Home » Mailing list archives » linux-netdev » 2008 » March » 26

Re: [AX25] kernel panic

Previous thread: [PATCH] Add Fixed PHY support for ucc_geth by Joakim Tjernlund on Tuesday, March 18, 2008 - 5:46 am. (7 messages)

Next thread: [PATCH] forcedeth: Use round_jiffies for stats timer by Daniel Drake on Tuesday, March 18, 2008 - 7:07 am. (2 messages)
[view original message]
From: Bernard Pidoux F6BVP <f6bvp@...>
To: Linux Netdev List <netdev@...>
Subject: [AX25] kernel panic
Date: Tuesday, March 18, 2008 - 6:44 am

Hi,


I used netconsole to document a kernel panic related to AX25 on one of

my systems.

This occurs systematically as soon as I close an AX25 session

established through a level 3 ROSE network (FPAC) via AXUDP link (ax25ipd).

It is not triggered when connecting locally or via AX25 call through LAN.


This is observed with kernel-2.6.24.2 and -3 with most recent AX25

patches applied.


Bernard Pidoux


BUG: unable to handle kernel paging request at virtual address 6b6b6c2b

printing eip: c0139585 *pde = 00000000

Oops: 0002 [#1]

Modules linked in: netconsole configfs rose netrom mkiss crc16 ax25 mga

drm nfsd exportfs nfs lockd nfs_acl sunrpc af_packet ipv6 snd_seq_dummy

snd_seq_oss snd_seq_midi_event snd_seq snd_pcm_oss snd_mixer_oss

binfmt_misc loop floppy usbhid snd_ens1371 snd_rawmidi snd_seq_device

snd_ac97_codec snd_pcm uhci_hcd snd_timer snd soundcore snd_page_alloc

usbcore ne2k_pci intel_agp ac97_bus ide_cd 8390 agpgart ext3 jbd

ide_disk piix ide_core


Pid: 3527, comm: ax25ipd Not tainted (2.6.24.2 #2)

EIP: 0060:[<c0139585>] EFLAGS: 00010012 CPU: 0

EIP is at __lock_acquire+0x65/0x11d0

EAX: 00000012 EBX: 00000001 ECX: c7670068 EDX: 00000002

ESI: 00000000 EDI: 6b6b6b6b EBP: c6bf5bc0 ESP: c6bf5b38

  DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 0068

Process ax25ipd (pid: 3527, ti=c6bf4000 task=c60251d0 task.ti=c6bf4000)

Stack: 0000580b 00000000 02c059e8 00000000 c04f73d8 02c059e8 00000000

c6bf5be4

        c01397d8 00000001 c017ecc0 00000000 00000002 00000000 c7670068

c60251d0

        0000580b 00000000 02c059d1 00000000 c04f3ce0 00000002 00000000

c75290f8

Call Trace:

  [<c010513a>] show_trace_log_lvl+0x1a/0x30

  [<c01051f9>] show_stack_log_lvl+0xa9/0xd0

  [<c01052e6>] show_registers+0xc6/0x1b0

  [<c01054b5>] die+0xe5/0x220

  [<c011244f>] do_page_fault+0x15f/0x610

  [<c02c2cfa>] error_code+0x6a/0x70

  [<c013a773>] lock_acquire+0x83/0xa0

  [<c02c2abe>] _spin_lock_irqsave+0x3e/0x70

  [<c011458b>] __wake_up+...
[ message continues ]
[view original message]
From: Francois Romieu <romieu@...>
To: Bernard Pidoux F6BVP <f6bvp@...>
Cc: Linux Netdev List <netdev@...>
Subject: Re: [AX25] kernel panic
Date: Tuesday, March 18, 2008 - 5:25 pm

Bernard Pidoux F6BVP <f6bvp@free.fr> :


May be replacing netif_rx() in drivers/net/hamradio/mkiss.c::ax_bump()

by netif_receive_skb() ?


--

Ueimor

--
[view original message]
From: Bernard Pidoux F6BVP <f6bvp@...>
To: Francois Romieu <romieu@...>
Cc: Linux Netdev List <netdev@...>
Subject: Re: [AX25] kernel panic
Date: Wednesday, March 19, 2008 - 3:36 am

Sorry, this does not made it.

I replaced netif_rx() by netif_receive_skb(skb)

but this gives a worst situation : system is frozen.

Thank you for the suggestion.


  =======================

BUG: soft lockup - CPU#0 stuck for 11s! [ax25ipd:3584]


Pid: 3584, comm: ax25ipd Not tainted (2.6.24.3 #7)

EIP: 0060:[<c01ed892>] EFLAGS: 00000246 CPU: 0

EIP is at delay_tsc+0x12/0x20

EAX: 7752234b EBX: 00000001 ECX: 7752234b EDX: 00000016

ESI: 087022c8 EDI: 00000000 EBP: c6bb5bc0 ESP: c6bb5bbc

  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068

CR0: 8005003b CR2: b7dd86f8 CR3: 07960000 CR4: 000000d0

DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000

DR6: ffff0ff0 DR7: 00000400

  [<c010535a>] show_trace_log_lvl+0x1a/0x30

  [<c0105d82>] show_trace+0x12/0x20

  [<c01024cc>] show_regs+0x1c/0x20

  [<c015555b>] softlockup_tick+0x11b/0x150

  [<c0126c12>] run_local_timers+0x12/0x20

  [<c012708f>] update_process_times+0x2f/0x60

  [<c013a355>] tick_periodic+0x25/0x80

  [<c013a3c9>] tick_handle_periodic+0x19/0x80

  [<c0107aca>] timer_interrupt+0x4a/0x60

  [<c0155928>] handle_IRQ_event+0x28/0x60

  [<c015725f>] handle_level_irq+0x7f/0xf0

  [<c0106e68>] do_IRQ+0x48/0xa0

  [<c0104d16>] common_interrupt+0x2e/0x34

  [<c01ed839>] __delay+0x9/0x10

  [<c01f15f2>] _raw_spin_lock+0xb2/0x140

  [<c02cd350>] _spin_lock_bh+0x50/0x60

  [<c8c59e72>] ax_xmit+0x112/0x370 [mkiss]

  [<c025e947>] dev_hard_start_xmit+0x207/0x270

  [<c026f454>] __qdisc_run+0x54/0x1b0

  [<c0261226>] dev_queue_xmit+0x266/0x340

  [<c8c27a87>] ax25_queue_xmit+0x47/0x70 [ax25]

