Skip to contentSkip to conversation

2.5.60-bk pdflush oops.

DISCUSSION Dave Jones 3 participants 5 messages Feb 13, 2003

lore
Share & tools
lore

Dave Jones

author
#0 lore raw

Bitkeeper pull from ~5 hrs ago.

Something went splat just after booting.
I think this may have happened as I mounted an NFS mount. Hard to tell, but the box booted at 20:30, this happened at 20:37, and I started NFS testing at 20:40 which was when I noticed it.

Feb 13 20:37:41 mesh kernel:  printing eip:
Feb 13 20:37:41 mesh kernel: c012e276
Feb 13 20:37:41 mesh kernel: Oops: 0002
Feb 13 20:37:41 mesh kernel: CPU:    0
Feb 13 20:37:41 mesh kernel: EIP:    0060:[<c012e276>]    Not tainted
Feb 13 20:37:41 mesh kernel: EFLAGS: 00010046
Feb 13 20:37:41 mesh kernel: EIP is at mod_timer+0x96/0x7e0
Feb 13 20:37:41 mesh kernel: eax: 00000000   ebx: c0147440   ecx: 00000007   edx: 00001388
Feb 13 20:37:41 mesh kernel: esi: 850fc085   edi: c06436c0   ebp: c11c7ed0   esp: c11c7ea0
Feb 13 20:37:41 mesh kernel: ds: 007b   es: 007b   ss: 0068
Feb 13 20:37:41 mesh kernel: Process pdflush (pid: 5, threadinfo=c11c6000 task=c113d980)
Feb 13 20:37:41 mesh kernel: Stack: c11c7ec0 c0146229 c11c7f04 00000000 00000080 c0147320 c11c7ee4 00000000 
Feb 13 20:37:41 mesh kernel:        00000297 000733cf c11c7ee4 00000000 c11c7f90 c0147426 c06436c0 000733cf 
Feb 13 20:37:41 mesh kernel:        0006ab17 00000000 00000000 c11c7ee0 00000000 00000001 00000000 00000001 
Feb 13 20:37:41 mesh kernel: Call Trace:
Feb 13 20:37:41 mesh kernel:  [<c0146229>] __get_page_state+0x29/0x90
Feb 13 20:37:41 mesh kernel:  [<c0147320>] wb_kupdate+0x0/0x120
Feb 13 20:37:41 mesh kernel:  [<c0147426>] wb_kupdate+0x106/0x120
Feb 13 20:37:41 mesh kernel:  [<c0147320>] wb_kupdate+0x0/0x120
Feb 13 20:37:41 mesh kernel:  [<c0147cf9>] __pdflush+0x259/0x5b0
Feb 13 20:37:41 mesh kernel:  [<c011e5fc>] schedule_tail+0x9c/0xe0
Feb 13 20:37:41 mesh kernel:  [<c0107d98>] __switch_to+0x148/0x150
Feb 13 20:37:41 mesh kernel:  [<c0148050>] pdflush+0x0/0x20
Feb 13 20:37:41 mesh kernel:  [<c0148061>] pdflush+0x11/0x20
Feb 13 20:37:41 mesh kernel:  [<c0147320>] wb_kupdate+0x0/0x120
Feb 13 20:37:41 mesh kernel:  [<c01075cd>] kernel_thread_helper+0x5/0x18
Feb 13 20:37:41 mesh kernel: 
Feb 13 20:37:41 mesh kernel: Code: 1c 52 fb 50 c0 c7 47 20 e1 00 00 00 8d b4 26 00 00 00 00 8d 
Feb 13 20:37:41 mesh kernel:  <6>note: pdflush[5] exited with preempt_count 1

Looking back through the logs, this also this bizarre snippet during boot:-

Feb 13 20:30:24 mesh kernel: Checking if this processor honours the WP bit even in supervisor mode... Ok.
Feb 13 20:30:24 mesh kernel: Call Trace:
Feb 13 20:30:24 mesh kernel:  [<c014a8b4>] kmem_cache_alloc+0x134/0x140
Feb 13 20:30:24 mesh kernel:  [<c014916f>] kmem_cache_create+0xbf/0x5a0
Feb 13 20:30:24 mesh kernel:  [<c0105000>] _stext+0x0/0x30
Feb 13 20:30:24 mesh kernel: 
Feb 13 20:30:24 mesh kernel: Dentry cache hash table entries: 16384 (order: 5, 131072 bytes)

Looks like part of 'something', but the other bit is nowhere to be seen. Odd.

Dave

Signature · 1 line
| Dave Jones.        http://www.codemonkey.org.uk
#1 lore raw
Quoted 4 lines

Looking back through the logs, this also this bizarre snippet during boot:-

Feb 13 20:30:24 mesh kernel: Checking if this processor honours the WP bit ev

en in supervisor mode... Ok.

