On 2/29/2016 1:29 PM, [email protected] wrote: > He said looking into it, he didn't find any legitimate uses of such ioctl.
That was the other thing I was wondering about: why would such a silly and security problematic ioctl exist in the first place? I guess that answers it, and removing or restricting it means we don't break job control. That sounds like the proper fix instead of using setsid().
