1. Patchew
  2. linux
  3. View series

[PATCH] L2CAP: Fix null-ptr-deref in l2cap_sock_set_shutdown_cb

Sungwoo Kim posted 1 patch 2 years, 7 months ago
There is a newer version of this series
net/bluetooth/l2cap_sock.c | 8 +++++---
1 file changed, 5 insertions(+), 3 deletions(-)
[PATCH] L2CAP: Fix null-ptr-deref in l2cap_sock_set_shutdown_cb
Posted by Sungwoo Kim 2 years, 7 months ago 
The L2CAP socket shutdown invokes l2cap_sock_destruct without a lock
on conn->chan_lock, assigning NULL to chan->data *just before*
the l2cap_disconnect_req thread that accesses to chan->data.
This patch prevent it by adding a null check for a workaround, instead
of fixing a lock.

This bug is found by FuzzBT, a modified Syzkaller by Sungwoo Kim(me).
Ruoyu Wu(wuruoyu@me.com) and Hui Peng(benquike@gmail.com) has helped
the FuzzBT project.

Signed-off-by: Sungwoo Kim <iam@sung-woo.kim>
---
 net/bluetooth/l2cap_sock.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c
index ca8f07f35..350c7afdf 100644
--- a/net/bluetooth/l2cap_sock.c
+++ b/net/bluetooth/l2cap_sock.c
@@ -1681,9 +1681,11 @@ static void l2cap_sock_set_shutdown_cb(struct l2cap_chan *chan)
 {
 	struct sock *sk = chan->data;
 
-	lock_sock(sk);
-	sk->sk_shutdown = SHUTDOWN_MASK;
-	release_sock(sk);
+	if (!sk) {
+		lock_sock(sk);
+		sk->sk_shutdown = SHUTDOWN_MASK;
+		release_sock(sk);
+	}
 }
 
 static long l2cap_sock_get_sndtimeo_cb(struct l2cap_chan *chan)
-- 
2.25.1
Re: [PATCH] L2CAP: Fix null-ptr-deref in l2cap_sock_set_shutdown_cb
Posted by Eric Dumazet 2 years, 7 months ago 
On Thu, Jan 19, 2023 at 2:35 AM Sungwoo Kim <iam@sung-woo.kim> wrote:
>
> The L2CAP socket shutdown invokes l2cap_sock_destruct without a lock
> on conn->chan_lock, assigning NULL to chan->data *just before*
> the l2cap_disconnect_req thread that accesses to chan->data.

This is racy then ?

> This patch prevent it by adding a null check for a workaround, instead
> of fixing a lock.

This would at least require some barriers I think.

What about other _cb helpers also reading/using chan->data ?

>
> This bug is found by FuzzBT, a modified Syzkaller by Sungwoo Kim(me).
> Ruoyu Wu(wuruoyu@me.com) and Hui Peng(benquike@gmail.com) has helped
> the FuzzBT project.
>
> Signed-off-by: Sungwoo Kim <iam@sung-woo.kim>

I would also add

Fixes: 1bff51ea59a9 ("Bluetooth: fix use-after-free error in
lock_sock_nested()")

> ---
>  net/bluetooth/l2cap_sock.c | 8 +++++---
>  1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c
> index ca8f07f35..350c7afdf 100644
> --- a/net/bluetooth/l2cap_sock.c
> +++ b/net/bluetooth/l2cap_sock.c
> @@ -1681,9 +1681,11 @@ static void l2cap_sock_set_shutdown_cb(struct l2cap_chan *chan)
>  {
>         struct sock *sk = chan->data;
>

Other similar fixes simply do:

     if (!sk)
          return;

I would chose to use the same coding style in net/bluetooth/l2cap_sock.c

> -       lock_sock(sk);
> -       sk->sk_shutdown = SHUTDOWN_MASK;
> -       release_sock(sk);
> +       if (!sk) {
> +               lock_sock(sk);
> +               sk->sk_shutdown = SHUTDOWN_MASK;
> +               release_sock(sk);
> +       }
>  }
>
>  static long l2cap_sock_get_sndtimeo_cb(struct l2cap_chan *chan)
> --
> 2.25.1
>
Re: [PATCH] L2CAP: Fix null-ptr-deref in l2cap_sock_set_shutdown_cb
Posted by Luiz Augusto von Dentz 2 years, 7 months ago 
Hi Kim, Eric,

