BUG: unable to handle kernel paging request at virtual address feededed (was: Re: Linux v2.6.17-rc5)

[Tomasz Torcz <zdzichu@irc.pl>]

  After 2 days and few hours uptime, during updatedb run I got:

BUG: unable to handle kernel paging request at virtual address feededed
 printing eip:
c0160fb1
*pde = 00000000
Oops: 0000 [#1]
Modules linked in: mga drm ipv6 sha256 dm_crypt binfmt_misc uhci_hcd i2c_nforce2 eth1394 ohci_hcd ehci_hcd forcedeth snd_intel8x0 snd_ac97_codec snd_ac97_bus sata_nv ohci1394 ieee1394 tuner tvaudio bttv video_buf firmware_class ir_common btcx_risc tveeprom e1000 udf usb_storage usbcore cpufreq_nforce2 lp parport w83627hf hwmon_vid i2c_isa
CPU:    0
EIP:    0060:[<c0160fb1>]    Not tainted VLI
EFLAGS: 00010206   (2.6.17-rc5 #63)
EIP is at __d_find_alias+0x1c/0xa2
eax: 00008000   ebx: feededed   ecx: feededed   edx: cfe98954
esi: cde98918   edi: cde98918   ebp: cfe9893c   esp: c4d27d10
ds: 007b   es: 007b   ss: 0068
Process updatedb (pid: 18083, threadinfo=c4d27000 task=d92b7570)
Stack: 00000001 cfe9893c dc3ed190 00000000 d509e42c c0161825 00000001 c4d27d3c
       cfe9893c c0180050 dc3ed190 00000001 00000000 00000000 00000000 00000000
       00000000 df4d39f4 00000000 df4d39c4 00000070 c66b8154 00000003 c599e784
Call Trace:
 <c0161825> d_splice_alias+0x17/0x86  <c0180050> reiserfs_lookup+0xe2/0xf1
 <c01597a3> real_lookup+0x53/0xad  <c01599c0> do_lookup+0x49/0x78
 <c015a00c> __link_path_walk+0x61d/0x9d6  <c0191d9b> pathrelse+0x1b/0x26
 <c0188909> reiserfs_readdir+0x3db/0x3ea  <c015a406> link_path_walk+0x41/0xaa
 <c01de7cc> strncpy_from_user+0x2d/0x4c  <c015a770> do_path_lookup+0x199/0x1e4
 <c015a9cb> __user_walk_fd+0x29/0x3a  <c0156448> vfs_lstat_fd+0x12/0x39
 <c01569d4> sys_lstat64+0xf/0x23  <c0102933> syscall_call+0x7/0xb
Code: 89 50 04 89 02 89 5b 04 89 59 24 5b 89 c8 c3 55 89 c5 57 56 31 f6 53 51 89 14 24 8b 48 18 8d 50 18 39 d1 0f 84 80 00 00 00 89 cb <8b> 09 0f 18 01 90 0f b7 45 28 8d 7b c4 25 00 f0 00 00 3d 00 40
EIP: [<c0160fb1>] __d_find_alias+0x1c/0xa2 SS:ESP 0068:c4d27d10
 BUG: updatedb/18083, lock held at task exit time!
 [d509e49c] {inode_init_once}
.. held by:          updatedb:18083 [d92b7570, 118]
... acquired at:               real_lookup+0x1c/0xad


 It never happened before. d_splice_alias in bt is very strange, as I don't
think anything on my system uses splice(). It's too new, and my system is
Slackware -current (which seems to return ENOSUPORTED even for old stuff
like posix_fadvise()).
  Anyway, my hardware is Sempron CPU on NForce2 MB. Seagate SATA disk
driven by sata_sil. LVM2 with few volumes, one of them (my ~) on
dm-crypt, but rootfs is not crypted. All filesystems are reiserfs mounted
with "user_xattr,acl". I will provide more info if needed. .config
attached.

-- 
Tomasz Torcz                 "God, root, what's the difference?"
zdzichu@irc.-nie.spam-.pl         "God is more forgiving."

Attachment: config.gz
Description: Binary data

Attachment: pgpHKQllfjpRw.pgp
Description: PGP signature