Hi Nadav,
On Sat, Jan 29, 2022 at 10:23:55PM -0800, Nadav Amit wrote:
> Using userfautlfd and looking at the kernel code, I encountered a usability
> issue that complicates userspace UFFD-monitor implementation. I obviosuly
> might be wrong, so I would appreciate a (polite?) feedback. I do have a
> userspace workaround, but I thought it is worthy to share and to hear your
> opinion, as well as feedback from other UFFD users.
>
> The issue I encountered regards the ordering of UFFD events tbat might not
> reflect the actual order in which events took place.
>
> In more detail, UFFD events (e.g., unmap, fork) are not ordered against
> themselves [*]. The mm-lock is dropped before notifying the userspace
> UFFD-monitor, and therefore there is no guarantee as to whether the order of
> the events actually reflects the order in which the events took place.
> This can prevent a UFFD-monitor from using the events to track which
> ranges are mapped. Specifically, UFFD_EVENT_FORK message and a
> UFFD_EVENT_UNMAP message (which reflects unmap in the parent process) can
> be reordered, if the events are triggered by two different threads. In
> this case the UFFD-monitor cannot figure from the events whether the
> child process has the unmapped memory range still mapped (because fork
> happened first) or not.
Yeah, it seems that something like this is possible:
fork() munmap()
mmap_write_unlock();
mmap_write_lock_killable();
do_things();
mmap_{read,write}_unlock();
userfaultfd_unmap_complete();
dup_userfaultfd_complete();
A solution could be to split uffd_*_complete() to two parts: one that
queues up the event message and the second one that waits for it to be read
by the monitor. The first part then can run befor mm-lock is released.
If you can think of something nicer, it'll be really great!
> Obviously, it does not make sense to keep holding mm-lock while notifying the
> user, as it can even lead to deadlocks. Userspace UFFD-monitors can
> workaround this issue by using seccomp+ptrace instead of UFFD-events to
> obtain order of the events or examine /proc/[pid]/smaps. Yet, this introduces
> overheads, is complicated, and I doubt anyone does so. I wonder if the API is
> reasonable, or whether I am missing something.
>
> Thanks,
> Nadav
>
> [*] Note that I do not discuss UFFD-monitor issued ioctl's, but the order
> between UFFD-events.
>
--
Sincerely yours,
Mike.