  [<c8c27b77>] ax25_transmit_buffer+0xc7/0x110 [ax25]

  [<c8c27c3d>] ax25_send_iframe+0x7d/0x110 [ax25]

  [<c8c27d80>] ax25_kick+0xb0/0x1b0 [ax25]

  [<c8c28e49>] ax25_std_frame_in+0x69/0x890 [ax25]

  [<c8c26dbb>] ax25_kiss_rcv+0x2bb/0x7c0 [ax25]

  [<c025e24a>] netif_receive_skb+0x38a/0x480

  [<c8c...
[ message continues ]
[view original message]
From: Bernard Pidoux <bpidoux@...>
To: Bernard Pidoux F6BVP <f6bvp@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Wednesday, March 19, 2008 - 4:57 pm

[Empty message]
[view original message]
From: Jarek Poplawski <jarkao2@...>
To: Bernard Pidoux <bpidoux@...>
Cc: Bernard Pidoux F6BVP <f6bvp@...>, Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Thursday, March 20, 2008 - 3:30 am

On Wed, Mar 19, 2008 at 09:57:18PM +0100, Bernard Pidoux wrote:


Well, the rest looks very similarly and not trivial... Probably a

socket is destructed too soon (or something...). Could you try to

reproduce this with some debugging turned off, e.g.:

CONFIG_DEBUG_LOCK_ALLOC, CONFIG_DEBUG_PAGEALLOC, (CONFIG_SLUB_DEBUG?).


Regards,

Jarek P.

--
[view original message]
From: Bernard Pidoux <bpidoux@...>
To: Jarek Poplawski <jarkao2@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Thursday, March 20, 2008 - 6:58 pm

I recompiled the kernel after turning off the above options.

However there is still a kernel panic documented below.

Does it help ?


Bernard Pidoux


BUG: spinlock bad magic on CPU#0, swapper/0

BUG: unable to handle kernel paging request at virtual address 6b6b6c2f

printing eip: c01e730b *pde = 00000000

Oops: 0000 [#1] SMP

Modules linked in: rose netrom mkiss crc16 ax25 mga drm netconsole

configfs nfsd exportfs nfs lockd nfs_acl sunrpc af_packet ipv6

snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_pcm_oss

snd_mixer_oss binfmt_misc loop usb_storage scsi_mod floppy usbhid

snd_ens1371 snd_rawmidi snd_seq_device snd_ac97_codec snd_pcm snd_timer

snd uhci_hcd intel_agp ne2k_pci ide_cd usbcore soundcore snd_page_alloc

ac97_bus 8390 agpgart genrtc ext3 jbd ide_disk piix ide_core


Pid: 0, comm: swapper Not tainted (2.6.24.3 #8)

EIP: 0060:[<c01e730b>] EFLAGS: 00010002 CPU: 0

EIP is at spin_bug+0xdb/0xf0

EAX: 0000002f EBX: 6b6b6b6b ECX: 6b6b6b6b EDX: 00000082

ESI: c07e5a38 EDI: c0333633 EBP: c0379c78 ESP: c0379c54

  DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068

Process swapper (pid: 0, ti=c0378000 task=c0353340 task.ti=c0378000)

Stack: c0333724 c0333633 00000000 c0353509 00000000 00000010 c07e5a38

00000001

        c7b8abb8 c0379cb0 c01e7409 c11448c8 c1147740 00000000 c11448c0

c0379cb8

        c114775c c11448c8 c1147740 00000000 00000282 00000001 c7b8abb8

c0379cbc

Call Trace:

  [<c010527a>] show_trace_log_lvl+0x1a/0x30

  [<c0105339>] show_stack_log_lvl+0xa9/0xd0

  [<c0105427>] show_registers+0xc7/0x1b0

  [<c010561a>] die+0x10a/0x230

  [<c0114855>] do_page_fault+0x2f5/0x600

  [<c02c039a>] error_code+0x72/0x78

  [<c01e7409>] _raw_spin_lock+0x69/0x140

  [<c02c01cc>] _spin_lock_irqsave+0xc/0x20

  [<c0117c9b>] __wake_up+0x1b/0x50

  [<c024badc>] sock_def_write_space+0x9c/0xc0

  [<c024bdcc>] sock_wfree+0x3c/0x50

  [<c024dfc7>] skb_release_all+0x57/0x130

  [<c024d75b>] __kfree_skb+0x...
[ message continues ]
[view original message]
From: Jarek Poplawski <jarkao2@...>
To: Bernard Pidoux <bpidoux@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Saturday, March 22, 2008 - 7:36 am

On Thu, Mar 20, 2008 at 11:58:35PM +0100, Bernard Pidoux wrote:

...


Hi,


Bernard, could you try this debugging patch?


Thanks,

Jarek P.


---


diff -Nurp 2.6.24.3-/net/core/sock.c 2.6.24.3+/net/core/sock.c

--- 2.6.24.3-/net/core/sock.c	2008-01-24 23:58:37.000000000 +0100

+++ 2.6.24.3+/net/core/sock.c	2008-03-22 12:10:10.000000000 +0100

@@ -1517,14 +1517,21 @@ static void sock_def_write_space(struct

 	 * progress.  --DaveM

 	 */

 	if ((atomic_read(&sk->sk_wmem_alloc) << 1) <= sk->sk_sndbuf) {

-		if (sk->sk_sleep && waitqueue_active(sk->sk_sleep))

+		if (sk->sk_sleep && waitqueue_active(sk->sk_sleep)) {

+			if (!sk->sk_socket || sk->sk_socket->sk != sk) {

+				WARN_ON_ONCE(1);

+				sk->sk_socket = NULL;

+				sk->sk_sleep = NULL;

+				goto out;

+			}

 			wake_up_interruptible(sk->sk_sleep);

+		}


 		/* Should agree with poll, otherwise some programs break */

 		if (sock_writeable(sk))

 			sk_wake_async(sk, 2, POLL_OUT);

 	}

-

+out:

 	read_unlock(&sk->sk_callback_lock);

 }


--
[view original message]
From: Bernard Pidoux <bpidoux@...>
To: Jarek Poplawski <jarkao2@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Sunday, March 23, 2008 - 1:00 pm

Hi Jarek,


Here is the result of your patch :


WARNING: at net/core/sock.c:1522 sock_def_write_space()

