Re: [PATCH] lockdep: annotate sk_lock nesting in AF_BLUETOOTH

[Marcel Holtmann <marcel@holtmann.org>]

Hi Peter,

> =============================================
> [ INFO: possible recursive locking detected ]
> 2.6.18-1.2726.fc6 #1
> ---------------------------------------------
> hidd/2271 is trying to acquire lock:
>  (sk_lock-AF_BLUETOOTH){--..}, at: [<f8d16241>] bt_accept_dequeue+0x26/0xc6
> [bluetooth]
> 
> but task is already holding lock:
>  (sk_lock-AF_BLUETOOTH){--..}, at: [<f8bce088>] l2cap_sock_accept+0x41/0x11e [l2cap]
> 
> other info that might help us debug this:
> 1 lock held by hidd/2271:
>  #0:  (sk_lock-AF_BLUETOOTH){--..}, at: [<f8bce088>]
> l2cap_sock_accept+0x41/0x11e [l2cap]
> 
> stack backtrace:
>  [<c04051ed>] show_trace_log_lvl+0x58/0x16a
>  [<c04057fa>] show_trace+0xd/0x10
>  [<c0405913>] dump_stack+0x19/0x1b
>  [<c043b7dc>] __lock_acquire+0x6ea/0x90d
>  [<c043bf70>] lock_acquire+0x4b/0x6b
>  [<c05b203b>] lock_sock+0xac/0xbc
>  [<f8d16241>] bt_accept_dequeue+0x26/0xc6 [bluetooth]
>  [<f8bce129>] l2cap_sock_accept+0xe2/0x11e [l2cap]
>  [<c05b142e>] sys_accept+0xd8/0x179
>  [<c05b1576>] sys_socketcall+0xa7/0x186
>  [<c0403fb7>] syscall_call+0x7/0xb
> 
> classical case of nesting; bt_accept_dequeue() locks the children of the object
> locked by l2cap_sock_accept().
> 
> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
> ---
>  include/net/sock.h    |   13 ++++++++++++-
>  net/bluetooth/l2cap.c |    2 +-
>  net/core/sock.c       |    6 +++---
>  3 files changed, 16 insertions(+), 5 deletions(-)
> 
> Index: linux-2.6/include/net/sock.h
> ===================================================================
> --- linux-2.6.orig/include/net/sock.h
> +++ linux-2.6/include/net/sock.h
> @@ -745,7 +745,18 @@ static inline int sk_stream_wmem_schedul
>   */
>  #define sock_owned_by_user(sk)	((sk)->sk_lock.owner)
>  
> -extern void FASTCALL(lock_sock(struct sock *sk));
> +extern void FASTCALL(_lock_sock(struct sock *sk, int subclass));
> +
> +static inline void lock_sock(struct sock *sk)
> +{
> +	_lock_sock(sk, 0);
> +}
> +
> +static inline void lock_sock_nested(struct sock *sk, int subclass)
> +{
> +	_lock_sock(sk, subclass);
> +}
> +
>  extern void FASTCALL(release_sock(struct sock *sk));
>  
>  /* BH context may only use the following locking interface. */
> Index: linux-2.6/net/bluetooth/l2cap.c
> ===================================================================
> --- linux-2.6.orig/net/bluetooth/l2cap.c
> +++ linux-2.6/net/bluetooth/l2cap.c
> @@ -770,7 +770,7 @@ static int l2cap_sock_accept(struct sock
>  	long timeo;
>  	int err = 0;
>  
> -	lock_sock(sk);
> +	lock_sock_nested(sk, SINGLE_DEPTH_NESTING);
>  
>  	if (sk->sk_state != BT_LISTEN) {
>  		err = -EBADFD;
> Index: linux-2.6/net/core/sock.c
> ===================================================================
> --- linux-2.6.orig/net/core/sock.c
> +++ linux-2.6/net/core/sock.c
> @@ -1527,7 +1527,7 @@ void sock_init_data(struct socket *sock,
>  	atomic_set(&sk->sk_refcnt, 1);
>  }
>  
> -void fastcall lock_sock(struct sock *sk)
> +void fastcall _lock_sock(struct sock *sk, int subclass)
>  {
>  	might_sleep();
>  	spin_lock_bh(&sk->sk_lock.slock);
> @@ -1538,11 +1538,11 @@ void fastcall lock_sock(struct sock *sk)
>  	/*
>  	 * The sk_lock has mutex_lock() semantics here:
>  	 */
> -	mutex_acquire(&sk->sk_lock.dep_map, 0, 0, _RET_IP_);
> +	mutex_acquire(&sk->sk_lock.dep_map, subclass, 0, _RET_IP_);
>  	local_bh_enable();
>  }
>  
> -EXPORT_SYMBOL(lock_sock);
> +EXPORT_SYMBOL(_lock_sock);
>  
>  void fastcall release_sock(struct sock *sk)
>  {

for the Bluetooth subsystem part of this patch:

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>

Regards

Marcel