Re: [RESEND 1/1] Bluetooth: Use chan_list_lock to protect the whole put/destroy invokation

[Lee Jones <lee.jones@linaro.org>]

On Tue, 28 Jun 2022, Luiz Augusto von Dentz wrote:

> Hi Eric, Lee,
> 
> On Mon, Jun 27, 2022 at 4:39 PM Luiz Augusto von Dentz
> <luiz.dentz@gmail.com> wrote:
> >
> > Hi Eric, Lee,
> >
> > On Mon, Jun 27, 2022 at 7:41 AM Eric Dumazet <edumazet@google.com> wrote:
> > >
> > > On Wed, Jun 22, 2022 at 10:27 AM Lee Jones <lee.jones@linaro.org> wrote:
> > > >
> > > > This change prevents a use-after-free caused by one of the worker
> > > > threads starting up (see below) *after* the final channel reference
> > > > has been put() during sock_close() but *before* the references to the
> > > > channel have been destroyed.
> > > >
> > > >   refcount_t: increment on 0; use-after-free.
> > > >   BUG: KASAN: use-after-free in refcount_dec_and_test+0x20/0xd0
> > > >   Read of size 4 at addr ffffffc114f5bf18 by task kworker/u17:14/705
> > > >
> > > >   CPU: 4 PID: 705 Comm: kworker/u17:14 Tainted: G S      W       4.14.234-00003-g1fb6d0bd49a4-dirty #28
> > > >   Hardware name: Qualcomm Technologies, Inc. SM8150 V2 PM8150 Google Inc. MSM sm8150 Flame DVT (DT)
> > > >   Workqueue: hci0 hci_rx_work
> > > >   Call trace:
> > > >    dump_backtrace+0x0/0x378
> > > >    show_stack+0x20/0x2c
> > > >    dump_stack+0x124/0x148
> > > >    print_address_description+0x80/0x2e8
> > > >    __kasan_report+0x168/0x188
> > > >    kasan_report+0x10/0x18
> > > >    __asan_load4+0x84/0x8c
> > > >    refcount_dec_and_test+0x20/0xd0
> > > >    l2cap_chan_put+0x48/0x12c
> > > >    l2cap_recv_frame+0x4770/0x6550
> > > >    l2cap_recv_acldata+0x44c/0x7a4
> > > >    hci_acldata_packet+0x100/0x188
> > > >    hci_rx_work+0x178/0x23c
> > > >    process_one_work+0x35c/0x95c
> > > >    worker_thread+0x4cc/0x960
> > > >    kthread+0x1a8/0x1c4
> > > >    ret_from_fork+0x10/0x18
> > > >
> > > > Cc: stable@kernel.org
> > >
> > > When was the bug added ? (Fixes: tag please)
> > >
> > > > Cc: Marcel Holtmann <marcel@holtmann.org>
> > > > Cc: Johan Hedberg <johan.hedberg@gmail.com>
> > > > Cc: Luiz Augusto von Dentz <luiz.dentz@gmail.com>
> > > > Cc: "David S. Miller" <davem@davemloft.net>
> > > > Cc: Eric Dumazet <edumazet@google.com>
> > > > Cc: Jakub Kicinski <kuba@kernel.org>
> > > > Cc: Paolo Abeni <pabeni@redhat.com>
> > > > Cc: linux-bluetooth@vger.kernel.org
> > > > Cc: netdev@vger.kernel.org
> > > > Signed-off-by: Lee Jones <lee.jones@linaro.org>
> > > > ---
> > > >  net/bluetooth/l2cap_core.c | 4 ++--
> > > >  1 file changed, 2 insertions(+), 2 deletions(-)
> > > >
> > > > diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
> > > > index ae78490ecd3d4..82279c5919fd8 100644
> > > > --- a/net/bluetooth/l2cap_core.c
> > > > +++ b/net/bluetooth/l2cap_core.c
> > > > @@ -483,9 +483,7 @@ static void l2cap_chan_destroy(struct kref *kref)
> > > >
> > > >         BT_DBG("chan %p", chan);
> > > >
> > > > -       write_lock(&chan_list_lock);
> > > >         list_del(&chan->global_l);
> > > > -       write_unlock(&chan_list_lock);
> > > >
> > > >         kfree(chan);
> > > >  }
> > > > @@ -501,7 +499,9 @@ void l2cap_chan_put(struct l2cap_chan *c)
> > > >  {
> > > >         BT_DBG("chan %p orig refcnt %u", c, kref_read(&c->kref));
> > > >
> > > > +       write_lock(&chan_list_lock);
> > > >         kref_put(&c->kref, l2cap_chan_destroy);
> > > > +       write_unlock(&chan_list_lock);
> > > >  }
> > > >  EXPORT_SYMBOL_GPL(l2cap_chan_put);
> > > >
> > > >
> > >
> > > I do not think this patch is correct.
> > >
> > > a kref does not need to be protected by a write lock.
> > >
> > > This might shuffle things enough to work around a particular repro you have.
> > >
> > > If the patch was correct why not protect kref_get() sides ?
> > >
> > > Before the &hdev->rx_work is scheduled (queue_work(hdev->workqueue,
> > > &hdev->rx_work),
> > > a reference must be taken.
> > >
> > > Then this reference must be released at the end of hci_rx_work() or
> > > when hdev->workqueue
> > > is canceled.
> > >
> > > This refcount is not needed _if_ the workqueue is properly canceled at
> > > device dismantle,
> > > in a synchronous way.
> > >
> > > I do not see this hdev->rx_work being canceled, maybe this is the real issue.
> > >
> > > There is a call to drain_workqueue() but this is not enough I think,
> > > because hci_recv_frame()
> > > can re-arm
> > >    queue_work(hdev->workqueue, &hdev->rx_work);
> >
> > I suspect this likely a refcount problem, we do l2cap_get_chan_by_scid:
> >
> > /* Find channel with given SCID.
> >  * Returns locked channel. */
> > static struct l2cap_chan *l2cap_get_chan_by_scid(struct l2cap_conn
> > *conn, u16 cid)
> >
> > So we return a locked channel but that doesn't prevent another thread
> > to call l2cap_chan_put which doesn't care about l2cap_chan_lock so
> > perhaps we actually need to host a reference while we have the lock,
> > at least we do something like that on l2cap_sock.c:
> >
> > l2cap_chan_hold(chan);
> > l2cap_chan_lock(chan);
> >
> > __clear_chan_timer(chan);
> > l2cap_chan_close(chan, ECONNRESET);
> > l2cap_sock_kill(sk);
> >
> > l2cap_chan_unlock(chan);
> > l2cap_chan_put(chan);
> 
> Perhaps something like this:

