On Mon, 2013-09-02 at 13:25 +0200, Gerd Hoffmann wrote:
> +static void uas_zap_dead(struct uas_dev_info *devinfo)
> +{
> + struct uas_cmd_info *cmdinfo;
> + struct uas_cmd_info *temp;
> + struct list_head list;
> + unsigned long flags;
> +
> + spin_lock_irq(&uas_lists_lock);
> + list_replace_init(&uas_dead_list, &list);
> + spin_unlock_irq(&uas_lists_lock);
This looks like a window for a race.
> + spin_lock_irqsave(&devinfo->lock, flags);
> + list_for_each_entry_safe(cmdinfo, temp, &list, dead) {
> +
What happens if list entries are on the private list, when
the function is called for another device? It looks to me like
the di==devinfo test could put them back on the list although
they would need to be canceled.
Regards
Oliver
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
