2.6.19-rc5: lockdep warnings starting irattach

2.6.19-rc5: lockdep warnings starting irattach

[Andrey Borzenkov]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

I experimented with SyncCE; after starting IrDA I got this:

Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing enabled
serial8250: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
00:09: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
ip_tables: (C) 2000-2006 Netfilter Core Team

=============================================
[ INFO: possible recursive locking detected ]
2.6.19-rc5-2avb #2
- ---------------------------------------------
pppd/26425 is trying to acquire lock:
 (&hashbin->hb_spinlock){....}, at: [<dfdea87a>] irlmp_slsap_inuse+0x5a/0x170 
[irda]

but task is already holding lock:
 (&hashbin->hb_spinlock){....}, at: [<dfdea857>] irlmp_slsap_inuse+0x37/0x170 
[irda]

other info that might help us debug this:
1 lock held by pppd/26425:
 #0:  (&hashbin->hb_spinlock){....}, at: [<dfdea857>] 
irlmp_slsap_inuse+0x37/0x170 [irda]

stack backtrace:
 [<c010413c>] dump_trace+0x1cc/0x200
 [<c010418a>] show_trace_log_lvl+0x1a/0x30
 [<c01047f2>] show_trace+0x12/0x20
 [<c01048c9>] dump_stack+0x19/0x20
 [<c01346ca>] __lock_acquire+0x8fa/0xc20
 [<c0134d2d>] lock_acquire+0x5d/0x80
 [<c02a851c>] _spin_lock+0x2c/0x40
 [<dfdea87a>] irlmp_slsap_inuse+0x5a/0x170 [irda]
 [<dfdebab2>] irlmp_open_lsap+0x62/0x180 [irda]
 [<dfdf35d1>] irttp_open_tsap+0x181/0x230 [irda]
 [<dfdc0c3d>] ircomm_open_tsap+0x5d/0xa0 [ircomm]
 [<dfdc05d8>] ircomm_open+0xb8/0xd0 [ircomm]
 [<dfdd0477>] ircomm_tty_open+0x4f7/0x570 [ircomm_tty]
 [<c020bbe4>] tty_open+0x174/0x340
 [<c016bd69>] chrdev_open+0x89/0x170
 [<c0167bd6>] __dentry_open+0xa6/0x1d0
 [<c0167da5>] nameidata_to_filp+0x35/0x40
 [<c0167df9>] do_filp_open+0x49/0x50
 [<c0167e47>] do_sys_open+0x47/0xd0
 [<c0167f0c>] sys_open+0x1c/0x20
 [<c010307d>] sysenter_past_esp+0x56/0x8d
 [<b7f86410>] 0xb7f86410
 =======================

This apparently happens when irattach is run. Interface itself (irda0) is 
created as far as I can tell.

Please let me know if additional information is required.

- -andrey

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (GNU/Linux)

iD8DBQFFXwbDR6LMutpd94wRAiUpAJwIK5hSusOwFUloh0jXOb5hk5iVwgCfX7Iq
c5kRWItPWR9WGw4jORPd21k=
=oax8
-----END PGP SIGNATURE-----

Re: 2.6.19-rc5: lockdep warnings starting irattach

[Peter Zijlstra]

On Sat, 2006-11-18 at 16:12 +0300, Andrey Borzenkov wrote:

