On Wed, Dec 6, 2023 at 11:52 AM Hyunwoo Kim [off-list ref] wrote:

Dear,

On Wed, Dec 06, 2023 at 11:33:15AM +0100, Eric Dumazet wrote:

quoted

On Wed, Dec 6, 2023 at 5:13 AM Hyunwoo Kim [off-list ref] wrote:

quoted

Because rose_ioctl() accesses sk->sk_receive_queue
without holding a sk->sk_receive_queue.lock, it can
cause a race with rose_accept().
A use-after-free for skb occurs with the following flow.

rose_ioctl() -> skb_peek()
rose_accept() -> skb_dequeue() -> kfree_skb()

Add sk->sk_receive_queue.lock to rose_ioctl() to fix this issue.

Please add a Fixes: tag

quoted

Signed-off-by: Hyunwoo Kim <redacted>
---
v1 -> v2: Use sk->sk_receive_queue.lock instead of lock_sock.
---
 net/rose/af_rose.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/rose/af_rose.c b/net/rose/af_rose.c
index 0cc5a4e19900..841c238de222 100644
--- a/net/rose/af_rose.c
+++ b/net/rose/af_rose.c

@@ -1316,8 +1316,10 @@ static int rose_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
                struct sk_buff *skb;
                long amount = 0L;
                /* These two are safe on a single CPU system as only user tasks fiddle here */
+               spin_lock(&sk->sk_receive_queue.lock);

You need interrupt safety here.

sk_receive_queue can be fed from interrupt, that would potentially deadlock.

I want to change spin_lock to spin_lock_irqsave, is this okay?

Either spin_lock_irq() or spin_lock_irqsave() will work.

Regards,
Hyunwoo Kim

quoted

quoted

                if ((skb = skb_peek(&sk->sk_receive_queue)) != NULL)
                        amount = skb->len;
+               spin_unlock(&sk->sk_receive_queue.lock);
                return put_user(amount, (unsigned int __user *) argp);
        }

--
2.25.1