Pid: 3452, comm: mkiss Not tainted 2.6.24.3 #2

  [<c010513a>] show_trace_log_lvl+0x1a/0x30

  [<c0105b52>] show_trace+0x12/0x20

  [<c01064ac>] dump_stack+0x6c/0x80

  [<c024d54b>] sock_def_write_space+0x10b/0x120

  [<c024d93a>] sock_wfree+0x3a/0x40

  [<c024fafd>] skb_release_all+0x4d/0xf0

  [<c024f2ab>] __kfree_skb+0xb/0x90

  [<c024f349>] kfree_skb+0x19/0x30

  [<c8cb5e37>] ax_xmit+0xf7/0x360 [mkiss]

  [<c0255367>] dev_hard_start_xmit+0x207/0x270

  [<c02657f4>] __qdisc_run+0x54/0x1a0

  [<c0257bac>] dev_queue_xmit+0x25c/0x330

  [<c8cc8a77>] ax25_queue_xmit+0x47/0x70 [ax25]

  [<c8cc8b65>] ax25_transmit_buffer+0xc5/0x110 [ax25]

  [<c8cc8c2d>] ax25_send_iframe+0x7d/0x110 [ax25]

  [<c8cc8d70>] ax25_kick+0xb0/0x1b0 [ax25]

  [<c8cc9e29>] ax25_std_frame_in+0x69/0x890 [ax25]

  [<c8cc7dba>] ax25_kiss_rcv+0x2ba/0x7b0 [ax25]

  [<c0254cc4>] netif_receive_skb+0x244/0x2e0

  [<c0257835>] process_backlog+0x75/0xe0

  [<c025734b>] net_rx_action+0x12b/0x1d0

  [<c011d595>] __do_softirq+0x55/0xc0

  [<c011d647>] do_softirq+0x47/0x50

  [<c011d9f5>] local_bh_enable_ip+0x95/0xe0

  [<c02babe5>] _spin_unlock_bh+0x25/0x30

  [<c8cb56b0>] mkiss_receive_buf+0x360/0x420 [mkiss]

  [<c020dfe7>] pty_write+0x47/0x60

  [<c020b5b2>] write_chan+0x222/0x330

  [<c0208c76>] tty_write+0x126/0x1b0

  [<c01728df>] vfs_write+0xaf/0x120

  [<c0172e2d>] sys_write+0x3d/0x70

  [<c010412a>] sysenter_past_esp+0x5f/0xa5

  =======================


Happy Easter,


Bernard Pidoux


--
[view original message]
From: Jarek Poplawski <jarkao2@...>
To: Bernard Pidoux <bpidoux@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Monday, March 24, 2008 - 4:51 pm

On Sun, Mar 23, 2008 at 06:00:17PM +0100, Bernard Pidoux wrote:

...


Thanks. Here is one more (to apply after the previous one).


Happy Holidays...

Jarek P.


---


diff -Nurp 2.6.24.3-with1/include/net/sock.h 2.6.24.3-with2/include/net/sock.h

--- 2.6.24.3-with1/include/net/sock.h	2008-01-24 23:58:37.000000000 +0100

+++ 2.6.24.3-with2/include/net/sock.h	2008-03-24 21:14:43.000000000 +0100

@@ -263,6 +263,7 @@ struct sock {

   	int			(*sk_backlog_rcv)(struct sock *sk,

 						  struct sk_buff *skb);

 	void                    (*sk_destruct)(struct sock *sk);

+	int			sk_ax25_debug;

 };


 /*

diff -Nurp 2.6.24.3-with1/net/ax25/af_ax25.c 2.6.24.3-with2/net/ax25/af_ax25.c

--- 2.6.24.3-with1/net/ax25/af_ax25.c	2008-01-24 23:58:37.000000000 +0100

+++ 2.6.24.3-with2/net/ax25/af_ax25.c	2008-03-23 00:53:51.000000000 +0100

@@ -936,6 +936,7 @@ struct sock *ax25_make_new(struct sock *

 	}


 	sk->sk_protinfo = ax25;

+	sk->sk_ax25_debug = 1;

 	ax25->sk    = sk;


 	return sk;

diff -Nurp 2.6.24.3-with1/net/core/sock.c 2.6.24.3-with2/net/core/sock.c

--- 2.6.24.3-with1/net/core/sock.c	2008-03-22 12:10:10.000000000 +0100

+++ 2.6.24.3-with2/net/core/sock.c	2008-03-24 21:43:46.000000000 +0100

@@ -1519,6 +1519,9 @@ static void sock_def_write_space(struct

 	if ((atomic_read(&sk->sk_wmem_alloc) << 1) <= sk->sk_sndbuf) {

 		if (sk->sk_sleep && waitqueue_active(sk->sk_sleep)) {

 			if (!sk->sk_socket || sk->sk_socket->sk != sk) {

+				if (net_ratelimit())

+					printk(KERN_INFO "sk_ax25_debug = %d, %p\n",

+						 sk->sk_ax25_debug, sk->sk_socket);

 				WARN_ON_ONCE(1);

 				sk->sk_socket = NULL;

 				sk->sk_sleep = NULL;

--
[view original message]
From: Bernard Pidoux <bpidoux@...>
To: Jarek Poplawski <jarkao2@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Tuesday, March 25, 2008 - 9:23 am

After applying this patch I did not see any sk_ax25_debug = 1 into

/var/log/messages. It was always = 0.


