oops in 2.6.4rc2+preempt

Viktors Rotanovs viktors at rotanovs.com
Wed Mar 17 14:27:01 UTC 2004


Hi,

I got the following on 2.6.4rc2+preempt with TUX patch made for 2.6.3 after 
load increased on one of our servers. I know that this combination is highly 
experemental and most likely it's my fault, I'm posting this just because it 
may be interesting for someone.

Unable to handle kernel NULL pointer dereference at virtual address 00000115
 printing eip:
c0241f41
*pde = 00000000
Oops: 0002 [#1]
PREEMPT
CPU:    0
EIP:    0060:[<c0241f41>]    Not tainted
EFLAGS: 00010293
EIP is at do_tcp_sendpages+0x191/0xac0
eax: 00000000   ebx: e5451c00   ecx: 00000111   edx: 00000111
esi: 00000000   edi: dca774c0   ebp: ddca1080   esp: ec125e70
ds: 007b   es: 007b   ss: 0068
Process tux (pid: 1503, threadinfo=ec124000 task=ec9cb410)
Stack: ec125ea8 c010a3ef 00000000 00000014 dd791ef4 ea581d2c ea581c00 ddca10d4
       00000111 00000076 00000111 00000000 000005b4 ddca124c 00001076 ec125ed0
       00000000 00001076 ddca1080 0000c040 00000111 c02428e2 00000111 0000c040
Call Trace:
 [<c010a3ef>] handle_IRQ_event+0x2f/0x60
 [<c02428e2>] tcp_sendpage+0x72/0x80
 [<c028a7d4>] do_send_abuf+0x54/0x160
 [<c027b9fc>] tux_schedule_atom+0xc/0x10
 [<c027c7da>] process_requests+0xaa/0xd0
 [<c0287f19>] event_loop+0x129/0x190
 [<c028971d>] __sys_tux+0x21d/0xb20
 [<c015267a>] path_release+0xa/0x30
 [<c014560d>] sys_chdir+0x5d/0x70
 [<c0108c67>] syscall_call+0x7/0xb

Code: ff 42 04 8b 7c 24 28 8b 83 a4 00 00 00 8d 04 f0 8d 50 10 89
 <5>TUX: thread 0 stopping ...



Best Wishes,
Viktors





More information about the tux-list mailing list