Distribution: Debian/testing Hardware Environment: Intel(R) Pentium(R) 4 CPU 2.80GHz Steps to reproduce: Unknown (low reproduction rate) Problem Description: Kernel oops on boot in cifs_mount kern.log from boot earlier that day: Oct 31 10:15:56 kernel: e1000e: Intel(R) PRO/1000 Network Driver - 0.3.3.3-k6 Oct 31 10:15:56 kernel: e1000e: Copyright (c) 1999-2008 Intel Corporation. Oct 31 10:15:56 kernel: e1000e 0000:01:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16 Oct 31 10:15:56 kernel: e1000e 0000:01:00.0: setting latency timer to 64 [...snip...] Oct 31 10:15:56 meowmix kernel: NET: Registered protocol family 10 Oct 31 10:15:56 meowmix kernel: lo: Disabled Privacy Extensions Oct 31 10:15:56 meowmix kernel: ADDRCONF(NETDEV_UP): eth0: link is not ready Oct 31 10:15:56 meowmix kernel: 0000:01:00.0: eth0: Link is Up 1000 Mbps Full Duplex, Flow Control: RX/TX Oct 31 10:15:56 meowmix kernel: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready Oct 31 10:15:56 meowmix kernel: lsb_release used greatest stack depth: 356 bytes left Oct 31 10:15:56 meowmix kernel: eth0: no IPv6 routers present Oct 31 10:15:56 meowmix kernel: CIFS VFS: Error -104 sending data on socket to server Oct 31 10:15:56 meowmix kernel: BUG: unable to handle kernel NULL pointer dereference at 00000518 Oct 31 10:15:56 meowmix kernel: IP: [<c0151c35>] __lock_acquire+0x5d/0xf79 Oct 31 10:15:56 meowmix kernel: *pde = 00000000 Oct 31 10:15:56 meowmix kernel: Oops: 0000 [#1] PREEMPT SMP Oct 31 10:15:56 meowmix kernel: Modules linked in: ipv6 dm_snapshot dm_mirror dm_log dm_mod it87 hwmon_vid sbp2 psmouse ide_disk ide_cd_mod cx88_dvb cx88_vp3054 _i2c mt352 s5h1409 dvb_pll s5h1411 or51132 videobuf_dvb nxt200x isl6421 zl10353 cx24123 lgdt330x dvb_core cx22702 tuner tea5767 tda8290 tda18271 tda827x tuner_x c2028 xc5000 tda9887 tuner_simple tuner_types mt20xx snd_intel8x0 tea5761 snd_ac97_codec v4l2_common ac97_bus snd_pcm_oss cx8802 snd_mixer_oss cx88xx videodev v 4l1_compat snd_pcm ir_common snd_timer i2c_algo_bit pcspkr tveeprom i2c_i801 rtc_cmos btcx_risc snd videobuf_dma_sg videobuf_core i2c_core soundcore iTCO_wdt sn d_page_alloc intel_agp agpgart button joydev evdev usb_storage sg usbhid hid ff_memless sr_mod cdrom sd_mod ide_pci_generic piix ahci ide_core ata_generic ohci1 394 ieee1394 floppy firewire_ohci firewire_core pata_acpi ata_piix r8169 libata scsi_mod dock ehci_hcd uhci_hcd e1000e usbcore thermal processor fan thermal_sys hwmon Oct 31 10:15:56 meowmix kernel: Oct 31 10:15:56 meowmix kernel: Pid: 2973, comm: mount.cifs Not tainted (2.6.27 #50) Oct 31 10:15:56 meowmix kernel: EIP: 0060:[<c0151c35>] EFLAGS: 00010046 CPU: 0 Oct 31 10:15:56 meowmix kernel: EIP is at __lock_acquire+0x5d/0xf79 Oct 31 10:15:56 meowmix kernel: EAX: 00000002 EBX: 00000046 ECX: 00000514 EDX: 00000000 Oct 31 10:15:56 meowmix kernel: ESI: 00000002 EDI: 00000001 EBP: c3179d04 ESP: c3179c9c Oct 31 10:15:56 meowmix kernel: DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068 Oct 31 10:15:56 meowmix kernel: Process mount.cifs (pid: 2973, ti=c3179000 task=c3068000 task.ti=c3179000) Oct 31 10:15:56 meowmix kernel: Stack: c087c080 00000009 1417a047 00000047 00000000 c09d541c c3068000 00000000 Oct 31 