On Wed, Jan 18, 2023 at 8:16 PM Eric Dumazet <edumazet@google.com> wrote:
>
> On Thu, Jan 19, 2023 at 2:35 AM Sungwoo Kim <iam@sung-woo.kim> wrote:
> >
> > The L2CAP socket shutdown invokes l2cap_sock_destruct without a lock
> > on conn->chan_lock, assigning NULL to chan->data *just before*
> > the l2cap_disconnect_req thread that accesses to chan->data.
>
> This is racy then ?
>
> > This patch prevent it by adding a null check for a workaround, instead
> > of fixing a lock.
>
> This would at least require some barriers I think.
>
> What about other _cb helpers also reading/using chan->data ?

Perhaps it would be a good idea to include the stack backtrace so we
can better understand it, at some point we might need to refactor or
locks to avoid circular dependencies.

> >
> > This bug is found by FuzzBT, a modified Syzkaller by Sungwoo Kim(me).
> > Ruoyu Wu(wuruoyu@me.com) and Hui Peng(benquike@gmail.com) has helped
> > the FuzzBT project.
> >
> > Signed-off-by: Sungwoo Kim <iam@sung-woo.kim>
>
> I would also add
>
> Fixes: 1bff51ea59a9 ("Bluetooth: fix use-after-free error in
> lock_sock_nested()")

+1

> > ---
> >  net/bluetooth/l2cap_sock.c | 8 +++++---
> >  1 file changed, 5 insertions(+), 3 deletions(-)
> >
> > diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c
> > index ca8f07f35..350c7afdf 100644
> > --- a/net/bluetooth/l2cap_sock.c
> > +++ b/net/bluetooth/l2cap_sock.c
> > @@ -1681,9 +1681,11 @@ static void l2cap_sock_set_shutdown_cb(struct l2cap_chan *chan)
> >  {
> >         struct sock *sk = chan->data;
> >
>
> Other similar fixes simply do:
>
>      if (!sk)
>           return;
>
> I would chose to use the same coding style in net/bluetooth/l2cap_sock.c

Yep, at least l2cap_sock_close_cb and l2cap_sock_shutdown do that already.

>
> > -       lock_sock(sk);
> > -       sk->sk_shutdown = SHUTDOWN_MASK;
> > -       release_sock(sk);
> > +       if (!sk) {
> > +               lock_sock(sk);
> > +               sk->sk_shutdown = SHUTDOWN_MASK;
> > +               release_sock(sk);
> > +       }
> >  }
> >
> >  static long l2cap_sock_get_sndtimeo_cb(struct l2cap_chan *chan)
> > --
> > 2.25.1
> >



-- 
Luiz Augusto von Dentz
BUG: KASAN: null-ptr-deref in _raw_spin_lock_bh+0x4c/0xc0
Posted by Sungwoo Kim 2 years, 7 months ago 
Write of size 4 at addr 0000000000000098 by task kworker/u3:0/76