  Mar 25 12:03:07 f6bvp-9 kernel: sk_ax25_debug = 0, c08de2e0

  Mar 25 12:11:55 f6bvp-9 kernel: sk_ax25_debug = 0, c08de2e0

  Mar 25 12:13:07 f6bvp-9 kernel: sk_ax25_debug = 0, c08de2e0

  Mar 25 12:16:39 f6bvp-9 kernel: sk_ax25_debug = 0, c08de040

  Mar 25 12:18:57 f6bvp-9 kernel: sk_ax25_debug = 0, c08de040

  Mar 25 12:23:12 f6bvp-9 kernel: sk_ax25_debug = 0, c08de580

  Mar 25 12:26:50 f6bvp-9 kernel: sk_ax25_debug = 0, c08de580

  Mar 25 12:32:34 f6bvp-9 kernel: sk_ax25_debug = 0, c08de040

  Mar 25 12:33:14 f6bvp-9 kernel: sk_ax25_debug = 0, c08de2e0

  Mar 25 12:33:57 f6bvp-9 kernel: sk_ax25_debug = 0, c08de040

  Mar 25 12:36:56 f6bvp-9 kernel: sk_ax25_debug = 0, c08de580

  Mar 25 12:43:16 f6bvp-9 kernel: sk_ax25_debug = 0, c08de580

  Mar 25 12:48:34 f6bvp-9 kernel: sk_ax25_debug = 0, c08de580

  Mar 25 12:48:57 f6bvp-9 kernel: sk_ax25_debug = 0, c08de040

  Mar 25 12:53:10 f6bvp-9 kernel: sk_ax25_debug = 0, c08de580

  Mar 25 13:00:31 f6bvp-9 kernel: sk_ax25_debug = 0, c08de2e0

  Mar 25 13:03:24 f6bvp-9 kernel: sk_ax25_debug = 0, c08de2e0

  Mar 25 13:03:56 f6bvp-9 kernel: sk_ax25_debug = 0, c08de040

  Mar 25 13:10:52 f6bvp-9 kernel: sk_ax25_debug = 0, c08de2e0

  Mar 25 13:13:22 f6bvp-9 kernel: sk_ax25_debug = 0, c08de040


Here is an extract of /var/log/kernel/warnings.log


  Mar 25 11:54:04 f6bvp-9 kernel: NET: Registered protocol family 3

  Mar 25 11:54:04 f6bvp-9 kernel: mkiss: AX.25 Multikiss, Hans Albas PE1AYX

  Mar 25 11:54:04 f6bvp-9 mkiss[3443]: starting

  Mar 25 11:54:04 f6bvp-9 kernel: mkiss: ax0: crc mode is auto.

  Mar 25 11:54:04 f6bvp-9 kernel: ADDRCONF(NETDEV_CHANGE): ax0: link

becomes ready

  Mar 25 11:54:04 f6bvp-9 kernel: mkiss: ax1: crc mode is auto.

  Mar 25 11:54:04 f6bvp-9 kernel: ADDRCONF(NETDEV_CHANGE): ax1: link

becomes ready

  Mar 25 11:54:05 f6bvp-9 mkiss[3455]: starting

  Mar 25 11:54:05 f6bvp-9 kernel: mkiss: ax2: crc mode is auto.

  Mar...
[ message continues ]
[view original message]
From: Jarek Poplawski <jarkao2@...>
To: Bernard Pidoux <bpidoux@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Wednesday, March 26, 2008 - 2:35 pm

On Tue, Mar 25, 2008 at 02:23:46PM +0100, Bernard Pidoux wrote:

...


OK, here is the next try. (Apply after all previous patches to any 2.6.24.x;

there would be more noise, so revert it after catching some data.)


Thanks,

Jarek P.


(debugging patch #3)

---


diff -Nurp 2.6.24.3-with2/include/net/sock.h 2.6.24.3-with3/include/net/sock.h

--- 2.6.24.3-with2/include/net/sock.h	2008-03-24 21:14:43.000000000 +0100

+++ 2.6.24.3-with3/include/net/sock.h	2008-03-26 19:10:33.000000000 +0100

@@ -1414,4 +1414,20 @@ extern int sysctl_optmem_max;

 extern __u32 sysctl_wmem_default;

 extern __u32 sysctl_rmem_default;


+static inline int sock_debug_ax25(struct sk_buff *skb, struct sock *ax25_sk)

+{

+	struct sock *sk = skb->sk;

+

+	if (sk && sk->sk_socket && sk->sk_socket->sk != sk) {

+		if (net_ratelimit())

+			printk(KERN_INFO "AX25: %s, %d, %p, %p, %p, %p, %p\n",

+				__FUNCTION__, __LINE__, sk, sk->sk_socket,

+				sk->sk_socket->sk, ax25_sk,

+				ax25_sk == NULL ? NULL : ax25_sk->sk_socket);

+		skb->destructor = NULL;

+		return 1;

+	}

+	return 0;

+}

+

 #endif	/* _SOCK_H */

diff -Nurp 2.6.24.3-with2/net/ax25/af_ax25.c 2.6.24.3-with3/net/ax25/af_ax25.c

--- 2.6.24.3-with2/net/ax25/af_ax25.c	2008-03-23 00:53:51.000000000 +0100

+++ 2.6.24.3-with3/net/ax25/af_ax25.c	2008-03-26 18:52:54.000000000 +0100

@@ -952,6 +952,8 @@ static int ax25_release(struct socket *s


 	sock_hold(sk);

 	sock_orphan(sk);

+	if (net_ratelimit())

+		printk(KERN_INFO "AX25- %s, %d, %p, %p\n", __FUNCTION__, __LINE__, sk, sock);

 	lock_sock(sk);

 	ax25 = ax25_sk(sk);


@@ -1372,12 +1374,16 @@ static int ax25_accept(struct socket *so

 	newsk		 = skb->sk;

 	newsk->sk_socket = newsock;

 	newsk->sk_sleep	 = &newsock->wait;

+	newsk->sk_ax25_debug = 2;


 	/* Now attach up the new socket */

 	kfree_skb(skb);

 	sk->sk_ack_backlog--;

 	newsock->sk    = newsk;

 	newsock->state = SS_CONNECTED;

+	if (net_ratelimit())

+		printk(KERN_INFO "AX25=...
[ message continues ]
[view original message]
From: Jarek Poplawski <jarkao2@...>
To: Bernard Pidoux <bpidoux@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Friday, March 28, 2008 - 8:07 am

On Wed, Mar 26, 2008 at 07:35:38PM +0100, Jarek Poplawski wrote:


Bernard, I fucked it up with this #3: (changed macro to function at

the last minute). I'm currently reading this, and maybe I'll figure

out something, but if it's not a big problem for you to repeat this

after adding this #4 (after all previous) it should be 'a bit' easier.

(But no need to hurry.)


Very sorry,

Jarek P.