10:15:56 meowmix kernel: 00000514 c3068000 0000000e 00000046 c3179cd8 c014f09c 906a8245 c3179ce0 Oct 31 10:15:56 meowmix kernel: c014f12a c3068000 0000000e 00000046 c3179cf8 c014f09c 906a8560 00000046 Oct 31 10:15:56 meowmix kernel: Call Trace: Oct 31 10:15:56 meowmix kernel: [<c014f09c>] ? trace_hardirqs_off_caller+0x15/0x98 Oct 31 10:15:56 meowmix kernel: [<c014f12a>] ? trace_hardirqs_off+0xb/0xd Oct 31 10:15:56 meowmix kernel: [<c014f09c>] ? trace_hardirqs_off_caller+0x15/0x98 Oct 31 10:15:56 meowmix kernel: [<c0152bc2>] ? lock_acquire+0x71/0x94 Oct 31 10:15:56 meowmix kernel: [<c013ae4a>] ? force_sig_info+0x25/0xbd Oct 31 10:15:56 meowmix kernel: [<c063d235>] ? _spin_lock_irqsave+0x61/0x8e Oct 31 10:15:56 meowmix kernel: [<c013ae4a>] ? force_sig_info+0x25/0xbd Oct 31 10:15:56 meowmix kernel: [<c013ae4a>] ? force_sig_info+0x25/0xbd Oct 31 10:15:56 meowmix kernel: [<c013aef6>] ? force_sig+0x14/0x16 Oct 31 10:15:56 meowmix kernel: [<c0302f66>] ? cifs_mount+0x991/0x2358 Oct 31 10:15:56 meowmix kernel: [<c014f09c>] ? trace_hardirqs_off_caller+0x15/0x98 Oct 31 10:15:56 meowmix kernel: [<c0151982>] ? trace_hardirqs_on+0xb/0xd Oct 31 10:15:56 meowmix kernel: [<c015193e>] ? trace_hardirqs_on_caller+0x10a/0x143 Oct 31 10:15:56 meowmix kernel: [<c02f6d68>] ? cifs_get_sb+0x89/0x1b1 Oct 31 10:15:56 meowmix kernel: [<c01a09fd>] ? vfs_kern_mount+0x5e/0x11c Oct 31 10:15:56 meowmix kernel: [<c01a0b0f>] ? do_kern_mount+0x38/0xc2 Oct 31 10:15:56 meowmix kernel: [<c01b701b>] ? do_new_mount+0x5e/0x7c Oct 31 10:15:56 meowmix kernel: [<c01b7540>] ? do_mount+0x1ce/0x1e4 Oct 31 10:15:56 meowmix kernel: [<c0103f16>] ? restore_nocheck_notrace+0x0/0xe Oct 31 10:15:56 meowmix kernel: [<c051332f>] ? _raw_spin_trylock+0xf/0x31 Oct 31 10:15:56 meowmix kernel: [<c01b75db>] ? sys_mount+0x85/0xba Oct 31 10:15:56 meowmix kernel: [<c0103e2f>] ? sysenter_do_call+0x12/0x43 Oct 31 10:15:56 meowmix kernel: ======================= Oct 31 10:15:56 meowmix kernel: Code: 2b 01 00 00 9c 58 90 8d b4 26 00 00 00 00 f6 c4 02 0f 85 13 03 00 00 83 fa 07 0f 87 51 03 00 00 85 d2 0f 85 11 01 00 00 8b 4d b8 <8b> 59 04 85 db 0f 84 03 01 00 00 8b 7d bc 8b bf b8 04 00 00 89 Oct 31 10:15:56 meowmix kernel: EIP: [<c0151c35>] __lock_acquire+0x5d/0xf79 SS:ESP 0068:c3179c9c Oct 31 10:15:56 meowmix kernel: ---[ end trace f3f72a65f106052a ]--- Oct 31 10:15:56 meowmix kernel: note: mount.cifs[2973] exited with preempt_count 2 Oct 31 10:15:56 meowmix kernel: BUG: unable to handle kernel NULL pointer dereference at 00000504 Oct 31 10:15:56 meowmix kernel: IP: [<c0119718>] __ticket_spin_trylock+0x6/0x20 Oct 31 10:15:56 meowmix kernel: *pde = 00000000 Oct 31 10:15:56 meowmix kernel: Oops: 0000 [#2] PREEMPT SMP Oct 31 10:15:56 meowmix kernel: Modules linked in: ipv6 dm_snapshot dm_mirror dm_log dm_mod it87 hwmon_vid sbp2 psmouse ide_disk ide_cd_mod cx88_dvb cx88_vp3054 _i2c mt352 s5h1409 dvb_pll s5h1411 or51132 videobuf_dvb nxt200x isl6421 zl10353 cx24123 lgdt330x dvb_core cx22702 tuner tea5767 tda8290 tda18271 tda827x tuner_x c2028 xc5000 tda9887 tuner_simple tuner_types mt20xx snd_intel8x0 tea5761 snd_ac97_codec v4l2_common ac97_bus snd_pcm_oss cx8802 snd_mixer_oss cx88xx videodev v 4l1_compat snd_pcm ir_common snd_timer