> =============================================
> [ INFO: possible recursive locking detected ]
> 2.6.19-rc5-2avb #2
> - ---------------------------------------------
> pppd/26425 is trying to acquire lock:
>  (&hashbin->hb_spinlock){....}, at: [<dfdea87a>] irlmp_slsap_inuse+0x5a/0x170 
> [irda]
> 
> but task is already holding lock:
>  (&hashbin->hb_spinlock){....}, at: [<dfdea857>] irlmp_slsap_inuse+0x37/0x170 
> [irda]
> 
> other info that might help us debug this:
> 1 lock held by pppd/26425:
>  #0:  (&hashbin->hb_spinlock){....}, at: [<dfdea857>] 
> irlmp_slsap_inuse+0x37/0x170 [irda]
> 
> stack backtrace:
>  [<c010413c>] dump_trace+0x1cc/0x200
>  [<c010418a>] show_trace_log_lvl+0x1a/0x30
>  [<c01047f2>] show_trace+0x12/0x20
>  [<c01048c9>] dump_stack+0x19/0x20
>  [<c01346ca>] __lock_acquire+0x8fa/0xc20
>  [<c0134d2d>] lock_acquire+0x5d/0x80
>  [<c02a851c>] _spin_lock+0x2c/0x40
>  [<dfdea87a>] irlmp_slsap_inuse+0x5a/0x170 [irda]
>  [<dfdebab2>] irlmp_open_lsap+0x62/0x180 [irda]
>  [<dfdf35d1>] irttp_open_tsap+0x181/0x230 [irda]
>  [<dfdc0c3d>] ircomm_open_tsap+0x5d/0xa0 [ircomm]
>  [<dfdc05d8>] ircomm_open+0xb8/0xd0 [ircomm]
>  [<dfdd0477>] ircomm_tty_open+0x4f7/0x570 [ircomm_tty]
>  [<c020bbe4>] tty_open+0x174/0x340
>  [<c016bd69>] chrdev_open+0x89/0x170
>  [<c0167bd6>] __dentry_open+0xa6/0x1d0
>  [<c0167da5>] nameidata_to_filp+0x35/0x40
>  [<c0167df9>] do_filp_open+0x49/0x50
>  [<c0167e47>] do_sys_open+0x47/0xd0
>  [<c0167f0c>] sys_open+0x1c/0x20
>  [<c010307d>] sysenter_past_esp+0x56/0x8d
>  [<b7f86410>] 0xb7f86410
>  =======================

The comment at the nesting lock says:

	/* Careful for priority inversions here !
	 * irlmp->links is never taken while another IrDA
	 * spinlock is held, so we are safe. Jean II */

So, under the assumption the author was right, it just needs a lockdep
annotation.

(depends on patches in -mm for spin_lock_irqsave_nested())

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 net/irda/irlmp.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Index: linux-2.6-mm/net/irda/irlmp.c
===================================================================
--- linux-2.6-mm.orig/net/irda/irlmp.c	2006-11-20 09:54:50.000000000 +0100
+++ linux-2.6-mm/net/irda/irlmp.c	2006-11-20 09:57:39.000000000 +0100
@@ -1678,7 +1678,7 @@ static int irlmp_slsap_inuse(__u8 slsap_
 	 *  every IrLAP connection and check every LSAP associated with each
 	 *  the connection.
 	 */
-	spin_lock_irqsave(&irlmp->links->hb_spinlock, flags);
+	spin_lock_irqsave_nested(&irlmp->links->hb_spinlock, flags, 1);
 	lap = (struct lap_cb *) hashbin_get_first(irlmp->links);
 	while (lap != NULL) {
 		IRDA_ASSERT(lap->magic == LMP_LAP_MAGIC, goto errlap;);

Re: 2.6.19-rc5: lockdep warnings starting irattach

[Ingo Molnar]

* Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:

> The comment at the nesting lock says:
> 
> 	/* Careful for priority inversions here !
> 	 * irlmp->links is never taken while another IrDA
> 	 * spinlock is held, so we are safe. Jean II */
> 
> So, under the assumption the author was right, it just needs a lockdep
> annotation.
> 
> (depends on patches in -mm for spin_lock_irqsave_nested())
> 
> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>

Acked-by: Ingo Molnar <mingo@elte.hu>

	Ingo

[PATCH] lockdep: annotate irda warning

[Peter Zijlstra]

(repost with slightly changed patch)

On Sat, 2006-11-18 at 16:12 +0300, Andrey Borzenkov wrote:

