[PATCH 3.16 190/328] nbd: don't allow invalid blocksize settings

[Ben Hutchings]

3.16.62-rc1 review patch.  If anyone has any objections, please let me know.

------------------

From: Jens Axboe <axboe@xxxxxxxxx>

commit bc811f05d77f47059c197a98b6ad242eb03999cb upstream.

syzbot reports a divide-by-zero off the NBD_SET_BLKSIZE ioctl.
We need proper validation of the input here. Not just if it's
zero, but also if the value is a power-of-2 and in a valid
range. Add that.

Reported-by: syzbot <syzbot+25dbecbec1e62c6b0dd4@xxxxxxxxxxxxxxxxxxxxxxxxx>
Reviewed-by: Josef Bacik <josef@xxxxxxxxxxxxxx>
Signed-off-by: Jens Axboe <axboe@xxxxxxxxx>
[bwh: Backported to 3.16: adjust context]
Signed-off-by: Ben Hutchings <ben@xxxxxxxxxxxxxxx>
---
 drivers/block/nbd.c | 3 +++
 1 file changed, 3 insertions(+)

--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -667,6 +667,9 @@ static int __nbd_ioctl(struct block_devi
 	}
 
 	case NBD_SET_BLKSIZE:
+		if (!arg || !is_power_of_2(arg) || arg < 512 ||
+		    arg > PAGE_SIZE)
+			return -EINVAL;
 		nbd->blksize = arg;
 		nbd->bytesize &= ~(nbd->blksize-1);
 		bdev->bd_inode->i_size = nbd->bytesize;