This is a note to let you know that I've just added the patch titled
io_uring: fix UAF due to missing POLLFREE handling
to the 5.15-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
io_uring-fix-uaf-due-to-missing-pollfree-handling.patch
and it can be found in the queue-5.15 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
>From foo@baz Thu Sep 1 11:59:31 AM CEST 2022
From: Pavel Begunkov <asml.silence@xxxxxxxxx>
Date: Mon, 29 Aug 2022 14:30:24 +0100
Subject: io_uring: fix UAF due to missing POLLFREE handling
To: stable@xxxxxxxxxxxxxxx
Cc: Jens Axboe <axboe@xxxxxxxxx>, asml.silence@xxxxxxxxx, Eric Biggers <ebiggers@xxxxxxxxxx>, syzbot+5426c7ed6868c705ca14@xxxxxxxxxxxxxxxxxxxxxxxxx
Message-ID: <23cf800d9e5e1ec5d7d3776cf57b4db87e5758dc.1661594698.git.asml.silence@xxxxxxxxx>
From: Pavel Begunkov <asml.silence@xxxxxxxxx>
[ upstream commmit 791f3465c4afde02d7f16cf7424ca87070b69396 ]
Fixes a problem described in 50252e4b5e989
("aio: fix use-after-free due to missing POLLFREE handling")
and copies the approach used there.
In short, we have to forcibly eject a poll entry when we meet POLLFREE.
We can't rely on io_poll_get_ownership() as can't wait for potentially
running tw handlers, so we use the fact that wqs are RCU freed. See
Eric's patch and comments for more details.
Reported-by: Eric Biggers <ebiggers@xxxxxxxxxx>
Link: https://lore.kernel.org/r/20211209010455.42744-6-ebiggers@xxxxxxxxxx
Reported-and-tested-by: syzbot+5426c7ed6868c705ca14@xxxxxxxxxxxxxxxxxxxxxxxxx
Fixes: 221c5eb233823 ("io_uring: add support for IORING_OP_POLL")
Signed-off-by: Pavel Begunkov <asml.silence@xxxxxxxxx>
Link: https://lore.kernel.org/r/4ed56b6f548f7ea337603a82315750449412748a.1642161259.git.asml.silence@xxxxxxxxx
[axboe: drop non-functional change from patch]
Signed-off-by: Jens Axboe <axboe@xxxxxxxxx>
[pavel: backport]
Signed-off-by: Pavel Begunkov <asml.silence@xxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
fs/io_uring.c | 58 ++++++++++++++++++++++++++++++++++++++++++++++++++--------
1 file changed, 50 insertions(+), 8 deletions(-)
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -5369,12 +5369,14 @@ static void io_init_poll_iocb(struct io_
static inline void io_poll_remove_entry(struct io_poll_iocb *poll)
{
- struct wait_queue_head *head = poll->head;
+ struct wait_queue_head *head = smp_load_acquire(&poll->head);
- spin_lock_irq(&head->lock);
- list_del_init(&poll->wait.entry);
- poll->head = NULL;
- spin_unlock_irq(&head->lock);
+ if (head) {
+ spin_lock_irq(&head->lock);
+ list_del_init(&poll->wait.entry);
+ poll->head = NULL;
+ spin_unlock_irq(&head->lock);
+ }
}
static void io_poll_remove_entries(struct io_kiocb *req)
@@ -5382,10 +5384,26 @@ static void io_poll_remove_entries(struc
struct io_poll_iocb *poll = io_poll_get_single(req);
struct io_poll_iocb *poll_double = io_poll_get_double(req);
- if (poll->head)
- io_poll_remove_entry(poll);
- if (poll_double && poll_double->head)
+ /*
+ * While we hold the waitqueue lock and the waitqueue is nonempty,
+ * wake_up_pollfree() will wait for us. However, taking the waitqueue
+ * lock in the first place can race with the waitqueue being freed.
+ *
+ * We solve this as eventpoll does: by taking advantage of the fact that
+ * all users of wake_up_pollfree() will RCU-delay the actual free. If
+ * we enter rcu_read_lock() and see that the pointer to the queue is
+ * non-NULL, we can then lock it without the memory being freed out from
+ * under us.
+ *
+ * Keep holding rcu_read_lock() as long as we hold the queue lock, in
+ * case the caller deletes the entry from the queue, leaving it empty.
+ * In that case, only RCU prevents the queue memory from being freed.
+ */
+ rcu_read_lock();
+ io_poll_remove_entry(poll);
+ if (poll_double)
io_poll_remove_entry(poll_double);
+ rcu_read_unlock();
}
/*
@@ -5523,6 +5541,30 @@ static int io_poll_wake(struct wait_queu
wait);
__poll_t mask = key_to_poll(key);
+ if (unlikely(mask & POLLFREE)) {
+ io_poll_mark_cancelled(req);
+ /* we have to kick tw in case it's not already */
+ io_poll_execute(req, 0);
+
+ /*
+ * If the waitqueue is being freed early but someone is already
+ * holds ownership over it, we have to tear down the request as
+ * best we can. That means immediately removing the request from
+ * its waitqueue and preventing all further accesses to the
+ * waitqueue via the request.
+ */
+ list_del_init(&poll->wait.entry);
+
+ /*
+ * Careful: this *must* be the last step, since as soon
+ * as req->head is NULL'ed out, the request can be
+ * completed and freed, since aio_poll_complete_work()
+ * will no longer need to take the waitqueue lock.
+ */
+ smp_store_release(&poll->head, NULL);
+ return 1;
+ }
+
/* for instances that support it check for an event match first */
if (mask && !(mask & poll->events))
return 0;
Patches currently in stable-queue which might be from asml.silence@xxxxxxxxx are
queue-5.15/io_uring-fix-wrong-arm_poll-error-handling.patch
queue-5.15/io_uring-poll-rework.patch
queue-5.15/io_uring-remove-unused-function-req_ref_put.patch
queue-5.15/io_uring-fail-links-when-poll-fails.patch
queue-5.15/io_uring-inline-io_poll_complete.patch
queue-5.15/io_uring-bump-poll-refs-to-full-31-bits.patch
queue-5.15/io_uring-clean-cqe-filling-functions.patch
queue-5.15/io_uring-fix-uaf-due-to-missing-pollfree-handling.patch
queue-5.15/io_uring-correct-fill-events-helpers-types.patch
queue-5.15/io_uring-refactor-poll-update.patch
queue-5.15/io_uring-kill-poll-linking-optimisation.patch
queue-5.15/io_uring-remove-poll-entry-from-list-when-canceling-all.patch
queue-5.15/io_uring-move-common-poll-bits.patch