(debugging patch #4)

---


diff -Nurp 2.6.24.3-with3/include/net/sock.h 2.6.24.3-with4/include/net/sock.h

--- 2.6.24.3-with3/include/net/sock.h	2008-03-26 19:10:33.000000000 +0100

+++ 2.6.24.3-with4/include/net/sock.h	2008-03-28 12:42:19.000000000 +0100

@@ -1414,20 +1414,18 @@ extern int sysctl_optmem_max;

 extern __u32 sysctl_wmem_default;

 extern __u32 sysctl_rmem_default;


-static inline int sock_debug_ax25(struct sk_buff *skb, struct sock *ax25_sk)

-{

-	struct sock *sk = skb->sk;

-

-	if (sk && sk->sk_socket && sk->sk_socket->sk != sk) {

-		if (net_ratelimit())

-			printk(KERN_INFO "AX25: %s, %d, %p, %p, %p, %p, %p\n",

-				__FUNCTION__, __LINE__, sk, sk->sk_socket,

-				sk->sk_socket->sk, ax25_sk,

-				ax25_sk == NULL ? NULL : ax25_sk->sk_socket);

-		skb->destructor = NULL;

-		return 1;

-	}

-	return 0;

-}

+#define sock_debug_ax25(__skb, __ax25_sk) 					\

+({										\

+	struct sock *__sk = __skb->sk;						\

+										\

+	if (__sk && __sk->sk_socket && __sk->sk_socket->sk != __sk) {		\

+		if (net_ratelimit())						\

+			printk(KERN_INFO "AX25: %s, %d, %p, %p, %p, %p, %p\n",	\

+				__FUNCTION__, __LINE__, __sk, __sk->sk_socket,	\

+				__sk->sk_socket->sk, __ax25_sk,			\

+				__ax25_sk == NULL ? NULL : __ax25_sk->sk_socket);	\

+		__skb->destructor = NULL;					\

+	}									\

+})


 #endif	/* _SOCK_H */

diff -Nurp 2.6.24.3-with3/net/ax25/af_ax25.c 2.6.24.3-with4/net/ax25/af_ax25.c

--- 2.6.24.3-with3/net/ax25/af_ax25.c	2008-03-26 18:52:54.000000000 +0100

+++ 2.6.24.3-with4/net/ax25/af_ax25.c	200...
[ message continues ]
[view original message]
From: Bernard Pidoux <bpidoux@...>
To: Jarek Poplawski <jarkao2@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Saturday, March 29, 2008 - 8:04 am

Please find below an extract of /var/log/messages

The time of interest is around 12:03


  Mar 29 11:52:53 f6bvp-9 kernel: AX25+ ax25_create, 866, c15602e0, c0f0a820

  Mar 29 11:52:53 f6bvp-9 kernel: AX25- ax25_release, 959, c15602e0,

c0f0a820

  Mar 29 11:52:53 f6bvp-9 kernel: AX25+ ax25_create, 866, c15602e0, c0f0a820

  Mar 29 11:52:53 f6bvp-9 kernel: AX25- ax25_release, 959, c15602e0,

c0f0a820

  Mar 29 11:52:59 f6bvp-9 kernel: AX25: ax25_frames_acked, 59, c6ae2bc8,

c0f0a040, 6b6b6b6b, 00000000, 00000000

  Mar 29 11:53:53 f6bvp-9 kernel: AX25+ ax25_create, 866, c15602e0, c0f0a580

  Mar 29 11:53:53 f6bvp-9 kernel: AX25- ax25_release, 959, c15602e0,

c0f0a580

  Mar 29 11:53:53 f6bvp-9 kernel: AX25+ ax25_create, 866, c15602e0, c0f0a580

  Mar 29 11:53:53 f6bvp-9 kernel: AX25- ax25_release, 959, c15602e0,

c0f0a580

  Mar 29 11:54:53 f6bvp-9 kernel: AX25+ ax25_create, 866, c15602e0, c0f0a580

  Mar 29 11:54:53 f6bvp-9 kernel: AX25- ax25_release, 959, c15602e0,

c0f0a580

  Mar 29 11:54:53 f6bvp-9 kernel: AX25+ ax25_create, 866, c15602e0, c0f0a580

  Mar 29 11:54:53 f6bvp-9 kernel: AX25- ax25_release, 959, c15602e0,

c0f0a580

  Mar 29 11:55:53 f6bvp-9 kernel: AX25+ ax25_create, 866, c15602e0, c0f60820

  Mar 29 11:55:53 f6bvp-9 kernel: AX25- ax25_release, 959, c15602e0,

c0f60820

  Mar 29 11:55:53 f6bvp-9 kernel: AX25+ ax25_create, 866, c15602e0, c0f60820

  Mar 29 11:55:53 f6bvp-9 kernel: AX25- ax25_release, 959, c15602e0,

c0f60820

  Mar 29 11:56:52 f6bvp-9 kernel: AX25+ ax25_create, 866, c15602e0, c0f60820

  Mar 29 11:56:52 f6bvp-9 kernel: AX25- ax25_release, 959, c15602e0,

c0f60820

  Mar 29 11:56:53 f6bvp-9 kernel: AX25+ ax25_create, 866, c15602e0, c0f60820

  Mar 29 11:56:53 f6bvp-9 kernel: AX25- ax25_release, 959, c15602e0,

c0f60820

  Mar 29 11:56:53 f6bvp-9 kernel: AX25+ ax25_create, 866, c15602e0, c0f60820

  Mar 29 11:56:53 f6bvp-9 kernel: AX25- ax25_release, 959, c15602e0,

c0f60820

  Mar 29 11:56:53 f6bvp-9 kernel: AX25+ ax25_create, 866, c15602e0, c0f60820

  Mar 29 11...
[ message continues ]
[view original message]
From: Jarek Poplawski <jarkao2@...>
To: Bernard Pidoux <bpidoux@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Saturday, March 29, 2008 - 8:24 am

On Sat, Mar 29, 2008 at 01:04:16PM +0100, Bernard Pidoux wrote:

...


Thanks Bernard! ...On the other hand, if you could hold these patches

a while yet and maybe find more of these "AX25:" it should be very

helpful (of course with other AX25* lines).


Regards,

Jarek P.

--
[view original message]
From: Bernard Pidoux <bpidoux@...>
To: Jarek Poplawski <jarkao2@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Sunday, March 30, 2008 - 5:00 am

Jarek,


I enlarged grep filter in order to grab AX.25 data that might be

interesting too.


