Message-ID: <20210510102950.12551-1-varad.gautam@suse.com>
Date:   Mon, 10 May 2021 12:29:46 +0200
From:   Varad Gautam <varad.gautam@...e.com>
To:     linux-kernel@...r.kernel.org
CC:     varad.gautam@...e.com,
        Matthias von Faber <matthias.vonfaber@...-tech.de>,
        Christian Brauner <christian.brauner@...ntu.com>,
        Oleg Nesterov <oleg@...hat.com>,
        "Eric W. Biederman" <ebiederm@...ssion.com>,
        Manfred Spraul <manfred@...orfullife.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Davidlohr Bueso <dbueso@...e.de>,
        James Morris <jamorris@...ux.microsoft.com>,
        Serge Hallyn <serge@...lyn.com>,
        Jules Irenge <jbi.octave@...il.com>,
        "Gustavo A. R. Silva" <gustavoars@...nel.org>,
        Alexey Dobriyan <adobriyan@...il.com>
Subject: [PATCH v4] ipc/mqueue, msg, sem: Avoid relying on a stack reference past its expiry

do_mq_timedreceive calls wq_sleep with a stack local address. The
sender (do_mq_timedsend) uses this address to later call
pipelined_send.

This leads to a very hard to trigger race where a do_mq_timedreceive call
might return and leave do_mq_timedsend to rely on an invalid address,
causing the following crash:

[  240.739977] RIP: 0010:wake_q_add_safe+0x13/0x60
[  240.739991] Call Trace:
[  240.739999]  __x64_sys_mq_timedsend+0x2a9/0x490
[  240.740003]  ? auditd_test_task+0x38/0x40
[  240.740007]  ? auditd_test_task+0x38/0x40
[  240.740011]  do_syscall_64+0x80/0x680
[  240.740017]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
[  240.740019] RIP: 0033:0x7f5928e40343

The race occurs as:

1. do_mq_timedreceive calls wq_sleep with the address of
`struct ext_wait_queue` on function stack (aliased as `ewq_addr` here)
- it holds a valid `struct ext_wait_queue *` as long as the stack has
not been overwritten.

2. `ewq_addr` gets added to info->e_wait_q[RECV].list in wq_add, and
do_mq_timedsend receives it via wq_get_first_waiter(info, RECV) to call
__pipelined_op.

3. Sender calls __pipelined_op::smp_store_release(&this->state, STATE_READY).
Here is where the race window begins. (`this` is `ewq_addr`.)

4. If the receiver wakes up now in do_mq_timedreceive::wq_sleep, it
will see `state == STATE_READY` and break.

5. do_mq_timedreceive returns, and `ewq_addr` is no longer guaranteed
to be a `struct ext_wait_queue *` since it was on do_mq_timedreceive's
stack. (Although the address may not get overwritten until another
function happens to touch it, which means it can persist around for an
indefinite time.)

6. do_mq_timedsend::__pipelined_op() still believes `ewq_addr` is a
`struct ext_wait_queue *`, and uses it to find a task_struct to pass
to the wake_q_add_safe call. In the lucky case where nothing has
overwritten `ewq_addr` yet, `ewq_addr->task` is the right task_struct.
In the unlucky case, __pipelined_op::wake_q_add_safe gets handed a
bogus address as the receiver's task_struct causing the crash.

do_mq_timedsend::__pipelined_op() should not dereference `this` after
setting STATE_READY, as the receiver counterpart is now free to return.
Change __pipelined_op to call wake_q_add_safe on the receiver's
task_struct returned by get_task_struct, instead of dereferencing
`this` which sits on the receiver's stack.

As Manfred pointed out, the race potentially also exists in
ipc/msg.c::expunge_all and ipc/sem.c::wake_up_sem_queue_prepare. Fix
those in the same way.

