Re: unable to handle kernel NULL pointer dereference...

[Dan Williams <dcbw redhat com>]

On Sat, 2009-05-16 at 11:12 +0200, Gour wrote:
> Hi!
> 
> After managing to make NM playing nicely with my Option Icon 225
> usb-stick, I experienced few crashes yesterday and one of those are
> captured in this log snippet:
> 
> -- BUG: unable to handle kernel NULL pointer dereference at 0000000000000020
> IP: [<ffffffff802676e4>] debug_mutex_wake_waiter+0x44/0x160
> PGD 4626a067 PUD 4626b067 PMD 0 
> Oops: 0000 [#1] PREEMPT SMP 
> last sysfs file: /sys/devices/pci0000:00/0000:00:13.0/usb2/2-3/2-3:1.0/net/hso0/statistics/collisions
> CPU 0 
> Modules linked in: radeon drm ipv6 sco bridge stp llc bnep l2cap
> bluetooth adm1021 hso rfkill usb_storage joydev serio_raw pcspkr psmouse
> ohci1394 ssb pcmcia yenta_socket rsrc_nonstatic ieee1394 pcmcia_core wmi
> container video output k8temp sg i2c_piix4 i2c_core shpchp pci_hotplug
> evdev thermal fan snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device
> button battery ac vboxdrv snd_atiixp snd_ac97_codec snd_pcm_oss snd_pcm
> snd_timer snd_page_alloc snd_mixer_oss snd soundcore ac97_bus md_mod
> cpufreq_ondemand powernow_k8 freq_table processor tg3 libphy rtc_cmos
> rtc_core rtc_lib ext3 jbd mbcache dm_mod sr_mod cdrom ohci_hcd sd_mod
> ehci_hcd usbcore ata_generic pata_atiixp libata scsi_mod
> 
> Pid: 5808, comm: NetworkManager Tainted: G   M       2.6.29-ARCH #1 HP Compaq nx6125 (PY417EA#AKN)  
> RIP: 0010:[<ffffffff802676e4>]  [<ffffffff802676e4>] debug_mutex_wake_waiter+0x44/0x160
> RSP: 0018:ffff88004620ddc8  EFLAGS: 00010017
> RAX: ffff880043e6fc78 RBX: 0000000000000000 RCX: 0000000000000000
> RDX: 0000000000000100 RSI: 0000000000000000 RDI: ffff880043e6fc74
> RBP: ffff880043e6fc70 R08: 0000000000000000 R09: 0000000000000000
> R10: 0000000000000000 R11: 2222222222222222 R12: 0000000000000246
> R13: 0000000000000000 R14: 0000000000000008 R15: 0000000000000000
> FS:  00007f9926236740(0000) GS:ffffffff8067a000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 0000000000000020 CR3: 0000000046ba7000 CR4: 00000000000006e0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> Process NetworkManager (pid: 5808, threadinfo ffff88004620c000, task ffff880046155200)
> Stack:
>  ffff880043e6f09c ffff880043e6fc70 ffff880043e6fc74 ffffffff804b2896
>  ffffffff804b4775 ffff880046052800 0000000000000000 ffff88002cca6300
>  ffff88002bebac08 ffffffff803c4812 0000000000000001 0000000000000000
> Call Trace:
>  [<ffffffff804b2896>] ? __mutex_unlock_slowpath+0x96/0x180
>  [<ffffffff804b4775>] ? _spin_lock+0x15/0x20
>  [<ffffffff803c4812>] ? tty_release_dev+0x162/0x5a0
>  [<ffffffff8040cb6d>] ? sys_sendto+0x11d/0x180
>  [<ffffffff802d9a7d>] ? vfs_ioctl+0x1d/0xb0
>  [<ffffffff803c4c61>] ? tty_release+0x11/0x20
>  [<ffffffff802cbcd2>] ? __fput+0xc2/0x210
>  [<ffffffff802c859b>] ? filp_close+0x5b/0x90
>  [<ffffffff802c8684>] ? sys_close+0xb4/0x120
>  [<ffffffff8020c6aa>] ? system_call_fastpath+0x16/0x1b
> Code: 89 fd 45 85 c9 75 36 48 8d 7f 04 e8 07 04 fc ff 90 85 c0 74 7c 44 8b 05 4b 97 42 00 45 85 c0 75 1c 48 8d 45 08 48 39 45 08 74 42 <48> 3b 5b 20 75 1b 48 3b 1b 0f 1f 00 0f 84 8f 00 00 00 48 8b 5c 
> RIP  [<ffffffff802676e4>] debug_mutex_wake_waiter+0x44/0x160
>  RSP <ffff88004620ddc8>
> CR2: 0000000000000020
> ---[ end trace 80249a4d042672e9 ]---
> 
> 
> 
> Is it of any help to NM-devs?

Since it's a kernel bug, there's nothing NM can do about it.  More
recent kernels have a much better 'hso' driver, which is the driver for
your Option iCON 225 card.  Since 2.6.26, the 'hso' driver has seen
quite a few stability and reliability updates.  So upgrading to the
latest kernel you can may help you.  Or asking your distro to backport
fixes to 'hso' from later kernels.

Dan