And this is the result :


# cat /var/log/kernel/info.log | grep AX...

Mar 30 10:19:00 f6bvp-9 kernel: AX25+ ax25_create, 866, c14e6e28, c74322e0

Mar 30 10:19:00 f6bvp-9 kernel: AX25- ax25_release, 959, c14e6e28, c74322e0

Mar 30 10:19:13 f6bvp-9 kernel: AX25: ax25_frames_acked, 59, c2aa4130,

c74322e0, 6b6b6b6b, 00000000, 00000000

Mar 30 10:19:59 f6bvp-9 kernel: AX25+ ax25_create, 866, c14e7a70, c74322e0

Mar 30 10:19:59 f6bvp-9 kernel: AX25- ax25_release, 959, c14e7a70, c74322e0

Mar 30 10:20:00 f6bvp-9 kernel: AX25+ ax25_create, 866, c14e7a70, c74322e0

Mar 30 10:20:00 f6bvp-9 kernel: AX25- ax25_release, 959, c14e7a70, c74322e0

Mar 30 10:20:59 f6bvp-9 kernel: AX25+ ax25_create, 866, c14e6e28, c74322e0

Mar 30 10:20:59 f6bvp-9 kernel: AX25- ax25_release, 959, c14e6e28, c74322e0

Mar 30 10:21:00 f6bvp-9 kernel: AX25+ ax25_create, 866, c14e6e28, c74322e0

Mar 30 10:21:00 f6bvp-9 kernel: AX25- ax25_release, 959, c14e6e28, c74322e0

Mar 30 10:21:59 f6bvp-9 kernel: AX25+ ax25_create, 866, c14e6e28, c74322e0

Mar 30 10:21:59 f6bvp-9 kernel: AX25- ax25_release, 959, c14e6e28, c74322e0

Mar 30 10:22:00 f6bvp-9 kernel: AX25+ ax25_create, 866, c14e6e28, c74322e0

Mar 30 10:22:00 f6bvp-9 kernel: AX25- ax25_release, 959, c14e6e28, c74322e0

Mar 30 10:22:37 f6bvp-9 kernel: AX25; ax25_make_new, 943, c14e7a70, c51ad820

Mar 30 10:22:37 f6bvp-9 kernel: AX25= ax25_accept, 1386, c14e7a70,

c51ad820, c74322e0

Mar 30 10:22:38 f6bvp-9 kernel: AX.25: sendto: Addresses built.

Mar 30 10:22:38 f6bvp-9 kernel: AX.25: sendto: building packet.

Mar 30 10:22:38 f6bvp-9 kernel: AX.25: Appending user data

Mar 30 10:22:38 f6bvp-9 kernel: AX.25: Transmitting buffer

Mar 30 10:22:39 f6bvp-9 kernel: AX.25: sendto: Addresses built.

Mar 30 10:22:39 f6bvp-9 kernel: AX.25: sendto: building packet.

Mar 30 10:22:39 f6bvp-9 kernel: AX.25: Appending user data

Mar 30 10:22:39 f6bvp-9 kernel: AX.25: Transmitting buffer

Mar 30 10:22:39 f6bvp-9 ker...
[ message continues ]
[view original message]
From: Bernard Pidoux <bpidoux@...>
To: Jarek Poplawski <jarkao2@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Sunday, March 30, 2008 - 4:43 am

Hi Jarek,


please find attached another collection of data from

/var/log/kernel/info.log that I bziped for it is rather large.


Bernard
[view original message]
From: Jarek Poplawski <jarkao2@...>
To: Bernard Pidoux <bpidoux@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Sunday, March 30, 2008 - 5:16 am

Thanks again. It looks like ROSE sockets aren't released properly.

Here is a try to fix this. Could you test it with all previous patches

present?


Regards,

Jarek P.


(patch #5)

---


diff -Nurp 2.6.24.4-/net/rose/af_rose.c 2.6.24.4+/net/rose/af_rose.c

--- 2.6.24.4-/net/rose/af_rose.c	2008-01-24 23:58:37.000000000 +0100

+++ 2.6.24.4+/net/rose/af_rose.c	2008-03-30 10:53:19.000000000 +0200

@@ -599,17 +599,24 @@ static int rose_release(struct socket *s


 	if (sk == NULL) return 0;


+	sock_hold(sk);

+	sock_orphan(sk);

+	lock_sock(sk);

 	rose = rose_sk(sk);


 	switch (rose->state) {

 	case ROSE_STATE_0:

+		release_sock(sk);

 		rose_disconnect(sk, 0, -1, -1);

+		lock_sock(sk);

 		rose_destroy_socket(sk);

 		break;


 	case ROSE_STATE_2:

 		rose->neighbour->use--;

+		release_sock(sk);

 		rose_disconnect(sk, 0, -1, -1);

+		lock_sock(sk);

 		rose_destroy_socket(sk);

 		break;


@@ -634,6 +641,8 @@ static int rose_release(struct socket *s

 	}


 	sock->sk = NULL;

+	release_sock(sk);

+	sock_put(sk);


 	return 0;

 }

--
[view original message]
From: Bernard Pidoux <bpidoux@...>
To: Jarek Poplawski <jarkao2@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Sunday, March 30, 2008 - 1:38 pm

Jarek,


It looks like you hit the right target !


There is no more warnings and no more ax25_frames_acked or

sk_ax25_debug.


We only get ax25_create and ax25_release data, plus ax25_make_new when

there is an AX25 connexion.


cat /var/log/messages | grep 'Mar 30 17' | grep AX...

Mar 30 17:00:58 f6bvp-9 kernel: AX25+ ax25_create, 866, c73a6738, c3300040

Mar 30 17:00:58 f6bvp-9 kernel: AX25- ax25_release, 959, c73a6738, c3300040

Mar 30 17:00:58 f6bvp-9 kernel: AX25+ ax25_create, 866, c73a6738, c3300040

Mar 30 17:00:58 f6bvp-9 kernel: AX25- ax25_release, 959, c73a6738, c3300040

Mar 30 17:01:23 f6bvp-9 kernel: AX25; ax25_make_new, 943, c73a6738, c762a820

Mar 30 17:01:23 f6bvp-9 kernel: AX25= ax25_accept, 1386, c73a6738,

c762a820, c3300040

Mar 30 17:01:58 f6bvp-9 kernel: AX25+ ax25_create, 866, c73a6f68, c7673ac0

Mar 30 17:01:58 f6bvp-9 kernel: AX25- ax25_release, 959, c73a6f68, c7673ac0

Mar 30 17:01:58 f6bvp-9 kernel: AX25+ ax25_create, 866, c73a6f68, c7673ac0


Please excuse the following questions if you find they are too naive.

Could you tell us how did you suspect ROSE socket from the data

collected ? and why improperly released ROSE sockets could interfere

with AX25 ?


