[PATCH] ublk: fix NULL pointer dereference in ublk_ctrl_set_size()

[PATCH] ublk: fix NULL pointer dereference in ublk_ctrl_set_size()

[Mehul Rao]

ublk_ctrl_set_size() unconditionally dereferences ub->ub_disk via
set_capacity_and_notify() without checking if it is NULL.

ub->ub_disk is NULL before UBLK_CMD_START_DEV completes (it is only
assigned in ublk_ctrl_start_dev()) and after UBLK_CMD_STOP_DEV runs
(ublk_detach_disk() sets it to NULL). Since the UBLK_CMD_UPDATE_SIZE
handler performs no state validation, a user can trigger a NULL pointer
dereference by sending UPDATE_SIZE to a device that has been added but
not yet started, or one that has been stopped.

Fix this by checking ub->ub_disk under ub->mutex before dereferencing
it, and returning -ENODEV if the disk is not available.

Fixes: 98b995660bff ("ublk: Add UBLK_U_CMD_UPDATE_SIZE")
Cc: stable@vger.kernel.org
Signed-off-by: Mehul Rao <mehulrao@gmail.com>
---
 drivers/block/ublk_drv.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c
index 004f36724..41ed30a18 100644
--- a/drivers/block/ublk_drv.c
+++ b/drivers/block/ublk_drv.c
@@ -5006,15 +5006,22 @@ static int ublk_ctrl_get_features(const struct ublksrv_ctrl_cmd *header)
 	return 0;
 }
 
-static void ublk_ctrl_set_size(struct ublk_device *ub, const struct ublksrv_ctrl_cmd *header)
+static int ublk_ctrl_set_size(struct ublk_device *ub, const struct ublksrv_ctrl_cmd *header)
 {
 	struct ublk_param_basic *p = &ub->params.basic;
 	u64 new_size = header->data[0];
+	int ret = 0;
 
 	mutex_lock(&ub->mutex);
+	if (!ub->ub_disk) {
+		ret = -ENODEV;
+		goto out;
+	}
 	p->dev_sectors = new_size;
 	set_capacity_and_notify(ub->ub_disk, p->dev_sectors);
+out:
 	mutex_unlock(&ub->mutex);
+	return ret;
 }
 
 struct count_busy {
@@ -5335,8 +5342,7 @@ static int ublk_ctrl_uring_cmd(struct io_uring_cmd *cmd,
 		ret = ublk_ctrl_end_recovery(ub, &header);
 		break;
 	case UBLK_CMD_UPDATE_SIZE:
-		ublk_ctrl_set_size(ub, &header);
-		ret = 0;
+		ret = ublk_ctrl_set_size(ub, &header);
 		break;
 	case UBLK_CMD_QUIESCE_DEV:
 		ret = ublk_ctrl_quiesce_dev(ub, &header);
-- 
2.53.0

Re: [PATCH] ublk: fix NULL pointer dereference in ublk_ctrl_set_size()

[Ming Lei]

On Thu, Mar 05, 2026 at 02:31:46PM -0500, Mehul Rao wrote:
> ublk_ctrl_set_size() unconditionally dereferences ub->ub_disk via
> set_capacity_and_notify() without checking if it is NULL.
> 
> ub->ub_disk is NULL before UBLK_CMD_START_DEV completes (it is only
> assigned in ublk_ctrl_start_dev()) and after UBLK_CMD_STOP_DEV runs
> (ublk_detach_disk() sets it to NULL). Since the UBLK_CMD_UPDATE_SIZE
> handler performs no state validation, a user can trigger a NULL pointer
> dereference by sending UPDATE_SIZE to a device that has been added but
> not yet started, or one that has been stopped.
> 
> Fix this by checking ub->ub_disk under ub->mutex before dereferencing
> it, and returning -ENODEV if the disk is not available.
> 
> Fixes: 98b995660bff ("ublk: Add UBLK_U_CMD_UPDATE_SIZE")
> Cc: stable@vger.kernel.org
> Signed-off-by: Mehul Rao <mehulrao@gmail.com>

Looks fine given ublk_detach_disk() is called with ub->mutex grabbed:

Reviewed-by: Ming Lei <ming.lei@redhat.com>


Thanks,
Ming

Re: [PATCH] ublk: fix NULL pointer dereference in ublk_ctrl_set_size()

[Jens Axboe]

On Thu, 05 Mar 2026 14:31:46 -0500, Mehul Rao wrote:
> ublk_ctrl_set_size() unconditionally dereferences ub->ub_disk via
> set_capacity_and_notify() without checking if it is NULL.
> 
> ub->ub_disk is NULL before UBLK_CMD_START_DEV completes (it is only
> assigned in ublk_ctrl_start_dev()) and after UBLK_CMD_STOP_DEV runs
> (ublk_detach_disk() sets it to NULL). Since the UBLK_CMD_UPDATE_SIZE
> handler performs no state validation, a user can trigger a NULL pointer
> dereference by sending UPDATE_SIZE to a device that has been added but
> not yet started, or one that has been stopped.
> 
> [...]

Applied, thanks!

[1/1] ublk: fix NULL pointer dereference in ublk_ctrl_set_size()
      commit: 25966fc097691e5c925ad080f64a2f19c5fd940a

Best regards,
-- 
Jens Axboe