Fixes: c5b2cbdbdac563 ("ipc/mqueue.c: update/document memory barriers")
Fixes: 8116b54e7e23ef ("ipc/sem.c: document and update memory barriers")
Fixes: 0d97a82ba830d8 ("ipc/msg.c: update and document memory barriers")
Signed-off-by: Varad Gautam <varad.gautam@...e.com>
Reported-by: Matthias von Faber <matthias.vonfaber@...-tech.de>
Cc: Christian Brauner <christian.brauner@...ntu.com>
Cc: Oleg Nesterov <oleg@...hat.com>
Cc: "Eric W. Biederman" <ebiederm@...ssion.com>
Cc: Manfred Spraul <manfred@...orfullife.com>
Cc: Andrew Morton <akpm@...ux-foundation.org>
Cc: Davidlohr Bueso <dbueso@...e.de>
Cc: Manfred Spraul <manfred@...orfullife.com>
---
v2: Call wake_q_add before smp_store_release, instead of using a
    get_task_struct/wake_q_add_safe combination across
    smp_store_release. (Davidlohr Bueso)
v3: Comment/commit message fixup.
v4: - v2 / v3 have potential for introducing lost wakeups. Return to v1
    as the path-of-least-surprises to fix the race at hand.
    - Also fix ipc/msg.c and ipc/sem.c which have the same usage
    pattern.(Manfred Spraul)

 ipc/mqueue.c | 6 ++++--
 ipc/msg.c    | 6 ++++--
 ipc/sem.c    | 6 ++++--
 3 files changed, 12 insertions(+), 6 deletions(-)

diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 8031464ed4ae..4e4e61111500 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -1004,12 +1004,14 @@ static inline void __pipelined_op(struct wake_q_head *wake_q,
 				  struct mqueue_inode_info *info,
 				  struct ext_wait_queue *this)
 {
+	struct task_struct *task;
+
 	list_del(&this->list);
-	get_task_struct(this->task);
+	task = get_task_struct(this->task);
 
 	/* see MQ_BARRIER for purpose/pairing */
 	smp_store_release(&this->state, STATE_READY);
-	wake_q_add_safe(wake_q, this->task);
+	wake_q_add_safe(wake_q, task);
 }
 
 /* pipelined_send() - send a message directly to the task waiting in
diff --git a/ipc/msg.c b/ipc/msg.c
index acd1bc7af55a..6e6c8e0c9380 100644
--- a/ipc/msg.c
+++ b/ipc/msg.c
@@ -251,11 +251,13 @@ static void expunge_all(struct msg_queue *msq, int res,
 	struct msg_receiver *msr, *t;
 
 	list_for_each_entry_safe(msr, t, &msq->q_receivers, r_list) {
-		get_task_struct(msr->r_tsk);
+		struct task_struct *r_tsk;
+
+		r_tsk = get_task_struct(msr->r_tsk);
 
 		/* see MSG_BARRIER for purpose/pairing */
 		smp_store_release(&msr->r_msg, ERR_PTR(res));
-		wake_q_add_safe(wake_q, msr->r_tsk);
+		wake_q_add_safe(wake_q, r_tsk);
 	}
 }
 
diff --git a/ipc/sem.c b/ipc/sem.c
index f6c30a85dadf..7d9c06b0ad6e 100644
--- a/ipc/sem.c
+++ b/ipc/sem.c
@@ -784,12 +784,14 @@ static int perform_atomic_semop(struct sem_array *sma, struct sem_queue *q)
 static inline void wake_up_sem_queue_prepare(struct sem_queue *q, int error,
 					     struct wake_q_head *wake_q)
 {
-	get_task_struct(q->sleeper);
+	struct task_struct *sleeper;
+
+	sleeper = get_task_struct(q->sleeper);
 
 	/* see SEM_BARRIER_2 for purpuse/pairing */
 	smp_store_release(&q->status, error);
 
-	wake_q_add_safe(wake_q, q->sleeper);
+	wake_q_add_safe(wake_q, sleeper);
 }
 
 static void unlink_queue(struct sem_array *sma, struct sem_queue *q)
-- 
2.30.2

Powered by blists - more mailing lists