i2c_algo_bit pcspkr tveeprom i2c_i801 rtc_cmos btcx_risc snd videobuf_dma_sg videobuf_core i2c_core soundcore iTCO_wdt sn d_page_alloc intel_agp agpgart button joydev evdev usb_storage sg usbhid hid ff_memless sr_mod cdrom sd_mod ide_pci_generic piix ahci ide_core ata_generic ohci1 394 ieee1394 floppy firewire_ohci firewire_core pata_acpi ata_piix r8169 libata scsi_mod dock ehci_hcd uhci_hcd e1000e usbcore thermal processor fan thermal_sys hwmon Oct 31 10:15:56 meowmix kernel: Oct 31 10:15:56 meowmix kernel: Pid: 2975, comm: mount.cifs Tainted: G D (2.6.27 #50) Oct 31 10:15:56 meowmix kernel: EIP: 0060:[<c0119718>] EFLAGS: 00010096 CPU: 0 Oct 31 10:15:56 meowmix kernel: EIP is at __ticket_spin_trylock+0x6/0x20 Oct 31 10:15:56 meowmix kernel: EAX: 00000504 EBX: 00000504 ECX: 00000000 EDX: 00000000 Oct 31 10:15:56 meowmix kernel: ESI: 00000504 EDI: 00000286 EBP: f66e4d10 ESP: f66e4d0c Oct 31 10:15:56 meowmix kernel: DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068 Oct 31 10:15:56 meowmix kernel: Process mount.cifs (pid: 2975, ti=f66e4000 task=f67b6680 task.ti=f66e4000) Oct 31 10:15:56 meowmix kernel: Stack: 00000514 f66e4d28 c051332f 00000000 c013ae4a 00000514 00000504 f66e4d4c Oct 31 10:15:56 meowmix kernel: c063d23c 00000000 00000002 00000000 c013ae4a c31e8b00 f6bc3d80 00000009 Oct 31 10:15:56 meowmix kernel: f66e4d78 c013ae4a c0dd4810 00000000 00000001 fffffffe c326b6c0 f66e4d80 Oct 31 10:15:56 meowmix kernel: Call Trace: Oct 31 10:15:56 meowmix kernel: [<c051332f>] ? _raw_spin_trylock+0xf/0x31 Oct 31 10:15:56 meowmix kernel: [<c013ae4a>] ? force_sig_info+0x25/0xbd Oct 31 10:15:56 meowmix kernel: [<c063d23c>] ? _spin_lock_irqsave+0x68/0x8e Oct 31 10:15:56 meowmix kernel: [<c013ae4a>] ? force_sig_info+0x25/0xbd Oct 31 10:15:56 meowmix kernel: [<c013ae4a>] ? force_sig_info+0x25/0xbd Oct 31 10:15:56 meowmix kernel: [<c013aef6>] ? force_sig+0x14/0x16 Oct 31 10:15:56 meowmix kernel: [<c03034da>] ? cifs_mount+0xf05/0x2358 Oct 31 10:15:56 meowmix kernel: [<c063fb55>] ? do_page_fault+0x1b5/0x970 Oct 31 10:15:56 meowmix kernel: [<c015184a>] ? trace_hardirqs_on_caller+0x16/0x143 Oct 31 10:15:56 meowmix kernel: [<c02f6d68>] ? cifs_get_sb+0x89/0x1b1 Oct 31 10:15:56 meowmix kernel: [<c01a09fd>] ? vfs_kern_mount+0x5e/0x11c Oct 31 10:15:56 meowmix kernel: [<c01a0b0f>] ? do_kern_mount+0x38/0xc2 Oct 31 10:15:56 meowmix kernel: [<c01b701b>] ? do_new_mount+0x5e/0x7c Oct 31 10:15:56 meowmix kernel: [<c01b7540>] ? do_mount+0x1ce/0x1e4 Oct 31 10:15:56 meowmix kernel: [<c0103f16>] ? restore_nocheck_notrace+0x0/0xe Oct 31 10:15:56 meowmix kernel: [<c051332f>] ? _raw_spin_trylock+0xf/0x31 Oct 31 10:15:56 meowmix kernel: [<c01b75db>] ? sys_mount+0x85/0xba Oct 31 10:15:56 meowmix kernel: [<c0103e2f>] ? sysenter_do_call+0x12/0x43 Oct 31 10:15:56 meowmix kernel: ======================= Oct 31 10:15:56 meowmix kernel: Code: 00 83 e8 01 0f 9f c0 0f b6 c0 5d c3 55 89 e5 ba 00 01 00 00 f0 66 0f c1 10 38 f2 74 06 f3 90 8a 10 eb f6 5d c3 55 89 e5 53 89 c3 <66> 8b 03 38 e0 75 0a 66 89 c1 fe c5 f0 66 0f b1 0b 0f 94 c1 0f Oct 31 10:15:56 meowmix kernel: EIP: [<c0119718>] __ticket_spin_trylock+0x6/0x20 SS:ESP 0068:f66e4d0c Oct 31 10:15:56 meowmix kernel: ---[ end trace f3f72a65f106052a ]--- Oct 31 10:15:56 meowmix kernel: note: mount.cifs[2975] exited with preempt_count 2 Some point after that I began receiving an endless number of these in the dmesg output: BUG: soft lockup - CPU#1 stuck for 61s! [kstop1:20486] Modules linked in: binfmt_misc ipt_MASQUERADE iptable_nat nf_nat nf_conntrack_ipv4 nf_conntrack ip_tables x_tables i915 drm rfcomm l2cap bluetooth tun ppdev par port_pc lp parport ac battery speedstep_lib cpufreq_userspace cpufreq_stats cpufreq_powersave cpufreq_ondemand freq_table cpufreq_conservative ipv6 dm_snapshot dm_mirror dm_log dm_mod it87 hwmon_vid sbp2 psmouse ide_disk ide_cd_mod cx88_dvb cx88_vp3054_i2c mt352 s5h1409 dvb_pll s5h1411 or51132 videobuf_dvb nxt200x isl6 421 zl10353 cx24123 lgdt330x dvb_core cx22702 tuner tea5767 tda8290 tda18271 tda827x tuner_xc2028 xc5000 tda9887 tuner_simple tuner_types mt20xx snd_intel8x0 te a5761 snd_ac97_codec v4l2_common ac97_bus snd_pcm_oss cx8802 snd_mixer_oss cx88xx videodev v4l1_compat snd_pcm ir_common snd_timer i2c_algo_bit pcspkr tveeprom i2c_i801 rtc_cmos btcx_risc snd videobuf_dma_sg videobuf_core i2c_core soundcore iTCO_wdt snd_page_alloc intel_agp agpgart button joydev evdev usb_storage sg us bhid hid ff_memless sr_mod cdrom sd_mod ide_pci_generic piix ahci ide_core ata_generic ohci1394 ieee1394 floppy firewire_ohci firewire_core pata_acpi ata_piix r 8169 libata scsi_mod dock ehci_hcd uhci_hcd e1000e usbcore thermal processor fan thermal_sys hwmon irq event stamp: 0 hardirqs last enabled at (0): [<00000000>] 0x0 hardirqs last disabled at (0): [<c012ca7b>] copy_process+0x4e8/0x11d3 softirqs last enabled at (0): [<c012ca99>] copy_process+0x506/0x11d3 softirqs last disabled at (0): [<00000000>] 0x0 Pid: 20486, comm: kstop1 Tainted: G D (2.6.27 #50) EIP: 0060:[<c015fc6d>] EFLAGS: 00000246 CPU: 1 EIP is at stop_cpu+0x67/0xaa EAX: 00000001 EBX: f7866f68 ECX: fffffffc EDX: c21f2900 ESI: 00000001 EDI: f7866f68 EBP: e7b24fd0 ESP: e7b24fc4 DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068 CR0: 8005003b CR2: b6db7000 CR3: 009a2000 CR4: 000006d0 DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000 DR6: ffff0ff0 DR7: 00000400 [<c015fc06>] ? stop_cpu+0x0/0xaa [<c0141826>] kthread+0x41/0x6b [<c01417e5>] ? kthread+0x0/0x6b [<c0104b6b>] kernel_thread_helper+0x7/0x10 =======================
Is this recreatable? What are the steps that lead up to this oops?
This may be due to the mount/umount races that were fixed last fall. Those *should* be fixed now in both >= 2.6.28 and in recent 2.6.27 stable kernels. Can you let us know whether you're still seeing this on more recent kernels?
I agree with Jeff - any lock related conflicts in this area were likely to have been cleared by the mount fixes which went into mainline and also into 2.6.27.9 (or later) stable kernel series. Any occurrences at that level or later (2.6.27.9 or later)?
The machine that I reproduced this on a couple times is now on 2.6.29-rc5 and I haven't seen the problem, but I haven't been using CIFS as much lately.
OK - please reopen this (or a new defect) if you see anything similar in the future. The locking in this area is much better and it is very likely the problem was resolved by that patch series.