Quoted 6 lines
Feb 13 20:30:24 mesh kernel: Call Trace:
Feb 13 20:30:24 mesh kernel:  [<c014a8b4>] kmem_cache_alloc+0x134/0x140
Feb 13 20:30:24 mesh kernel:  [<c014916f>] kmem_cache_create+0xbf/0x5a0
Feb 13 20:30:24 mesh kernel:  [<c0105000>] _stext+0x0/0x30
Feb 13 20:30:24 mesh kernel: 
Feb 13 20:30:24 mesh kernel: Dentry cache hash table entries: 16384 (order: 5

, 131072 bytes)

I caught this one too:

Memory: 255056k/262024k available (2010k kernel code, 6248k reserved, 895k data,
104k init, 0k highmem)
Debug: sleeping function called from illegal context at mm/slab.c:1618

Call Trace:
 [<c0135353>] kmem_cache_alloc+0x51/0x55
 [<c013490f>] kmem_cache_create+0x6c/0x448
 [<c0105000>] _stext+0x0/0x22

Security Scaffold v1.0.0 initialized
Dentry cache hash table entries: 32768 (order: 6, 262144 bytes)

So it's after test_wp_bit() is called, and before security_scaffolding_startup().

Interesting that you didn't get the 'sleeping function called' message?

#3 lore raw
In reply to Dave Jones
Quoted 8 lines · Dave Jones

Bitkeeper pull from ~5 hrs ago.

Something went splat just after booting.
I think this may have happened as I mounted an NFS mount. Hard to tell, but the box booted at 20:30, this happened at 20:37, and I started NFS testing at 20:40 which was when I noticed it.

There is missing text here as well.

        if (address < PAGE_SIZE)
                printk(KERN_ALERT "Unable to handle kernel NULL pointer dereference");
        else
                printk(KERN_ALERT "Unable to handle kernel paging request");
        printk(" at virtual address %08lx\n",address);
        printk(" printing eip:\n");
Quoted 9 lines
Feb 13 20:37:41 mesh kernel:  printing eip:
Feb 13 20:37:41 mesh kernel: c012e276
Feb 13 20:37:41 mesh kernel: Oops: 0002
Feb 13 20:37:41 mesh kernel: CPU:    0
Feb 13 20:37:41 mesh kernel: EIP:    0060:[<c012e276>]    Not tainted
Feb 13 20:37:41 mesh kernel: EFLAGS: 00010046
Feb 13 20:37:41 mesh kernel: EIP is at mod_timer+0x96/0x7e0
Feb 13 20:37:41 mesh kernel: eax: 00000000   ebx: c0147440   ecx: 00000007   edx: 00001388
Feb 13 20:37:41 mesh kernel: esi: 850fc085   edi: c06436c0   ebp: c11c7ed0   esp: c11c7ea0
Quoted 8 lines
Feb 13 20:37:41 mesh kernel: ds: 007b   es: 007b   ss: 0068
Feb 13 20:37:41 mesh kernel: Process pdflush (pid: 5, threadinfo=c11c6000 task=c113d980)
Feb 13 20:37:41 mesh kernel: Stack: c11c7ec0 c0146229 c11c7f04 00000000 00000080 c0147320 c11c7ee4 00000000 
Feb 13 20:37:41 mesh kernel:        00000297 000733cf c11c7ee4 00000000 c11c7f90 c0147426 c06436c0 000733cf 
Feb 13 20:37:41 mesh kernel:        0006ab17 00000000 00000000 c11c7ee0 00000000 00000001 00000000 00000001 
Feb 13 20:37:41 mesh kernel: Call Trace:
Feb 13 20:37:41 mesh kernel:  [<c0146229>] __get_page_state+0x29/0x90
Feb 13 20:37:41 mesh kernel:  [<c0147320>] wb_kupdate+0x0/0x120

That is a statically stored timer. It would appear that some other timer from some other random part of the kernel has got itself scribbled on.

Maybe networking?

author
#4 lore raw
Quoted 4 lines · Andrew Morton

That is a statically stored timer. It would appear that some other timer from some other random part of the kernel has got itself scribbled on.

Maybe networking?

Possible given this is the box I'm seeing that NFS bug on. Could even be both the same bug perhaps.

Dave

Signature · 2 lines
| Dave Jones.        http://www.codemonkey.org.uk
| SuSE Labs
author
#2 lore raw
In reply to [email protected]
Quoted 10 lines · On Thu, Feb 13, 2003 at 04:14:48PM -0500, [email protected] wrote:
Quoted 6 lines
Feb 13 20:30:24 mesh kernel: Call Trace:
Feb 13 20:30:24 mesh kernel:  [<c014a8b4>] kmem_cache_alloc+0x134/0x140
Feb 13 20:30:24 mesh kernel:  [<c014916f>] kmem_cache_create+0xbf/0x5a0
Feb 13 20:30:24 mesh kernel:  [<c0105000>] _stext+0x0/0x30
Feb 13 20:30:24 mesh kernel: 
Feb 13 20:30:24 mesh kernel: Dentry cache hash table entries: 16384 (order: 5

, 131072 bytes)
So it's after test_wp_bit() is called, and before security_scaffolding_startup().

Interesting that you didn't get the 'sleeping function called' message?

Could be that one went out to serial console, but not to the logs for some unknown reason. Unfortunatly my serial terminal has no docs, and the 'scroll backwards' keys are somewhat not so obvious.

Dave

Signature · 2 lines
| Dave Jones.        http://www.codemonkey.org.uk
| SuSE Labs