2.6.10-rc2-bk7 - lots of "page allocation failure. order:0, mode:0x20"

[Christopher S. Aker]

Host: 2.6.10-rc2-bk7, SMP, 4GB RAM, about 200M into swap

# cat /proc/sys/vm/min_free_kbytes
10240

I've seen this reported elsewhere, but thought I'd provide a few more examples.  It
seems to happen mostly during scp'ing of large files to the machine. The network
config of this machine is Linux bridge, going through an e1000 card.  Increasing
min_free_kbytes doesn't seem to help.

Some examples:

uml-2.4: page allocation failure. order:0, mode:0x20
 [<c0136fa9>] __alloc_pages+0x1b7/0x35b
 [<c0137172>] __get_free_pages+0x25/0x3f
 [<c031069a>] ip_rcv_finish+0x211/0x28b
 [<c013a224>] kmem_getpages+0x21/0xc9
 [<c013aeb7>] cache_grow+0xab/0x14d
 [<c013b0d5>] cache_alloc_refill+0x17c/0x221
 [<c0310489>] ip_rcv_finish+0x0/0x28b
 [<c013b347>] kmem_cache_alloc+0x4b/0x4d
 [<c037d936>] br_nf_pre_routing+0x191/0x3fc
 [<c0379604>] br_handle_frame_finish+0x0/0x10f
 [<c0304025>] nf_iterate+0x71/0xa2
 [<c0379604>] br_handle_frame_finish+0x0/0x10f
 [<c0379604>] br_handle_frame_finish+0x0/0x10f
 [<c0304381>] nf_hook_slow+0x6b/0xf9
 [<c0379604>] br_handle_frame_finish+0x0/0x10f
 [<c03797f7>] br_handle_frame+0xe4/0x1c9
 [<c0379604>] br_handle_frame_finish+0x0/0x10f
 [<c02fa752>] netif_receive_skb+0x114/0x275
 [<c0113447>] finish_task_switch+0x3a/0x83
 [<c02fa931>] process_backlog+0x7e/0x10b
 [<c02faa35>] net_rx_action+0x77/0xf6
 [<c011cba7>] __do_softirq+0xb7/0xc6
 [<c011cbe3>] do_softirq+0x2d/0x2f
 [<c02fa51d>] netif_rx_ni+0x3b/0x3d
 [<c028230e>] tun_chr_writev+0x11c/0x18a
 [<c028237c>] tun_chr_write+0x0/0x3b
 [<c02823b3>] tun_chr_write+0x37/0x3b
 [<c0151200>] vfs_write+0xb0/0x119
 [<c0152086>] fget_light+0x84/0x86
 [<c015133a>] sys_write+0x51/0x80
 [<c010243b>] syscall_call+0x7/0xb

swapper: page allocation failure. order:0, mode:0x20
 [<c0136fa9>] __alloc_pages+0x1b7/0x35b
 [<c0137172>] __get_free_pages+0x25/0x3f
 [<c013a224>] kmem_getpages+0x21/0xc9
 [<c013aeb7>] cache_grow+0xab/0x14d
 [<c013b0d5>] cache_alloc_refill+0x17c/0x221
 [<c013b347>] kmem_cache_alloc+0x4b/0x4d
 [<c037d936>] br_nf_pre_routing+0x191/0x3fc
 [<c0379604>] br_handle_frame_finish+0x0/0x10f
 [<c0304025>] nf_iterate+0x71/0xa2
 [<c0379604>] br_handle_frame_finish+0x0/0x10f
 [<c0379604>] br_handle_frame_finish+0x0/0x10f
 [<c0304381>] nf_hook_slow+0x6b/0xf9
 [<c0379604>] br_handle_frame_finish+0x0/0x10f
 [<c03797f7>] br_handle_frame+0xe4/0x1c9
 [<c0379604>] br_handle_frame_finish+0x0/0x10f
 [<c02fa752>] netif_receive_skb+0x114/0x275
 [<c026bb24>] e1000_clean_rx_irq+0x12e/0x447
 [<c026b765>] e1000_clean+0x51/0xca
 [<c02faa35>] net_rx_action+0x77/0xf6
 [<c011cba7>] __do_softirq+0xb7/0xc6
 [<c011cbe3>] do_softirq+0x2d/0x2f
 [<c01046b6>] do_IRQ+0x1e/0x24
 [<c0102db2>] common_interrupt+0x1a/0x20
 [<c01005da>] mwait_idle+0x31/0x48
 [<c01005a0>] cpu_idle+0x33/0x3c

-Chris

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/