Bluetooth: L2CAP: handle NULL sock pointer in l2cap_sock_alloc
authorFedor Pchelkin <pchelkin@ispras.ru>
Tue, 17 Dec 2024 21:19:59 +0000 (00:19 +0300)
committerBen Hutchings <benh@debian.org>
Mon, 24 Feb 2025 01:11:58 +0000 (02:11 +0100)
commit1c45e0e88ece59892c53f8e06978f29a6ae5c2e1
tree39caa7fc0cb1dfe21f3aef9bde1471737224c04c
parentfb37780ed99b1eeaa5e71845edbe8121365e316b
Bluetooth: L2CAP: handle NULL sock pointer in l2cap_sock_alloc

commit 5f397409f8ee5bc82901eeaf799e1cbc4f8edcf1 upstream.

A NULL sock pointer is passed into l2cap_sock_alloc() when it is called
from l2cap_sock_new_connection_cb() and the error handling paths should
also be aware of it.

Seemingly a more elegant solution would be to swap bt_sock_alloc() and
l2cap_chan_create() calls since they are not interdependent to that moment
but then l2cap_chan_create() adds the soon to be deallocated and still
dummy-initialized channel to the global list accessible by many L2CAP
paths. The channel would be removed from the list in short period of time
but be a bit more straight-forward here and just check for NULL instead of
changing the order of function calls.

Found by Linux Verification Center (linuxtesting.org) with SVACE static
analysis tool.

Fixes: 7c4f78cdb8e7 ("Bluetooth: L2CAP: do not leave dangling sk pointer on error in l2cap_sock_create()")
Cc: stable@vger.kernel.org
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Gbp-Pq: Topic bugfix/all
Gbp-Pq: Name bluetooth-l2cap-handle-null-sock-pointer-in-l2cap_sock_alloc.patch
net/bluetooth/l2cap_sock.c