Bug 1523481 (CVE-2017-15126)

A flaw was found in the Linux kernels handling of fork failure when dealing with event messages in the userfaultfd code.  Failure to fork correctly can lead to a situation where a fork event will be removed from an already freed list of events with  userfaultfd_ctx_put().  The userfault events are part of the slab cache, and the secondary free could only free other event list objects as the free mechanism uses the underlying kmem_cache_* implementation making generic use-after-free style attacks less effective.  

While technically a use after free style flaw, at this time Red Hat Product security believes this flaw to be difficult to exploit but this may change in the future.


From the commit:

When reading the event from the uffd, we put it on a temporary
fork_event list to detect if we can still access it after releasing and
retaking the event_wqh.lock.

If fork aborts and removes the event from the fork_event all is fine as
long as we're still in the userfault read context and fork_event head is
still alive.

We've [forgotten] to put the event allocated in the fork kernel stack, back from
fork_event list-head to the event_wqh head, before returning from
userfaultfd_ctx_read, because the fork_event head lifetime is limited to
the userfaultfd_ctx_read stack lifetime.

Forgetting to move the event back to its event_wqh place then results in
__remove_wait_queue(&ctx->event_wqh, &ewq->wq); in
userfaultfd_event_wait_completion to remove it from a head that has been
already freed from the reader stack.

This could only happen if resolve_userfault_fork failed (for example if
there are no file descriptors available to allocate the fork uffd).  If
it succeeded it was put back correctly.

Furthermore, after find_userfault_evt receives a fork event, the forked
userfault context in fork_nctx and uwq->msg.arg.reserved.reserved1 can
be released by the fork thread as soon as the event_wqh.lock is
released.  Taking a reference on the fork_nctx before dropping the lock
prevents an use after free in resolve_userfault_fork().

If the fork side aborted and it already released everything, we still
try to succeed resolve_userfault_fork(), if possible.
-- end ---

References:

https://marc.info/?t=150351212000001&r=1&w=2

An upstream patch:

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=384632e67e0829deb8015ee6ad916b180049d252