat_read(gp, merges[1]),
-		(unsigned long long)disk_stat_read(gp, sectors[1]),
-		jiffies_to_msecs(disk_stat_read(gp, ticks[1])),
-		gp->in_flight,
-		jiffies_to_msecs(disk_stat_read(gp, io_ticks)),
-		jiffies_to_msecs(disk_stat_read(gp, time_in_queue)));
-
-	/* now show all non-0 size partitions of it */
-	disk_part_iter_init(&piter, gp, 0);
+	disk_part_iter_init(&piter, gp, DISK_PITER_INCL_PART0);
 	while ((hd = disk_part_iter_next(&piter))) {
-		cpu = disk_stat_lock();
+		cpu = part_stat_lock();
 		part_round_stats(cpu, hd);
-		disk_stat_unlock();
+		part_stat_unlock();
 		seq_printf(seqf, "%4d %7d %s %lu %lu %llu "
 			   "%u %lu %lu %llu %u %u %u %u\n",
 			   MAJOR(part_devt(hd)), MINOR(part_devt(hd)),
@@ -1000,7 +951,7 @@ struct gendisk *alloc_disk_ext_node(int minors, int ext_minors, int node_id)
 		int tot_minors = minors + ext_minors;
 		int size = tot_minors * sizeof(struct hd_struct *);
 
-		if (!init_disk_stats(disk)) {
+		if (!init_part_stats(&disk->part0)) {
 			kfree(disk);
 			return NULL;
 		}
@@ -1008,7 +959,7 @@ struct gendisk *alloc_disk_ext_node(int minors, int ext_minors, int node_id)
 		disk->__part = kmalloc_node(size, GFP_KERNEL | __GFP_ZERO,
 					    node_id);
 		if (!disk->__part) {
-			free_disk_stats(disk);
+				free_part_stats(&disk->part0);
 			kfree(disk);
 			return NULL;
 		}
diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c
index 934800f979c9..961d29a53cab 100644
--- a/drivers/block/aoe/aoecmd.c
+++ b/drivers/block/aoe/aoecmd.c
@@ -758,15 +758,15 @@ diskstats(struct gendisk *disk, struct bio *bio, ulong duration, sector_t sector
 	struct hd_struct *part;
 	int cpu;
 
-	cpu = disk_stat_lock();
+	cpu = part_stat_lock();
 	part = disk_map_sector_rcu(disk, sector);
 
-	all_stat_inc(cpu, disk, part, ios[rw], sector);
-	all_stat_add(cpu, disk, part, ticks[rw], duration, sector);
-	all_stat_add(cpu, disk, part, sectors[rw], n_sect, sector);
-	all_stat_add(cpu, disk, part, io_ticks, duration, sector);
+	part_stat_inc(cpu, part, ios[rw]);
+	part_stat_add(cpu, part, ticks[rw], duration);
+	part_stat_add(cpu, part, sectors[rw], n_sect);
+	part_stat_add(cpu, part, io_ticks, duration);
 
-	disk_stat_unlock();
+	part_stat_unlock();
 }
 
 void
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 637806695bb9..327de03a5bdf 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -381,10 +381,10 @@ static void start_io_acct(struct dm_io *io)
 
 	io->start_time = jiffies;
 
-	cpu = disk_stat_lock();
-	disk_round_stats(cpu, dm_disk(md));
-	disk_stat_unlock();
-	dm_disk(md)->in_flight = atomic_inc_return(&md->pending);
+	cpu = part_stat_lock();
+	part_round_stats(cpu, &dm_disk(md)->part0);
+	part_stat_unlock();
+	dm_disk(md)->part0.in_flight = atomic_inc_return(&md->pending);
 }
 
 static int end_io_acct(struct dm_io *io)
@@ -395,12 +395,13 @@ static int end_io_acct(struct dm_io *io)
 	int pending, cpu;
 	int rw = bio_data_dir(bio);
 
-	cpu = disk_stat_lock();
-	disk_round_stats(cpu, dm_disk(md));
-	disk_stat_add(cpu, dm_disk(md), ticks[rw], duration);
-	disk_stat_unlock();
+	cpu = part_stat_lock();
+	part_round_stats(cpu, &dm_disk(md)->part0);
+	part_stat_add(cpu, &dm_disk(md)->part0, ticks[rw], duration);
+	part_stat_unlock();
 