I'm struggling to apply this for test:

  "error: corrupt patch at line 6"

Please could you attach a `git format-patch`ed patch instead?

> diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c
> index 09ecaf556de5..9050b6af3577 100644
> --- a/net/bluetooth/l2cap_core.c
> +++ b/net/bluetooth/l2cap_core.c
> @@ -111,7 +111,7 @@ static struct l2cap_chan
> *__l2cap_get_chan_by_scid(struct l2cap_conn *conn,
>  }
> 
>  /* Find channel with given SCID.
> - * Returns locked channel. */
> + * Returns a reference locked channel. */
>  static struct l2cap_chan *l2cap_get_chan_by_scid(struct l2cap_conn *conn,
>                                                  u16 cid)
>  {
> @@ -119,15 +119,17 @@ static struct l2cap_chan
> *l2cap_get_chan_by_scid(struct l2cap_conn *conn,
> 
>         mutex_lock(&conn->chan_lock);
>         c = __l2cap_get_chan_by_scid(conn, cid);
> -       if (c)
> +       if (c) {
> +               l2cap_chan_hold(c);
>                 l2cap_chan_lock(c);
> +       }
>         mutex_unlock(&conn->chan_lock);
> 
>         return c;
>  }
> 
>  /* Find channel with given DCID.
> - * Returns locked channel.
> + * Returns a reference locked channel.
>   */
>  static struct l2cap_chan *l2cap_get_chan_by_dcid(struct l2cap_conn *conn,
>                                                  u16 cid)
> @@ -136,8 +138,10 @@ static struct l2cap_chan
> *l2cap_get_chan_by_dcid(struct l2cap_conn *conn,
> 
>         mutex_lock(&conn->chan_lock);
>         c = __l2cap_get_chan_by_dcid(conn, cid);
> -       if (c)
> +       if (c) {
> +               l2cap_chan_hold(c);
>                 l2cap_chan_lock(c);
> +       }
>         mutex_unlock(&conn->chan_lock);
> 
>         return c;
> @@ -4464,6 +4468,7 @@ static inline int l2cap_config_req(struct
> l2cap_conn *conn,
> 
>  unlock:
>         l2cap_chan_unlock(chan);
> +       l2cap_chan_put(chan);
>         return err;
>  }
> 
> @@ -4578,6 +4583,7 @@ static inline int l2cap_config_rsp(struct
> l2cap_conn *conn,
> 
>  done:
>         l2cap_chan_unlock(chan);
> +       l2cap_chan_put(chan);
>         return err;
>  }
> 
> @@ -5305,6 +5311,7 @@ static inline int l2cap_move_channel_req(struct
> l2cap_conn *conn,
>         l2cap_send_move_chan_rsp(chan, result);
> 
>         l2cap_chan_unlock(chan);
> +       l2cap_chan_put(chan);
> 
>         return 0;
>  }
> @@ -5397,6 +5404,7 @@ static void l2cap_move_continue(struct
> l2cap_conn *conn, u16 icid, u16 result)
>         }
> 
>         l2cap_chan_unlock(chan);
> +       l2cap_chan_put(chan);
>  }
> 
>  static void l2cap_move_fail(struct l2cap_conn *conn, u8 ident, u16 icid,
> @@ -5489,6 +5497,7 @@ static int l2cap_move_channel_confirm(struct
> l2cap_conn *conn,
>         l2cap_send_move_chan_cfm_rsp(conn, cmd->ident, icid);
> 
>         l2cap_chan_unlock(chan);
> +       l2cap_chan_put(chan);
> 
>         return 0;
>  }
> @@ -5524,6 +5533,7 @@ static inline int
> l2cap_move_channel_confirm_rsp(struct l2cap_conn *conn,
>         }
> 
>         l2cap_chan_unlock(chan);
> +       l2cap_chan_put(chan);
> 
>         return 0;
>  }
> @@ -5896,12 +5906,11 @@ static inline int l2cap_le_credits(struct
> l2cap_conn *conn,
>         if (credits > max_credits) {
>                 BT_ERR("LE credits overflow");
>                 l2cap_send_disconn_req(chan, ECONNRESET);
> -               l2cap_chan_unlock(chan);
> 
>                 /* Return 0 so that we don't trigger an unnecessary
>                  * command reject packet.
>                  */
> -               return 0;
> +               goto unlock;
>         }
> 
>         chan->tx_credits += credits;
> @@ -5912,7 +5921,9 @@ static inline int l2cap_le_credits(struct
> l2cap_conn *conn,
>         if (chan->tx_credits)
>                 chan->ops->resume(chan);
> 
> +unlock:
>         l2cap_chan_unlock(chan);
> +       l2cap_chan_put(chan);
> 
>         return 0;
>  }
> @@ -7598,6 +7609,7 @@ static void l2cap_data_channel(struct l2cap_conn
> *conn, u16 cid,
> 
>  done:
>         l2cap_chan_unlock(chan);
> +       l2cap_chan_put(chan);
>  }
> 
>  static void l2cap_conless_channel(struct l2cap_conn *conn, __le16 psm,
> 
> 
> >
> 
> 
> 

-- 
Lee Jones [李琼斯]
Principal Technical Lead - Developer Services
Linaro.org │ Open source software for Arm SoCs
Follow Linaro: Facebook | Twitter | Blog