On 3/19/2021 2:28 AM, Borislav Petkov wrote:

On Thu, Mar 18, 2021 at 12:05:58PM -0700, Yu, Yu-cheng wrote:

quoted

Maybe I would add comments here.

Yap.

Also, looking forward in the set, I see prctl_set() and that is also
done on current so should be ok.

In any case, yes, documenting the assumptions and expectations wrt
current here is a good idea.

quoted

If vm_munmap() returns -EINTR, mmap_lock is held by something else. That
lock should not be held forever.  For other types of error, the loop stops.

Ok I guess. The subsequent WARN_ON_ONCE() looks weird too but that
should not fire, right? :)

Thx.

That should not fire.

Yu-cheng