-	dm_disk(md)->in_flight = pending = atomic_dec_return(&md->pending);
+	dm_disk(md)->part0.in_flight = pending =
+		atomic_dec_return(&md->pending);
 
 	return !pending;
 }
@@ -899,10 +900,10 @@ static int dm_request(struct request_queue *q, struct bio *bio)
 
 	down_read(&md->io_lock);
 
-	cpu = disk_stat_lock();
-	disk_stat_inc(cpu, dm_disk(md), ios[rw]);
-	disk_stat_add(cpu, dm_disk(md), sectors[rw], bio_sectors(bio));
-	disk_stat_unlock();
+	cpu = part_stat_lock();
+	part_stat_inc(cpu, &dm_disk(md)->part0, ios[rw]);
+	part_stat_add(cpu, &dm_disk(md)->part0, sectors[rw], bio_sectors(bio));
+	part_stat_unlock();
 
 	/*
 	 * If we're suspended we have to queue
diff --git a/drivers/md/linear.c b/drivers/md/linear.c
index 00cbc8e47294..c80ea90593d3 100644
--- a/drivers/md/linear.c
+++ b/drivers/md/linear.c
@@ -325,10 +325,11 @@ static int linear_make_request (struct request_queue *q, struct bio *bio)
 		return 0;
 	}
 
-	cpu = disk_stat_lock();
-	disk_stat_inc(cpu, mddev->gendisk, ios[rw]);
-	disk_stat_add(cpu, mddev->gendisk, sectors[rw], bio_sectors(bio));
-	disk_stat_unlock();
+	cpu = part_stat_lock();
+	part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]);
+	part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw],
+		      bio_sectors(bio));
+	part_stat_unlock();
 
 	tmp_dev = which_dev(mddev, bio->bi_sector);
 	block = bio->bi_sector >> 1;
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 2bd9cf416123..0a3a4bdcd4af 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -5546,8 +5546,8 @@ static int is_mddev_idle(mddev_t *mddev)
 	rcu_read_lock();
 	rdev_for_each_rcu(rdev, mddev) {
 		struct gendisk *disk = rdev->bdev->bd_contains->bd_disk;
-		curr_events = disk_stat_read(disk, sectors[0]) + 
-				disk_stat_read(disk, sectors[1]) - 
+		curr_events = part_stat_read(&disk->part0, sectors[0]) +
+				part_stat_read(&disk->part0, sectors[1]) -
 				atomic_read(&disk->sync_io);
 		/* sync IO will cause sync_io to increase before the disk_stats
 		 * as sync_io is counted when a request starts, and
diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c
index 182f5a94cdc5..8bb8794129b3 100644
--- a/drivers/md/multipath.c
+++ b/drivers/md/multipath.c
@@ -159,10 +159,11 @@ static int multipath_make_request (struct request_queue *q, struct bio * bio)
 	mp_bh->master_bio = bio;
 	mp_bh->mddev = mddev;
 
-	cpu = disk_stat_lock();
-	disk_stat_inc(cpu, mddev->gendisk, ios[rw]);
-	disk_stat_add(cpu, mddev->gendisk, sectors[rw], bio_sectors(bio));
-	disk_stat_unlock();
+	cpu = part_stat_lock();
+	part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]);
+	part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw],
+		      bio_sectors(bio));
+	part_stat_unlock();
 
 	mp_bh->path = multipath_map(conf);
 	if (mp_bh->path < 0) {
diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c
index e26030fa59ab..f52f442a735f 100644
--- a/drivers/md/raid0.c
+++ b/drivers/md/raid0.c
@@ -406,10 +406,11 @@ static int raid0_make_request (struct request_queue *q, struct bio *bio)
 		return 0;
 	}
 
-	cpu = disk_stat_lock();
-	disk_stat_inc(cpu, mddev->gendisk, ios[rw]);
-	disk_stat_add(cpu, mddev->gendisk, sectors[rw], bio_sectors(bio));
-	disk_stat_unlock();
+	cpu = part_stat_lock();
+	part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]);
+	part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw],
+		      bio_sectors(bio));
+	part_stat_unlock();
 
 	chunk_size = mddev->chunk_size >> 10;
 	chunk_sects = mddev->chunk_size >> 9;
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index babb13036f93..b9764429d856 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -804,10 +804,11 @@ static int make_request(struct request_queue *q, struct bio * bio)
 
 	bitmap = mddev->bitmap;
 
