block: fix disk->part[] dereferencing race

disk->part[] is protected by its matching bdev's lock.  However,
non-critical accesses like collecting stats and printing out sysfs and
proc information used to be performed without any locking.  As
partitions can come and go dynamically, partitions can go away
underneath those non-critical accesses.  As some of those accesses are
writes, this theoretically can lead to silent corruption.

This patch fixes the race by using RCU for the partition array and dev
reference counter to hold partitions.

* Rename disk->part[] to disk->__part[] to make sure no one outside
  genhd layer proper accesses it directly.

* Use RCU for disk->__part[] dereferencing.

* Implement disk_{get|put}_part() which can be used to get and put
  partitions from gendisk respectively.

* Iterators are implemented to help iterate through all partitions
  safely.

* Functions which require RCU readlock are marked with _rcu suffix.

* Use disk_put_part() in __blkdev_put() instead of directly putting
  the contained kobject.

Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

1 files changed, 16 insertions, 10 deletions

diff --git a/block/ioctl.c b/block/ioctl.c
index 403f7d7e0c28..a5f672ad55f6 100644
--- a/block/ioctl.c
+++ b/block/ioctl.c
@@ -12,11 +12,12 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
 {
 	struct block_device *bdevp;
 	struct gendisk *disk;
+	struct hd_struct *part;
 	struct blkpg_ioctl_arg a;
 	struct blkpg_partition p;
+	struct disk_part_iter piter;
 	long long start, length;
 	int partno;
-	int i;
 	int err;
 
 	if (!capable(CAP_SYS_ADMIN))
@@ -47,28 +48,33 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
 			mutex_lock(&bdev->bd_mutex);
 
 			/* overlap? */
-			for (i = 0; i < disk_max_parts(disk); i++) {
-				struct hd_struct *s = disk->part[i];
-
-				if (!s)
-					continue;
-				if (!(start+length <= s->start_sect ||
-				      start >= s->start_sect + s->nr_sects)) {
+			disk_part_iter_init(&piter, disk,
+					    DISK_PITER_INCL_EMPTY);
+			while ((part = disk_part_iter_next(&piter))) {
+				if (!(start + length <= part->start_sect ||
+				      start >= part->start_sect + part->nr_sects)) {
+					disk_part_iter_exit(&piter);
 					mutex_unlock(&bdev->bd_mutex);
 					return -EBUSY;
 				}
 			}
+			disk_part_iter_exit(&piter);
+
 			/* all seems OK */
 			err = add_partition(disk, partno, start, length,
 					    ADDPART_FLAG_NONE);
 			mutex_unlock(&bdev->bd_mutex);
 			return err;
 		case BLKPG_DEL_PARTITION:
-			if (!disk->part[partno - 1])
+			part = disk_get_part(disk, partno);
+			if (!part)
 				return -ENXIO;
-			bdevp = bdget_disk(disk, partno);
+
+			bdevp = bdget(part_devt(part));
+			disk_put_part(part);
 			if (!bdevp)
 				return -ENOMEM;
+
 			mutex_lock(&bdevp->bd_mutex);
 			if (bdevp->bd_openers) {
 				mutex_unlock(&bdevp->bd_mutex);