Permalink
Browse files

loop: fix concurrent lo_open/lo_release

范龙飞 reports that KASAN can report a use-after-free in __lock_acquire.
The reason is due to insufficient serialization in lo_release(), which
will continue to use the loop device even after it has decremented the
lo_refcnt to zero.

In the meantime, another process can come in, open the loop device
again as it is being shut down. Confusion ensues.

Reported-by: 范龙飞 <long7573@126.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
  • Loading branch information...
torvalds authored and axboe committed Jan 6, 2018
1 parent cbf3a95 commit ae6650163c66a7eff1acd6eb8b0f752dcfa8eba5
Showing with 8 additions and 2 deletions.
  1. +8 −2 drivers/block/loop.c
View
@@ -1581,9 +1581,8 @@ static int lo_open(struct block_device *bdev, fmode_t mode)
return err;
}
static void lo_release(struct gendisk *disk, fmode_t mode)
static void __lo_release(struct loop_device *lo)
{
struct loop_device *lo = disk->private_data;
int err;
if (atomic_dec_return(&lo->lo_refcnt))
@@ -1610,6 +1609,13 @@ static void lo_release(struct gendisk *disk, fmode_t mode)
mutex_unlock(&lo->lo_ctl_mutex);
}
static void lo_release(struct gendisk *disk, fmode_t mode)
{
mutex_lock(&loop_index_mutex);
__lo_release(disk->private_data);
mutex_unlock(&loop_index_mutex);
}
static const struct block_device_operations lo_fops = {
.owner = THIS_MODULE,
.open = lo_open,

0 comments on commit ae66501

Please sign in to comment.