Regards,


Bernard P.
[view original message]
From: Jarek Poplawski <jarkao2@...>
To: Bernard Pidoux <bpidoux@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Sunday, March 30, 2008 - 2:49 pm

On Sun, Mar 30, 2008 at 07:38:51PM +0200, Bernard Pidoux wrote:


No problem! Patch #3 added to "sk_ax25_debug =" sk->sk_family as the

2nd number, and 11 according to include/linux/socket.h means AF_ROSE.


sock_def_write() function in net/core/sock.c where the OOPs triggered

runs some cleaning on sockets when their skbs are kfreed, but this

shouldn't be done after socket had been closed (sock_orphan() removes

some pointers). When ROSE sends it's skbs with ax25_send_frame(), they

are added to AX25 queues, and kfreed after transmitting. If it's after

ROSE socket had been released, but not properly updated, the cleaning

is done on kfreed memory not used by the socket anymore.


Cheers,

Jarek P.

--
[view original message]
From: Bernard Pidoux <bpidoux@...>
To: Linux Netdev List <netdev@...>
Subject: Re: [AX25] kernel panic
Date: Friday, March 28, 2008 - 4:27 am

Hi Jarek,


Here is the result of cat /var/log/messages | grep -B1 -A50 core/sock

No event was detected before 16:33.


--

Mar 27 16:33:05 f6bvp-9 kernel: sk_ax25_debug = 0, 11, 0, 5, c08922e0,

6b6b6b6b, c7790190

Mar 27 16:33:05 f6bvp-9 kernel: WARNING: at net/core/sock.c:1561

sock_def_write_space()

Mar 27 16:33:05 f6bvp-9 kernel: Pid: 3454, comm: mkiss Not tainted

2.6.24.4 #4

Mar 27 16:33:05 f6bvp-9 kernel:  [<c010513a>] show_trace_log_lvl+0x1a/0x30

Mar 27 16:33:05 f6bvp-9 kernel:  [<c0105b52>] show_trace+0x12/0x20

Mar 27 16:33:05 f6bvp-9 kernel:  [<c01064ac>] dump_stack+0x6c/0x80

Mar 27 16:33:05 f6bvp-9 kernel:  [<c024d68b>]

sock_def_write_space+0x17b/0x190

Mar 27 16:33:05 f6bvp-9 kernel:  [<c024da7a>] sock_wfree+0x3a/0x40

Mar 27 16:33:05 f6bvp-9 kernel:  [<c024fc3d>] skb_release_all+0x4d/0xf0

Mar 27 16:33:05 f6bvp-9 kernel:  [<c024f3eb>] __kfree_skb+0xb/0x90

Mar 27 16:33:05 f6bvp-9 kernel:  [<c024f489>] kfree_skb+0x19/0x30

Mar 27 16:33:05 f6bvp-9 kernel:  [<c8cbfe37>] ax_xmit+0xf7/0x360 [mkiss]

Mar 27 16:33:05 f6bvp-9 kernel:  [<c02554a7>]

dev_hard_start_xmit+0x207/0x270

Mar 27 16:33:05 f6bvp-9 kernel:  [<c0265934>] __qdisc_run+0x54/0x1a0

Mar 27 16:33:05 f6bvp-9 kernel:  [<c0257cec>] dev_queue_xmit+0x25c/0x330

Mar 27 16:33:05 f6bvp-9 kernel:  [<c8befa77>] ax25_queue_xmit+0x47/0x70

[ax25]

Mar 27 16:33:05 f6bvp-9 kernel:  [<c8befb65>]

ax25_transmit_buffer+0xc5/0x110 [ax25]

Mar 27 16:33:05 f6bvp-9 kernel:  [<c8befc2d>]

ax25_send_iframe+0x7d/0x110 [ax25]

Mar 27 16:33:06 f6bvp-9 kernel:  [<c8befd70>] ax25_kick+0xb0/0x2b0 [ax25]

Mar 27 16:33:06 f6bvp-9 kernel:  [<c8bf10c9>]

ax25_std_frame_in+0x69/0x890 [ax25]

Mar 27 16:33:06 f6bvp-9 kernel:  [<c8beedba>] ax25_kiss_rcv+0x2ba/0x7b0

[ax25]

Mar 27 16:33:06 f6bvp-9 kernel:  [<c0254e04>] netif_receive_skb+0x244/0x2e0

Mar 27 16:33:06 f6bvp-9 kernel:  [<c0257975>] process_backlog+0x75/0xe0

Mar 27 16:33:06 f6bvp-9 ke...
[ message continues ]
[view original message]
From: Bernard Pidoux <bpidoux@...>
To: Linux Netdev List <netdev@...>
Subject: Re: [AX25] kernel panic
Date: Friday, March 28, 2008 - 4:24 am

Hi Jarek,


Thank you for the new debugging patch.

I think I have collected enough data for you to make some interesting

deductions.


Here are the data collected and filtered with


cat  /var/log/kernel/info.log | grep AX25

I have noticed that there is no AX25= result, although the system

ran a few hours.


