On Mon, 2009-08-17 at 14:40 -0700, Robert Fleming wrote: > Package: linux-image-2.6.26-2-686 > Version: 2.6.26-17lenny2 > Severity: normal > > I have this in my /var/log/kern.log: > Aug 17 11:54:27 mizar kernel: [8806733.883881] Not cloning cgroup for unused > subsystem ns > Aug 17 11:55:38 mizar kernel: [8806840.228151] ------------[ cut here > ]------------ > Aug 17 11:55:38 mizar kernel: [8806840.228160] kernel BUG at > kernel/exit.c:822! > Aug 17 11:55:38 mizar kernel: [8806840.228164] invalid opcode: 0000 [#1] SMP > Aug 17 11:55:38 mizar kernel: [8806840.228170] Modules linked in: udf > crc_itu_t visor usbserial ext2 xt_mac ipt_MASQUERADE iptable_nat nf_nat > nf_conntrack_ipv4 nf_conntrack nls_utf8 nls_cp437 vfat fat usb_storage isofs > nls_base zlib_inflate xt_multiport iptable_filter ip_tables x_tables > binfmt_misc nvidia(P) agpgart rfcomm l2cap ppdev lp autofs4 tun ipv6 nfsd > auth_rpcgss exportfs nfs lockd nfs_acl sunrpc fuse dm_snapshot dm_mirror > dm_log dm_mod sbp2 loop snd_intel8x0 snd_ac97_codec ac97_bus snd_pcm_oss > snd_mixer_oss snd_pcm snd_seq_dummy snd_seq_oss snd_seq_midi snd_rawmidi > snd_seq_midi_event snd_seq snd_timer snd_seq_device parport_pc parport snd > button soundcore hci_usb rng_core pcspkr i2c_i801 snd_page_alloc bluetooth > iTCO_wdt e752x_edac i2c_core shpchp pci_hotplug edac_core evdev dcdbas ext3 > jbd mbcache usbhid hid ff_memless raid1 md_mod ide_cd_mod cdrom sd_mod piix > ide_pci_generic ide_core floppy ohci1394 ata_piix ieee1394 ata_generic libata > scsi_mod dock ehci_hcd uhci_! > hcd usbcore e1000 thermal processor > Aug 17 11:55:38 mizar kernel: an thermal_sys > Aug 17 11:55:38 mizar kernel: [8806840.228301] > Aug 17 11:55:38 mizar kernel: [8806840.228306] Pid: 17496, comm: chrome > Tainted: P (2.6.26-2-686 #1) > Aug 17 11:55:38 mizar kernel: [8806840.228311] EIP: 0060:[<c012527d>] EFLAGS: > 00210287 CPU: 0 > Aug 17 11:55:38 mizar kernel: [8806840.228320] EIP is at do_exit+0x40c/0x5bb > Aug 17 11:55:38 mizar kernel: [8806840.228324] EAX: f74c0780 EBX: f74c0658 > ECX: f74c0780 EDX: f74c073c > Aug 17 11:55:38 mizar kernel: [8806840.228328] ESI: f74c0660 EDI: f74c0658 > EBP: f74c0660 ESP: f221bf80 > Aug 17 11:55:38 mizar kernel: [8806840.228332] DS: 007b ES: 007b FS: 00d8 > GS: 0000 SS: 0068 > Aug 17 11:55:38 mizar kernel: [8806840.228336] Process chrome (pid: 17496, > ti=f221a000 task=f74c0660 task.ti=f221a000) > Aug 17 11:55:38 mizar kernel: [8806840.228340] Stack: 00000000 00000000 > f221bf88 f221bf88 d047f380 f7492d00 00000000 f221a000 > Aug 17 11:55:38 mizar kernel: [8806840.228351] c0125490 00000000 > bf8bf594 bf8bfbfb c01254c6 c0103853 00000000 00000000 > Aug 17 11:55:38 mizar kernel: [8806840.228361] 00000000 bf8bf594 > bf8bfbfb bf8bf9c8 000000fc 0000007b 0000007b 00000000 > Aug 17 11:55:38 mizar kernel: [8806840.228371] Call Trace: > Aug 17 11:55:38 mizar kernel: [8806840.228400] [<c0125490>] > do_group_exit+0x64/0x8d > Aug 17 11:55:38 mizar kernel: [8806840.228417] [<c01254c6>] > sys_exit_group+0xd/0x10 > Aug 17 11:55:38 mizar kernel: [8806840.228423] [<c0103853>] > sysenter_past_esp+0x78/0xb1 > Aug 17 11:55:38 mizar kernel: [8806840.228482] ======================= > Aug 17 11:55:38 mizar kernel: [8806840.228484] Code: dc 00 00 00 39 c2 75 ce > f0 81 05 00 ea 36 c0 00 00 00 01 fb 0f 1f 84 00 00 00 00 00 90 8d 86 20 01 00 > 00 39 86 20 01 00 00 74 04 <0f> 0b eb fe 39 96 dc 00 00 00 74 04 0f 0b eb fe > 8b 5c 24 08 81 > Aug 17 11:55:38 mizar kernel: [8806840.228541] EIP: [<c012527d>] > do_exit+0x40c/0x5bb SS:ESP 0068:f221bf80 > Aug 17 11:55:38 mizar kernel: [8806840.228558] ---[ end trace > 4090fef249e8d577 ]--- > Aug 17 11:55:38 mizar kernel: [8806840.228562] Fixing recursive fault but > reboot is needed! > > At the moment, the process mentioned is still showing up in the output of > "ps": > > # ps -ef|grep 17496 > flemingr 17496 1 0 11:54 ? 00:00:00 [chrome] > > But the chrome web browser is no-where to be found on-screen. Not sure what > provoked this. [...]
The browser tried to exit but hit a kernel bug so the process was not
properly terminated.
Can you try to reproduce this without the nvidia module loaded?
Ben.
--
Ben Hutchings
Power corrupts. Absolute power is kind of neat.
- John Lehman, Secretary of the US Navy 1981-1987
signature.asc
Description: This is a digitally signed message part