-	cpu = disk_stat_lock();
-	disk_stat_inc(cpu, mddev->gendisk, ios[rw]);
-	disk_stat_add(cpu, mddev->gendisk, sectors[rw], bio_sectors(bio));
-	disk_stat_unlock();
+	cpu = part_stat_lock();
+	part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]);
+	part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw],
+		      bio_sectors(bio));
+	part_stat_unlock();
 
 	/*
 	 * make_request() can abort the operation when READA is being
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 5ec80da0a9d7..5f990133f5ef 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -844,10 +844,11 @@ static int make_request(struct request_queue *q, struct bio * bio)
 	 */
 	wait_barrier(conf);
 
-	cpu = disk_stat_lock();
-	disk_stat_inc(cpu, mddev->gendisk, ios[rw]);
-	disk_stat_add(cpu, mddev->gendisk, sectors[rw], bio_sectors(bio));
-	disk_stat_unlock();
+	cpu = part_stat_lock();
+	part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]);
+	part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw],
+		      bio_sectors(bio));
+	part_stat_unlock();
 
 	r10_bio = mempool_alloc(conf->r10bio_pool, GFP_NOIO);
 
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 5899f211515f..ae16794bef20 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -3396,10 +3396,11 @@ static int make_request(struct request_queue *q, struct bio * bi)
 
 	md_write_start(mddev, bi);
 
-	cpu = disk_stat_lock();
-	disk_stat_inc(cpu, mddev->gendisk, ios[rw]);
-	disk_stat_add(cpu, mddev->gendisk, sectors[rw], bio_sectors(bi));
-	disk_stat_unlock();
+	cpu = part_stat_lock();
+	part_stat_inc(cpu, &mddev->gendisk->part0, ios[rw]);
+	part_stat_add(cpu, &mddev->gendisk->part0, sectors[rw],
+		      bio_sectors(bi));
+	part_stat_unlock();
 
 	if (rw == READ &&
 	     mddev->reshape_position == MaxSector &&
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index 60592d9f43b6..f517869e8d10 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -210,15 +210,15 @@ ssize_t part_size_show(struct device *dev,
 	return sprintf(buf, "%llu\n",(unsigned long long)p->nr_sects);
 }
 
-static ssize_t part_stat_show(struct device *dev,
-			      struct device_attribute *attr, char *buf)
+ssize_t part_stat_show(struct device *dev,
+		       struct device_attribute *attr, char *buf)
 {
 	struct hd_struct *p = dev_to_part(dev);
 	int cpu;
 
-	cpu = disk_stat_lock();
+	cpu = part_stat_lock();
 	part_round_stats(cpu, p);
-	disk_stat_unlock();
+	part_stat_unlock();
 	return sprintf(buf,
 		"%8lu %8lu %8llu %8u "
 		"%8lu %8lu %8llu %8u "
@@ -575,8 +575,8 @@ void del_gendisk(struct gendisk *disk)
 	set_capacity(disk, 0);
 	disk->flags &= ~GENHD_FL_UP;
 	unlink_gendisk(disk);
-	disk_stat_set_all(disk, 0);
-	disk->stamp = 0;
+	part_stat_set_all(&disk->part0, 0);
+	disk->part0.stamp = 0;
 
 	kobject_put(disk->part0.holder_dir);
 	kobject_put(disk->slave_dir);
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index 3d15b42dc352..c90e1b4fbe5a 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -145,13 +145,6 @@ struct gendisk {
 	struct timer_rand_state *random;
 