Mar 27 15:50:49 f6bvp-9 kernel: AX25- ax25_release, 952, c340fad0, c513a820

Mar 27 15:50:54 f6bvp-9 kernel: AX25- ax25_release, 952, c0f07b10, c0cc0820

Mar 27 15:50:55 f6bvp-9 kernel: AX25- ax25_release, 952, c34676d8, c0cc0d60

Mar 27 15:50:55 f6bvp-9 kernel: AX25- ax25_release, 952, c340e240, c0cc0820

Mar 27 15:50:55 f6bvp-9 kernel: AX25- ax25_release, 952, c0f07b10, c0cc02e0

Mar 27 15:50:55 f6bvp-9 kernel: AX25- ax25_release, 952, c0f07b10, c0cc0d60

Mar 27 15:50:55 f6bvp-9 kernel: AX25- ax25_release, 952, c0f07b10, c0cc02e0

Mar 27 15:51:17 f6bvp-9 kernel: AX25- ax25_release, 952, c0fc4ad0, c148aac0

Mar 27 15:51:22 f6bvp-9 kernel: AX25- ax25_release, 952, c0d482c0, c148a2e0

Mar 27 15:51:22 f6bvp-9 kernel: AX25- ax25_release, 952, c0805758, c148a580

Mar 27 15:51:22 f6bvp-9 kernel: AX25- ax25_release, 952, c0804f28, c148a820

Mar 27 15:51:22 f6bvp-9 kernel: AX25- ax25_release, 952, c08046f8, c0cfdd60

Mar 27 15:51:27 f6bvp-9 kernel: AX25- ax25_release, 952, c0804f28, c148a820

Mar 27 15:51:27 f6bvp-9 kernel: AX25- ax25_release, 952, c0d48af0, c148a820

Mar 27 15:51:27 f6bvp-9 kernel: AX25- ax25_release, 952, c0fc4ad0, c148aac0

Mar 27 15:51:27 f6bvp-9 kernel: AX25- ax25_release, 952, c0804f28, c148a820

Mar 27 15:51:27 f6bvp-9 kernel: AX25- ax25_release, 952, c0804f28, c148a040

Mar 27 15:51:37 f6bvp-9 kernel: AX25- ax25_release, 952, c0fc4ad0, c148aac0

Mar 27 15:51:47 f6bvp-9 kernel: AX25- ax25_release, 952, c0fc4ad0, c148aac0

Mar 27 15:51:47 f6bvp-9 kernel: AX25- ax25_release, 952, c0d486d8, c0cfdd60

Mar 27 15:52:27 f6bvp-9 kernel: AX25- ax25_release, 952, c0d486d8, c0cfdd60

Mar 27 15:52:27 f6bvp-9 kernel: AX25- ax25_release, 952, c0d486d8, c0cfdd60

Mar 27 15:53:27 f6bvp-9 kernel: AX...
[ message continues ]
[view original message]
From: Jarek Poplawski <jarkao2@...>
To: Bernard Pidoux <bpidoux@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Tuesday, March 25, 2008 - 4:51 pm

On Tue, Mar 25, 2008 at 02:23:46PM +0100, Bernard Pidoux wrote:


I need some time yet; it looks like 2.6.24.4 doesn't collide with

these patches, but upgrade could be reasonable for other reasons.


Best regards,

Jarek P.

--
[view original message]
From: Bernard Pidoux <bpidoux@...>
To: Jarek Poplawski <jarkao2@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Friday, March 21, 2008 - 5:40 am

Hi Jarek,


Please ignore the spinlock report I sent in this previous message.

I simply forgot to apply back your mkiss ax_bump() patch when doing some

trials.


After applying again your patch and turning off the debug you asked for,

I observe exactly the same spinlock I first reported i.e. involving ax25ipd.


Sorry for the confusion.


Regards,


Bernard


--
[view original message]
From: Jarek Poplawski <jarkao2@...>
To: Bernard Pidoux <bpidoux@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Thursday, March 20, 2008 - 8:28 pm

On Thu, Mar 20, 2008 at 11:58:35PM +0100, Bernard Pidoux wrote:


I wanted to see something more on the stack than lockdep's code only.

Will see... It seems some memory debugging is still on and doing this

poisoning with '6b', but I can't check now what it is. (Sometimes this

could omit some warnings).


Thanks,

Jarek P.

--
[view original message]
From: Bernard Pidoux <bpidoux@...>
To: Jarek Poplawski <jarkao2@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Friday, March 21, 2008 - 5:52 am

The only difference I can see compared to the original spinlock report

is this line :


[<c01e7409>] _raw_spin_lock+0x69/0x140


Regards,


Bernard P.


BUG: spinlock bad magic on CPU#0, ax25ipd/3648

BUG: unable to handle kernel paging request at virtual address 6b6b6c2f

printing eip: c01e730b *pde = 00000000

Oops: 0000 [#1] SMP

Modules linked in: rose netrom mkiss crc16 ax25 mga drm netconsole

configfs nfsd exportfs nfs lockd nfs_acl sunrpc af_packet ipv6

snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_pcm_oss

snd_mixer_oss binfmt_misc loop usb_storage scsi_mod floppy usbhid

snd_ens1371 snd_rawmidi snd_seq_device snd_ac97_codec snd_pcm snd_timer

snd uhci_hcd soundcore snd_page_alloc usbcore ne2k_pci ide_cd 8390

ac97_bus intel_agp agpgart genrtc ext3 jbd ide_disk piix ide_core


Pid: 3648, comm: ax25ipd Not tainted (2.6.24.3 #8)

EIP: 0060:[<c01e730b>] EFLAGS: 00010002 CPU: 0

EIP is at spin_bug+0xdb/0xf0

EAX: 00000032 EBX: 6b6b6b6b ECX: 6b6b6b6b EDX: 00000096

ESI: c0e4bc30 EDI: c0333633 EBP: c5c7bbc4 ESP: c5c7bba0

  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068

Process ax25ipd (pid: 3648, ti=c5c7a000 task=c6abec90 task.ti=c5c7a000)

Stack: c0333724 c0333633 00000000 c6abee59 00000e40 00000007 c0e4bc30

00000001

        c697ebf8 c5c7bbfc c01e7409 00000000 00000001 c0180640 00000000

00000000

        00000003 c0a320b8 00000000 c6abec90 00000282 00000001 c697ebf8

c5c7bc08

Call Trace:

  [<c010527a>] show_trace_log_lvl+0x1a/0x30

  [<c0105339>] show_stack_log_lvl+0xa9/0xd0

  [<c0105427>] show_registers+0xc7/0x1b0

  [<c010561a>] die+0x10a/0x230

  [<c0114855>] do_page_fault+0x2f5/0x600

  [<c02c039a>] error_code+0x72/0x78

  [<c01e7409>] _raw_spin_lock+0x69/0x140

  [<c02c01cc>] _spin_lock_irqsave+0xc/0x20

  [<c0117c9b>] __wake_up+0x1b/0x50

  [<c024badc>] sock_def_write_space+0x9c/0xc0

  [<c024bdcc>] sock_wfree+0x3c/0x50

  [<c024dfc7>] skb_release_all+0x57/0x130

  [<c024d75b&...
[ message continues ]
[view original message]
From: Jarek Poplawski <jarkao2@...>
To: Bernard Pidoux <bpidoux@...>
Cc: Francois Romieu <romieu@...>, Linux Netdev List <netdev@...>, Jarek Poplawski <jarkao2@...>
Subject: Re: [AX25] kernel panic
Date: Friday, March 21, 2008 - 7:22 am

On Fri, Mar 21, 2008 at 10:52:27AM +0100, Bernard Pidoux wrote:


Bernard, thanks for these reports. I simply need more time to find

various places in ax25 which could matter here. Of course any

additional information could help, but these reports mostly repeat

the same. BTW, I wonder if it's possible to run this command for

closing the session which causes the bug with strace?:

strace -o strace.log a_command


Regards,

Jarek P.

--
Previous thread: [PATCH] Add Fixed PHY support for ucc_geth by Joakim Tjernlund on Tuesday, March 18, 2008 - 5:46 am. (7 messages)

Next thread: [PATCH] forcedeth: Use round_jiffies for stats timer by Daniel Drake on Tuesday, March 18, 2008 - 7:07 am. (2 messages)