regression 2.6.30-rc4+git: oops in rfcomm_tty_open klist_add_tail
Dear all,
I am seeing the oops below as soon as I try to bring up my internet connection over rfcomm via a bluetooth device. I think that things were working ok with 2.6.30-rc3 still but now I see this oops even after a clean reboot, reproducibly ...
Any ideas?
Soeren
------------[ cut here ]------------ WARNING: at kernel/workqueue.c:371 flush_cpu_workqueue+0x7d/0x80() Hardware name: NC10 Modules linked in: ath5k mac80211 cfg80211 tun ppp_mppe ppp_async crc_ccitt ppp_generic slhc rfcomm l2cap fuse binfmt_misc snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_oss snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq snd_timer uvcvideo snd_seq_device videodev v4l1_compat btusb bluetooth psmouse snd i2c_i801 rtc_cmos rtc_core rtc_lib sky2 evdev soundcore snd_page_alloc [last unloaded: cfg80211] Pid: 1418, comm: bluetooth Not tainted 2.6.30-rc4-sonne #8 Call Trace: [<c051dddf>] ? printk+0x18/0x21 [<c0136233>] warn_slowpath+0x73/0xd0 [<c0126709>] ? update_curr+0x99/0x1d0 [<c0127777>] ? set_next_entity+0xa7/0x190 [<c012c877>] ? finish_task_switch+0x97/0xd0 [<c051e5f8>] ? __schedule+0x4e8/0xb20 [<c0127777>] ? set_next_entity+0xa7/0x190 [<c012c877>] ? finish_task_switch+0x97/0xd0 [<c014752d>] flush_cpu_workqueue+0x7d/0x80 [<c0147760>] flush_workqueue+0x30/0x50 [<f8ab9cc6>] add_conn+0x16/0x40 [bluetooth] [<c0147050>] ? worker_thread+0x0/0x250 [<c01471aa>] worker_thread+0x15a/0x250 [<f8ab9cb0>] ? add_conn+0x0/0x40 [bluetooth] [<c014b370>] ? autoremove_wake_function+0x0/0x50 [<c0147050>] ? worker_thread+0x0/0x250 [<c014af56>] kthread+0x46/0x80 [<c014af10>] ? kthread+0x0/0x80 [<c010394f>] kernel_thread_helper+0x7/0x18 ---[ end trace c91b55c4d7c5c20d ]--- BUG: unable to handle kernel NULL pointer dereference at 0000000c IP: [<c050ad5b>] klist_node_init+0x2b/0x60 *pde = 00000000 Oops: 0000 [#1] PREEMPT SMP last sysfs file: /sys/devices/pci0000:00/0000:00:1c.0/0000:02:00.0/net/wlan1/statistics/collisions Modules linked in: ath5k mac80211 cfg80211 tun ppp_mppe ppp_async crc_ccitt ppp_generic slhc rfcomm l2cap fuse binfmt_misc snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_oss snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq snd_timer uvcvideo snd_seq_device videodev v4l1_compat btusb bluetooth psmouse snd i2c_i801 rtc_cmos rtc_core rtc_lib sky2 evdev soundcore snd_page_alloc [last unloaded: cfg80211]
Pid: 13288, comm: pppd Tainted: G W (2.6.30-rc4-sonne #8) NC10
EIP: 0060:[<c050ad5b>] EFLAGS: 00010246 CPU: 1
EIP is at klist_node_init+0x2b/0x60
EAX: e89da560 EBX: e89da554 ECX: c200cae0 EDX: e89da554
ESI: 00000000 EDI: ffffffea EBP: f697dd98 ESP: f697dd84
DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
Process pppd (pid: 13288, ti=f697c000 task=e8926b70 task.ti=f697c000)
Stack: c0679ce3 e1e527a0 e1e527a0 00000000 e89da554 f697dda8 c050ae6a eaacf840 00000000 f697ddd0 c03cb1a3 00000000 00000001 eaacf848 eba32edc eba32ed4 eba32e00 eba32e16 e88d6f80 f697de30 f91c7330 00000004 e88d6f80 eb9d4a58 Call Trace: [<c050ae6a>] ? klist_add_tail+0x1a/0x50 [<c03cb1a3>] ? device_move+0x83/0x230 [<f91c7330>] ? rfcomm_tty_open+0x300/0x330 [rfcomm] [<c035fc21>] ? check_tty_count+0x51/0xc0 [<c01315a0>] ? default_wake_function+0x0/0x10 [<c0362d9a>] ? tty_open+0x1aa/0x4c0 [<c01a6caa>] ? exact_lock+0xa/0x20 [<c03d067c>] ? kobj_lookup+0x17c/0x190 [<c01a6f62>] ? chrdev_open+0xf2/0x1d0 [<c01a24c2>] ? __dentry_open+0xd2/0x2c0 [<c01a2799>] ? nameidata_to_filp+0x59/0x70 [<c01a6e70>] ? chrdev_open+0x0/0x1d0 [<c01afdb1>] ? do_filp_open+0x1c1/0x7d0 [<c0120a75>] ? ptep_set_access_flags+0x35/0x50 [<c018c533>] ? do_wp_page+0x143/0x680 [<c01b9298>] ? alloc_fd+0xe8/0x100 [<c01a22a9>] ? do_sys_open+0x59/0xf0 [<c01a23a9>] ? sys_open+0x29/0x40 [<c0102f34>] ? sysenter_do_call+0x12/0x26 Code: 55 89 e5 83 ec 14 89 75 fc 89 c6 8d 42 04 89 5d f8 89 d3 89 42 04 89 42 08 8d 42 0c e8 af 2e dd ff f7 c6 01 00 00 00 89 33 75 15 <8b> 56 0c 85 d2 74 04 89 d8 ff d2 8b 5d f8 8b 75 fc 89 ec 5d c3 EIP: [<c050ad5b>] klist_node_init+0x2b/0x60 SS:ESP 0068:f697dd84 CR2: 000000000000000c ---[ end trace c91b55c4d7c5c20e ]---
Hi Soeren,
Quoted 4 lines
I am seeing the oops below as soon as I try to bring up my internet connection over rfcomm via a bluetooth device. I think that things were working ok with 2.6.30-rc3 still but now I see this oops even after a clean reboot, reproducibly ...
please test with 2.6.30-rc5 since this one should be fixed. There is one additional patch waiting in net-2.6 that fixes a memory allocation in a context that can't sleep.
Regards
Marcel
Quoted 10 lines · Marcel Holtmann
Hi Soeren,
Quoted 4 lines
I am seeing the oops below as soon as I try to bring up my internet connection over rfcomm via a bluetooth device. I think that things were working ok with 2.6.30-rc3 still but now I see this oops even after a clean reboot, reproducibly ...
please test with 2.6.30-rc5 since this one should be fixed. There is one additional patch waiting in net-2.6 that fixes a memory allocation in a context that can't sleep.
Yes indeed, things work again with -rc5, thanks for the quick reply!
Soeren
Signature · 2 lines
For the one fact about the future of which we can be certain is that it will be utterly fantastic. -- Arthur C. Clarke, 1962