 	atomic_t sync_io;		/* RAID */
-	unsigned long stamp;
-	int in_flight;
-#ifdef	CONFIG_SMP
-	struct disk_stats *dkstats;
-#else
-	struct disk_stats dkstats;
-#endif
 	struct work_struct async_notify;
 #ifdef  CONFIG_BLK_DEV_INTEGRITY
 	struct blk_integrity *integrity;
@@ -232,46 +225,18 @@ extern struct hd_struct *disk_map_sector_rcu(struct gendisk *disk,
  * internal use only.
  */
 #ifdef	CONFIG_SMP
-#define disk_stat_lock()	({ rcu_read_lock(); get_cpu(); })
-#define disk_stat_unlock()	do { put_cpu(); rcu_read_unlock(); } while (0)
-
-#define disk_stat_add(cpu, gendiskp, field, addnd)			\
-	(per_cpu_ptr(gendiskp->dkstats, cpu)->field += addnd)
-
-#define disk_stat_read(gendiskp, field)					\
-({									\
-	typeof(gendiskp->dkstats->field) res = 0;			\
-	int i;								\
-	for_each_possible_cpu(i)					\
-		res += per_cpu_ptr(gendiskp->dkstats, i)->field;	\
-	res;								\
-})
-
-static inline void disk_stat_set_all(struct gendisk *gendiskp, int value)
-{
-	int i;
-
-	for_each_possible_cpu(i)
-		memset(per_cpu_ptr(gendiskp->dkstats, i), value,
-				sizeof(struct disk_stats));
-}		
+#define part_stat_lock()	({ rcu_read_lock(); get_cpu(); })
+#define part_stat_unlock()	do { put_cpu(); rcu_read_unlock(); } while (0)
 
-#define part_stat_add(cpu, part, field, addnd)				\
-	(per_cpu_ptr(part->dkstats, cpu)->field += addnd)
-
-#define all_stat_add(cpu, gendiskp, part, field, addnd, sector)		\
-({									\
-	if (part)							\
-		part_stat_add(cpu, part, field, addnd);			\
-	disk_stat_add(cpu, gendiskp, field, addnd);			\
-})
+#define __part_stat_add(cpu, part, field, addnd)			\
+	(per_cpu_ptr((part)->dkstats, (cpu))->field += (addnd))
 
 #define part_stat_read(part, field)					\
 ({									\
-	typeof(part->dkstats->field) res = 0;				\
+	typeof((part)->dkstats->field) res = 0;				\
 	int i;								\
 	for_each_possible_cpu(i)					\
-		res += per_cpu_ptr(part->dkstats, i)->field;		\
+		res += per_cpu_ptr((part)->dkstats, i)->field;		\
 	res;								\
 })
 
@@ -284,109 +249,73 @@ static inline void part_stat_set_all(struct hd_struct *part, int value)
 				sizeof(struct disk_stats));
 }
 
-#else /* !CONFIG_SMP */
-#define disk_stat_lock()	({ rcu_read_lock(); 0; })
-#define disk_stat_unlock()	rcu_read_unlock()
-
-#define disk_stat_add(cpu, gendiskp, field, addnd)			\
-	(gendiskp->dkstats.field += addnd)
-#define disk_stat_read(gendiskp, field)	(gendiskp->dkstats.field)
-
-static inline void disk_stat_set_all(struct gendisk *gendiskp, int value)
+static inline int init_part_stats(struct hd_struct *part)
 {
-	memset(&gendiskp->dkstats, value, sizeof (struct disk_stats));
+	part->dkstats = alloc_percpu(struct disk_stats);
+	if (!part->dkstats)
+		return 0;
+	return 1;
 }
 
-#define part_stat_add(cpu, part, field, addnd)				\
-	(part->dkstats.field += addnd)
-
-#define all_stat_add(cpu, gendiskp, part, field, addnd, sector)		\
-({									\
-	if (part)							\
-		part_stat_add(cpu, part, field, addnd);			\
-	disk_stat_add(cpu, gendiskp, field, addnd);			\
-})
-
-#define part_stat_read(part, field)	(part->dkstats.field)
-
-static inline void part_stat_set_all(struct hd_struct *part, int value)
+static inline void free_part_stats(struct hd_struct *part)
 {
-	memset(&part->dkstats, value, sizeof(struct disk_stats));
+	free_percpu(part->dkstats);
 }
 
-#endif /* CONFIG_SMP */
-
-#define disk_stat_dec(cpu, gendiskp, field)				\
-	disk_stat_add(cpu, gendiskp, field, -1)
-#define disk_stat_inc(cpu, gendiskp, field)				\
-	disk_stat_add(cpu, gendiskp, field, 1)
-#define disk_stat_sub(cpu, gendiskp, field, subnd)			\
-	disk_stat_add(cpu, gendiskp, field, -subnd)
-
-#define part_stat_dec(cpu, gendiskp, field)				\
-	part_stat_add(cpu, gendiskp, field, -1)
-#define part_stat_inc(cpu, gendiskp, field)				\
-	part_stat_add(cpu, gendiskp, field, 1)
-#define part_stat_sub(cpu, gendiskp, field, subnd)			\
-	part_stat_add(cpu, gendiskp, field, -subnd)
+#else /* !CONFIG_SMP */
+#define part_stat_lock()	({ rcu_read_lock(); 0; })
+#define part_stat_unlock()	rcu_read_unlock()
 