CPU: 0 PID: 76 Comm: kworker/u3:0 Not tainted 6.1.0-rc2 #129
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014
Workqueue: hci0 hci_rx_work
Call Trace:
 <TASK>
 dump_stack_lvl+0x7b/0xb3
 print_report+0xed/0x200
 ? __virt_addr_valid+0x5c/0x240
 ? kasan_addr_to_slab+0xd/0xa0
 ? _raw_spin_lock_bh+0x4c/0xc0
 kasan_report+0xd3/0x100
 ? _raw_spin_lock_bh+0x4c/0xc0
 kasan_check_range+0x2d3/0x310
 __kasan_check_write+0x14/0x20
 _raw_spin_lock_bh+0x4c/0xc0
 lock_sock_nested+0x3f/0x160
 ? queue_work_on+0x90/0xd0
 l2cap_sock_set_shutdown_cb+0x3d/0x60
 l2cap_disconnect_req+0x1e3/0x2e0
 l2cap_bredr_sig_cmd+0x3d2/0x5ec0
 ? vprintk_emit+0x29b/0x4d0
 ? vprintk_default+0x2b/0x30
 ? vprintk+0xdc/0x100
 ? _printk+0x67/0x85
 ? bt_err+0x7f/0xc0
 ? bt_err+0x9a/0xc0
 l2cap_recv_frame+0x7bc/0x4e10
 ? _printk+0x67/0x85
 ? bt_err+0x7f/0xc0
 ? __wake_up_klogd+0xc4/0xf0
 l2cap_recv_acldata+0x327/0x650
 ? hci_conn_enter_active_mode+0x1b7/0x1f0
 hci_rx_work+0x6b7/0x7c0
 process_one_work+0x461/0xaf0
 worker_thread+0x5f8/0xba0
 kthread+0x1c5/0x200
 ? process_one_work+0xaf0/0xaf0
 ? kthread_blkcg+0x90/0x90
 ret_from_fork+0x1f/0x30
 </TASK>
Re: BUG: KASAN: null-ptr-deref in _raw_spin_lock_bh+0x4c/0xc0
Posted by Eric Dumazet 2 years, 7 months ago 
On Thu, Jan 19, 2023 at 2:41 AM Sungwoo Kim <iam@sung-woo.kim> wrote:
>
> Write of size 4 at addr 0000000000000098 by task kworker/u3:0/76
>
> CPU: 0 PID: 76 Comm: kworker/u3:0 Not tainted 6.1.0-rc2 #129
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1.1 04/01/2014
> Workqueue: hci0 hci_rx_work
> Call Trace:
>  <TASK>
>  dump_stack_lvl+0x7b/0xb3
>  print_report+0xed/0x200
>  ? __virt_addr_valid+0x5c/0x240
>  ? kasan_addr_to_slab+0xd/0xa0
>  ? _raw_spin_lock_bh+0x4c/0xc0
>  kasan_report+0xd3/0x100
>  ? _raw_spin_lock_bh+0x4c/0xc0
>  kasan_check_range+0x2d3/0x310
>  __kasan_check_write+0x14/0x20
>  _raw_spin_lock_bh+0x4c/0xc0
>  lock_sock_nested+0x3f/0x160
>  ? queue_work_on+0x90/0xd0
>  l2cap_sock_set_shutdown_cb+0x3d/0x60
>  l2cap_disconnect_req+0x1e3/0x2e0
>  l2cap_bredr_sig_cmd+0x3d2/0x5ec0
>  ? vprintk_emit+0x29b/0x4d0
>  ? vprintk_default+0x2b/0x30
>  ? vprintk+0xdc/0x100
>  ? _printk+0x67/0x85
>  ? bt_err+0x7f/0xc0
>  ? bt_err+0x9a/0xc0
>  l2cap_recv_frame+0x7bc/0x4e10
>  ? _printk+0x67/0x85
>  ? bt_err+0x7f/0xc0
>  ? __wake_up_klogd+0xc4/0xf0
>  l2cap_recv_acldata+0x327/0x650
>  ? hci_conn_enter_active_mode+0x1b7/0x1f0
>  hci_rx_work+0x6b7/0x7c0
>  process_one_work+0x461/0xaf0
>  worker_thread+0x5f8/0xba0
>  kthread+0x1c5/0x200
>  ? process_one_work+0xaf0/0xaf0
>  ? kthread_blkcg+0x90/0x90
>  ret_from_fork+0x1f/0x30
>  </TASK>

If you plan sending more stack traces like that in the future, always
run scripts/decode_stacktrace.sh
so that the public report includes symbols.

Thanks.
[PATCH] L2CAP: Fix null-ptr-deref in l2cap_sock_set_shutdown_cb
Posted by Sungwoo Kim 2 years, 7 months ago 
Fix a critical typo on the prev patch - Sorry!

Signed-off-by: Sungwoo Kim <iam@sung-woo.kim>
---
 net/bluetooth/l2cap_sock.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c
index ca8f07f35..b9381d45d 100644
--- a/net/bluetooth/l2cap_sock.c
+++ b/net/bluetooth/l2cap_sock.c
@@ -1681,9 +1681,11 @@ static void l2cap_sock_set_shutdown_cb(struct l2cap_chan *chan)
 {
 	struct sock *sk = chan->data;
 
-	lock_sock(sk);
-	sk->sk_shutdown = SHUTDOWN_MASK;
-	release_sock(sk);
+	if (sk) {
+		lock_sock(sk);
+		sk->sk_shutdown = SHUTDOWN_MASK;
+		release_sock(sk);
+	}
 }
 
 static long l2cap_sock_get_sndtimeo_cb(struct l2cap_chan *chan)
-- 
2.25.1

Patchew 2.1-devel-b67e4c2

© 2016 - 2025 Red Hat, Inc.