There is a race over dev->radio_tx_rds_owner between the two functions mentioned below: Thread-1 Thread-2 vivid_fop_release() vivid_radio_rx_read() mutex_unlock(&dev->mutex) mutex_lock_interruptible(&dev->mutex) ... dev->radio_rx_rds_owner = file->private_data; ... if (file->private_data == dev->radio_rx_rds_owner) { dev->radio_tx_rds_last_block = 0; dev->radio_tx_rds_owner = NULL; } This race can be fixed by only releasing the lock after vivid_fop_release() finishes the checks. Signed-off-by: Sishuai Gong <sishuai.system@xxxxxxxxx> --- drivers/media/test-drivers/vivid/vivid-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/media/test-drivers/vivid/vivid-core.c b/drivers/media/test-drivers/vivid/vivid-core.c index c2167ccfd222..e95bdccfc18e 100644 --- a/drivers/media/test-drivers/vivid/vivid-core.c +++ b/drivers/media/test-drivers/vivid/vivid-core.c @@ -628,7 +628,6 @@ static int vivid_fop_release(struct file *file) v4l2_info(&dev->v4l2_dev, "reconnect\n"); vivid_reconnect(dev); } - mutex_unlock(&dev->mutex); if (file->private_data == dev->radio_rx_rds_owner) { dev->radio_rx_rds_last_block = 0; dev->radio_rx_rds_owner = NULL; @@ -637,6 +636,7 @@ static int vivid_fop_release(struct file *file) dev->radio_tx_rds_last_block = 0; dev->radio_tx_rds_owner = NULL; } + mutex_unlock(&dev->mutex); if (vdev->queue) return vb2_fop_release(file); return v4l2_fh_release(file); -- 2.39.2 (Apple Git-143)