-#define all_stat_dec(cpu, gendiskp, field, sector)			\
-	all_stat_add(cpu, gendiskp, field, -1, sector)
-#define all_stat_inc(cpu, gendiskp, part, field, sector)		\
-	all_stat_add(cpu, gendiskp, part, field, 1, sector)
-#define all_stat_sub(cpu, gendiskp, part, field, subnd, sector)		\
-	all_stat_add(cpu, gendiskp, part, field, -subnd, sector)
+#define __part_stat_add(cpu, part, field, addnd)				\
+	((part)->dkstats.field += addnd)
 
-/* Inlines to alloc and free disk stats in struct gendisk */
-#ifdef  CONFIG_SMP
-static inline int init_disk_stats(struct gendisk *disk)
-{
-	disk->dkstats = alloc_percpu(struct disk_stats);
-	if (!disk->dkstats)
-		return 0;
-	return 1;
-}
+#define part_stat_read(part, field)	((part)->dkstats.field)
 
-static inline void free_disk_stats(struct gendisk *disk)
+static inline void part_stat_set_all(struct hd_struct *part, int value)
 {
-	free_percpu(disk->dkstats);
+	memset(&part->dkstats, value, sizeof(struct disk_stats));
 }
 
 static inline int init_part_stats(struct hd_struct *part)
 {
-	part->dkstats = alloc_percpu(struct disk_stats);
-	if (!part->dkstats)
-		return 0;
 	return 1;
 }
 
 static inline void free_part_stats(struct hd_struct *part)
 {
-	free_percpu(part->dkstats);
 }
 
-#else	/* CONFIG_SMP */
-static inline int init_disk_stats(struct gendisk *disk)
-{
-	return 1;
-}
+#endif /* CONFIG_SMP */
 
-static inline void free_disk_stats(struct gendisk *disk)
-{
-}
+#define part_stat_add(cpu, part, field, addnd)	do {			\
+	__part_stat_add((cpu), (part), field, addnd);			\
+	if ((part)->partno)						\
+		__part_stat_add((cpu), &part_to_disk((part))->part0,	\
+				field, addnd);				\
+} while (0)
 
-static inline int init_part_stats(struct hd_struct *part)
+#define part_stat_dec(cpu, gendiskp, field)				\
+	part_stat_add(cpu, gendiskp, field, -1)
+#define part_stat_inc(cpu, gendiskp, field)				\
+	part_stat_add(cpu, gendiskp, field, 1)
+#define part_stat_sub(cpu, gendiskp, field, subnd)			\
+	part_stat_add(cpu, gendiskp, field, -subnd)
+
+static inline void part_inc_in_flight(struct hd_struct *part)
 {
-	return 1;
+	part->in_flight++;
+	if (part->partno)
+		part_to_disk(part)->part0.in_flight++;
 }
 
-static inline void free_part_stats(struct hd_struct *part)
+static inline void part_dec_in_flight(struct hd_struct *part)
 {
+	part->in_flight--;
+	if (part->partno)
+		part_to_disk(part)->part0.in_flight--;
 }
-#endif	/* CONFIG_SMP */
 
 /* drivers/block/ll_rw_blk.c */
-extern void disk_round_stats(int cpu, struct gendisk *disk);
 extern void part_round_stats(int cpu, struct hd_struct *part);
 
 /* drivers/block/genhd.c */
@@ -595,6 +524,8 @@ extern void blk_unregister_region(dev_t devt, unsigned long range);
 
 extern ssize_t part_size_show(struct device *dev,
 			      struct device_attribute *attr, char *buf);
+extern ssize_t part_stat_show(struct device *dev,
+			      struct device_attribute *attr, char *buf);
 #ifdef CONFIG_FAIL_MAKE_REQUEST
 extern ssize_t part_fail_show(struct device *dev,
 			      struct device_attribute *attr, char *buf);
-- 
cgit v1.2.3