Home » Mailing list archives » linux-kernel » 2007 » September » 4

Re: kernel BUG at mm/slab.c:2980 (was Re: [<c019c63f>] xfs_bmap_search_multi_extents+0x6f/0xe0)

Previous thread: making local traffic to pass through bridge by rajneesh rana on Tuesday, September 4, 2007 - 8:19 am. (1 message)

Next thread: ICH Intel PATA short cable override... by Daniel J Blueman on Tuesday, September 4, 2007 - 8:37 am. (8 messages)
[view original message]
From: Marco Berizzi <pupilla@...>
To: Christoph Lameter <clameter@...>
Cc: <linux-kernel@...>, David Chinner <dgc@...>, <xfs@...>, Marco Berizzi <pupilla@...>
Subject: Re: kernel BUG at mm/slab.c:2980 (was Re: [&lt;c019c63f&gt;] xfs_bmap_search_multi_extents+0x6f/0xe0)
Date: Tuesday, September 4, 2007 - 8:22 am

Hi Christoph,


I have upgraded to 2.6.22.5 and I have selected

the SLUB. I have also added append=slub_debug to

lilo.conf

After a week uptime I got this error. I hope it

will be useful for you.


Linux version 2.6.22.5 (root@Mimosa) (gcc version 3.3.5) #1 Mon Aug 27

16:57:18 CEST 2007

BIOS-provided physical RAM map:

 BIOS-e820: 0000000000000000 - 000000000009f800 (usable)

 BIOS-e820: 000000000009f800 - 00000000000a0000 (reserved)

 BIOS-e820: 00000000000dc000 - 00000000000e0000 (reserved)

 BIOS-e820: 00000000000f0000 - 0000000000100000 (reserved)

 BIOS-e820: 0000000000100000 - 000000000a000000 (usable)

 BIOS-e820: 00000000ffff0000 - 0000000100000000 (reserved)

160MB LOWMEM available.

Entering add_active_range(0, 0, 40960) 0 entries of 256 used

Zone PFN ranges:

  DMA             0 ->     4096

  Normal       4096 ->    40960

early_node_map[1] active PFN ranges

    0:        0 ->    40960

On node 0 totalpages: 40960

  DMA zone: 32 pages used for memmap

  DMA zone: 0 pages reserved

  DMA zone: 4064 pages, LIFO batch:0

  Normal zone: 288 pages used for memmap

  Normal zone: 36576 pages, LIFO batch:7

DMI 2.1 present.

Allocating PCI resources starting at 10000000 (gap: 0a000000:f5ff0000)

Built 1 zonelists.  Total pages: 40640

Kernel command line: auto BOOT_IMAGE=Linux ro root=301 slub_debug

Local APIC disabled by BIOS -- you can enable it with "lapic"

mapped APIC to ffffd000 (01141000)

Enabling fast FPU save and restore... done.

Initializing CPU#0

PID hash table entries: 1024 (order: 10, 4096 bytes)

Detected 267.281 MHz processor.

Console: colour VGA+ 80x25

Dentry cache hash table entries: 32768 (order: 5, 131072 bytes)

Inode-cache hash table entries: 16384 (order: 4, 65536 bytes)

Memory: 158956k/163840k available (1975k kernel code, 4452k reserved,

628k data, 160k init, 0k highmem)

virtual kernel memory layout:

    fixmap  : 0xfffb7000 - 0xfffff000   ( 288 kB)

    vmalloc : 0xca800000 - 0xfffb5000   ( 855 MB)

    lowmem  : 0xc0000000 - 0xca000000   ( 160 MB)

   ...
[ message continues ]
[view original message]
From: Christoph Lameter <clameter@...>
To: Marco Berizzi <pupilla@...>
Cc: <linux-kernel@...>, David Chinner <dgc@...>, <xfs@...>, <mel@...>
Subject: Re: kernel BUG at mm/slab.c:2980 (was Re: [&lt;c019c63f&gt;] xfs_bmap_search_multi_extents+0x6f/0xe0)
Date: Tuesday, September 4, 2007 - 12:58 pm

Yes indeed but this is a different type of failure. Looks like a higher

allocation failure in the networking code. Someone created objects that


tcp_collapse? This is due to a network configuration that required an

order 2 kmalloc block. Jumbo frames?


-
Previous thread: making local traffic to pass through bridge by rajneesh rana on Tuesday, September 4, 2007 - 8:19 am. (1 message)

Next thread: ICH Intel PATA short cable override... by Daniel J Blueman on Tuesday, September 4, 2007 - 8:37 am. (8 messages)