> =============================================
> [ INFO: possible recursive locking detected ]
> 2.6.19-rc5-2avb #2
> - ---------------------------------------------
> pppd/26425 is trying to acquire lock:
>  (&hashbin->hb_spinlock){....}, at: [<dfdea87a>] irlmp_slsap_inuse+0x5a/0x170 
> [irda]
> 
> but task is already holding lock:
>  (&hashbin->hb_spinlock){....}, at: [<dfdea857>] irlmp_slsap_inuse+0x37/0x170 
> [irda]
> 
> other info that might help us debug this:
> 1 lock held by pppd/26425:
>  #0:  (&hashbin->hb_spinlock){....}, at: [<dfdea857>] 
> irlmp_slsap_inuse+0x37/0x170 [irda]
> 
> stack backtrace:
>  [<c010413c>] dump_trace+0x1cc/0x200
>  [<c010418a>] show_trace_log_lvl+0x1a/0x30
>  [<c01047f2>] show_trace+0x12/0x20
>  [<c01048c9>] dump_stack+0x19/0x20
>  [<c01346ca>] __lock_acquire+0x8fa/0xc20
>  [<c0134d2d>] lock_acquire+0x5d/0x80
>  [<c02a851c>] _spin_lock+0x2c/0x40
>  [<dfdea87a>] irlmp_slsap_inuse+0x5a/0x170 [irda]
>  [<dfdebab2>] irlmp_open_lsap+0x62/0x180 [irda]
>  [<dfdf35d1>] irttp_open_tsap+0x181/0x230 [irda]
>  [<dfdc0c3d>] ircomm_open_tsap+0x5d/0xa0 [ircomm]
>  [<dfdc05d8>] ircomm_open+0xb8/0xd0 [ircomm]
>  [<dfdd0477>] ircomm_tty_open+0x4f7/0x570 [ircomm_tty]
>  [<c020bbe4>] tty_open+0x174/0x340
>  [<c016bd69>] chrdev_open+0x89/0x170
>  [<c0167bd6>] __dentry_open+0xa6/0x1d0
>  [<c0167da5>] nameidata_to_filp+0x35/0x40
>  [<c0167df9>] do_filp_open+0x49/0x50
>  [<c0167e47>] do_sys_open+0x47/0xd0
>  [<c0167f0c>] sys_open+0x1c/0x20
>  [<c010307d>] sysenter_past_esp+0x56/0x8d
>  [<b7f86410>] 0xb7f86410
>  =======================

The comment at the nesting lock says:

	/* Careful for priority inversions here !
	 * irlmp->links is never taken while another IrDA
	 * spinlock is held, so we are safe. Jean II */

So, under the assumption the author was right, it just needs a lockdep
annotation.

(depends on patches in -mm for spin_lock_irqsave_nested())

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 net/irda/irlmp.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

Index: linux-2.6-mm/net/irda/irlmp.c
===================================================================
--- linux-2.6-mm.orig/net/irda/irlmp.c	2006-11-20 09:54:50.000000000 +0100
+++ linux-2.6-mm/net/irda/irlmp.c	2006-11-20 10:12:11.000000000 +0100
@@ -1678,7 +1678,8 @@ static int irlmp_slsap_inuse(__u8 slsap_
 	 *  every IrLAP connection and check every LSAP associated with each
 	 *  the connection.
 	 */
-	spin_lock_irqsave(&irlmp->links->hb_spinlock, flags);
+	spin_lock_irqsave_nested(&irlmp->links->hb_spinlock, flags,
+			SINGLE_DEPTH_NESTING);
 	lap = (struct lap_cb *) hashbin_get_first(irlmp->links);
 	while (lap != NULL) {
 		IRDA_ASSERT(lap->magic == LMP_LAP_MAGIC, goto errlap;);

Re: [PATCH] lockdep: annotate irda warning

[Samuel Ortiz]

Hi guys,

On Mon, Nov 20, 2006 at 10:15:39AM +0100, Peter Zijlstra wrote:
>
> So, under the assumption the author was right, it just needs a lockdep
> annotation.
> 
> (depends on patches in -mm for spin_lock_irqsave_nested())
> 
> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
> ---
>  net/irda/irlmp.c |    3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> Index: linux-2.6-mm/net/irda/irlmp.c
> ===================================================================
> --- linux-2.6-mm.orig/net/irda/irlmp.c	2006-11-20 09:54:50.000000000 +0100
> +++ linux-2.6-mm/net/irda/irlmp.c	2006-11-20 10:12:11.000000000 +0100
> @@ -1678,7 +1678,8 @@ static int irlmp_slsap_inuse(__u8 slsap_
>  	 *  every IrLAP connection and check every LSAP associated with each
>  	 *  the connection.
>  	 */
> -	spin_lock_irqsave(&irlmp->links->hb_spinlock, flags);
> +	spin_lock_irqsave_nested(&irlmp->links->hb_spinlock, flags,
> +			SINGLE_DEPTH_NESTING);
>  	lap = (struct lap_cb *) hashbin_get_first(irlmp->links);
>  	while (lap != NULL) {
>  		IRDA_ASSERT(lap->magic == LMP_LAP_MAGIC, goto errlap;);

This patch got pushed into mainline, while the spin_lock_irq_save_nested()
patches are not there yet. It breaks IrDA as irlmp.c doesn't build.
Linus, could you please revert the corresponding commit
(700f9672c9a61c12334651a94d17ec04620e1976) unless you are planning to pull
the spin_lock_irq_save_nested() patches soon from -mm ?
Thanks a lot in advance.

Cheers,
Samuel.