From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.9)
Gecko/20020408

Description of problem:
I am seeing this bug a lots of times. This is becos the dentry has
junk pointer.

Unable to handle kernel paging request at virtual address 80badb9b /g
printing eip:

c016654b
*pde = 1165b001
Oops: 0000 [#1]
SMP
Modules linked in: panfs nfsd exportfs ipv6 parport_pc lp parport nfs
lockd sunrpc e100 mii floppy sg microcode dm_mod ohci_hcd button
battery asus_acpi ac aic7xxx sd_mod scsi_mod
CPU: 0
EIP: 0060:[<c016654b>] Tainted: P
EFLAGS: 00010282 (2.6.6-1.435.2.3.lair6smp)
EIP is at __d_lookup+0x67/0x10b
eax: 80badb9b ebx: dd1ad720 ecx: 818d485a edx: c1520000
esi: d8047f2c edi: d8047ee0 ebp: 80badb9b esp: d8047e84
ds: 007b es: 007b ss: 0068
rocess perl (pid: 20032, threadinfo=d8044000 task=d981c750

MStack: 00000000 c1536738 d2e90015 818d485a 00000009 d8047ee0 dd3b9608
818d485a
d8047f2c d8047ee0 d8047ed8 c015d7dc dff738e0 818d485a fffffdee c5683320
d8047f2c c015de8a 00000001 00000000 d2e9001e dff738e0 dd3b9608 818d485a
Call Trace:
[<c015d7dc>] do_lookup+0x18/0x72
[<c015de8a>] link_path_walk+0x654/0x874
[<c015e397>] path_lookup+0x13f/0x16f
[<c015e4d3>] __user_walk+0x21/0x51
[<c0159ef7>] vfs_stat+0x14/0x3a
[<c011d4ae>] scheduler_tick+0x28f/0x425
[<c015a4d7>] sys_stat64+0xf/0x23

<c0117383>] smp_apic_timer_interrupt+0x124/0x129
[<c02c55cb>] syscall_call+0x7/0xb
[<c02c007b>] unix_stream_recvmsg+0x6/0x3af

Code: 8b 45 00 0f 18 00 90 8d 5d 90 f0 83 44 24 00 00 8b 4c 24 0c




 (gdb) bt  
#0  0xc0169bd0 in __d_lookup (parent=0xec3f4220, name=0xd473bf78) at
processor.h:660  
#1  0xc01608ee in cached_lookup (parent=0xec3f4220, name=0xd473bf78,
nd=0x0)  
    at fs/namei.c:287  
#2  0xc01617fe in __lookup_hash (name=0xd473bf78, base=0xec3f4220,
nd=0x0) at fs/namei.c:958  
#3  0xc01621df in lookup_create (nd=0xd473bf70, is_dir=1) at
fs/namei.c:1472  
#4  0xc01625f5 in sys_mkdir (pathname=0x4d4d4d4d <Address 0x4d4d4d4d
out of bounds>,  
    mode=511) at fs/namei.c:1598  
#5  0xc02d241f in syscall_call () at uaccess.h:516  
#6  0x09f007a0 in ?? ()  
#7  0x000001ff in ?? ()  
#8  0x0000013c in ?? ()  
#9  0x09f005f8 in ?? ()  
#10 0xb8097060 in ?? ()  
#11 0xbff22728 in ?? ()  
#12 0x00000027 in ?? ()  
#13 0xc02d007b in .text.lock.af_packet () at net/packet/af_packet.c:1858  
  
 
gdb) fr 0 
#0  0xc0169bd0 in __d_lookup (parent=0xec3f4220, name=0xd473bf78) at
processor.h:660 
660     in processor.h 
(gdb) p *parent 
$1 = {d_count = {counter = 3}, d_flags = 8, d_lock = {lock = 1},
d_inode = 0xc5ecc080, 
  d_parent = 0xec3f42bc, d_bucket = 0xf7f4e9f8, d_name = {hash =
282957070, 
    name = 0xec3f4298 "d_Rlk", len = 5}, d_lru = {next = 0xec3f42e0,
prev = 0xec3f41a8}, 
  d_child = {next = 0xec3f42f0, prev = 0xca97bad4}, d_subdirs = {next
= 0xdc774de0, 
    prev = 0xec3f41b0}, d_alias = {next = 0xc5ecc090, prev = 0xc5ecc090}, 
  d_time = 1109008540, d_op = 0xfa70ac54, d_sb = 0xf71ede00, d_mounted
= 0, d_fsdata = 0x0, 
  d_extra_attributes = 0x0, d_rcu = {list = {next = 0x15e82404, prev =
0x8dffffe9}, 
    func = 0xffbf3085, arg = 0x240489ff}, d_cookie = 0x0, d_hash =
{next = 0xf5fede24, 
    pprev = 0xd830f290}, d_iname = "d_Rlk\000\022���\215e�\211�[^_��
\203�\bh��:\b�u\b����"} 
(gdb) p name 
$2 = (struct qstr *) 0xd473bf78 
(gdb) p *name 
$3 = {hash = 440507695, name = 0xe4b58072 "d_oNRQvamNStrgO", len = 15} 
(gdb) p found 
$4 = (struct dentry *) 0x0 
(gdb) p str 
$5 = (const unsigned char *) 0xe4b58072 "d_oNRQvamNStrgO" 
(gdb) p head 
$8 = {0x0 <repeats 32 times>}


I saw a few links with related issues as 
http://www.redhat.com/archives/fedora-devel-list/2004-November/msg01279.html

http://www.ussg.iu.edu/hypermail/linux/kernel/0407.3/1187.html

http://www.thisishull.net/archive/index.php/t-31212.html

https://bugzilla.redhat.com/beta/show_bug.cgi?id=128670

http://www.webservertalk.com/message333123-4.html

http://lkml.org/lkml/2004/10/1/101

Is there any patch available ?

Version-Release number of selected component (if applicable):
kernel  2.6.6-1.435.2.3

How reproducible:
Always

Steps to Reproduce:
1. Run a few compiles annd the machine crashes after a few hrs
2.
3.
    

Actual Results:  crashes

Expected Results:  work fine

Additional info: