[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#666021: [powerpc] Kernel reports page allocation failure: order:1, mode:0x20



I installed the 3.2.12-1 and within within 2 hours around midnight there was this:

Mar 27 23:47:40 addict kernel: [    0.000000] Linux version 3.2.0-2-powerpc64 (Debian 3.2.12-1) (debian-kernel@lists.debian.org) (gcc version 4.6.2 (Debian 4.6.2-11) ) #1 SMP Tue Mar 20 20:35:04 UTC 2012
...
Mar 28 00:37:11 addict kernel: [ 3010.104653] swapper/0: page allocation failure: order:1, mode:0x20
Mar 28 00:37:11 addict kernel: [ 3010.108757] Call Trace:
Mar 28 00:37:11 addict kernel: [ 3010.112765] [c00000000ffff350] [c0000000000134f4] .show_stack+0x80/0x130 (unreliable)
Mar 28 00:37:11 addict kernel: [ 3010.121163] [c00000000ffff400] [c000000000119104] .warn_alloc_failed+0xf0/0x108
Mar 28 00:37:11 addict kernel: [ 3010.130015] [c00000000ffff4c0] [c00000000011c3d4] .__alloc_pages_nodemask+0x700/0x7c4
Mar 28 00:37:11 addict kernel: [ 3010.139288] [c00000000ffff650] [c0000000001589b8] .kmem_getpages+0x5c/0x140
Mar 28 00:37:11 addict kernel: [ 3010.144125] [c00000000ffff6f0] [c000000000158cdc] .fallback_alloc+0x174/0x200
Mar 28 00:37:11 addict kernel: [ 3010.148945] [c00000000ffff7d0] [c000000000159bf8] .__kmalloc+0x148/0x248
Mar 28 00:37:11 addict kernel: [ 3010.153672] [c00000000ffff890] [c0000000003c8de8] .pskb_expand_head+0xec/0x2cc
Mar 28 00:37:11 addict kernel: [ 3010.162737] [c00000000ffff950] [c00000000040a0b8] .ip_forward+0x2ec/0x454
Mar 28 00:37:11 addict kernel: [ 3010.167403] [c00000000ffff9e0] [c000000000408318] .ip_rcv_finish+0x374/0x394
Mar 28 00:37:11 addict kernel: [ 3010.171991] [c00000000ffffa70] [c0000000003d0d48] .__netif_receive_skb+0x698/0x6e8
Mar 28 00:37:11 addict kernel: [ 3010.180975] [c00000000ffffb60] [c0000000003d23a0] .netif_receive_skb+0x90/0x98
Mar 28 00:37:11 addict kernel: [ 3010.190225] [c00000000ffffc00] [c0000000003d4dc0] .napi_skb_finish+0x34/0x58
Mar 28 00:37:11 addict kernel: [ 3010.195132] [c00000000ffffc80] [d000000000aaebfc] .gem_poll+0x1c04/0x1ea4 [sungem]
Mar 28 00:37:11 addict kernel: [ 3010.204418] [c00000000ffffdc0] [c0000000003d2600] .net_rx_action+0xd0/0x30c
Mar 28 00:37:11 addict kernel: [ 3010.209294] [c00000000ffffeb0] [c00000000008b998] .__do_softirq+0x158/0x2a0
Mar 28 00:37:11 addict kernel: [ 3010.214192] [c00000000fffff90] [c00000000001c674] .call_do_softirq+0x14/0x24
Mar 28 00:37:11 addict kernel: [ 3010.219122] [c0000000007ef910] [c00000000000eda4] .do_softirq+0x7c/0xec
Mar 28 00:37:11 addict kernel: [ 3010.224051] [c0000000007ef9b0] [c00000000008bcec] .irq_exit+0x4c/0x9c
Mar 28 00:37:11 addict kernel: [ 3010.228983] [c0000000007efa30] [c00000000000ec04] .do_IRQ+0x1c4/0x240
Mar 28 00:37:11 addict kernel: [ 3010.233844] [c0000000007efae0] [c00000000000553c] hardware_interrupt_entry+0x18/0x1c
Mar 28 00:37:11 addict kernel: [ 3010.243320] --- Exception: 501 at .cpu_idle+0x10c/0x1cc
Mar 28 00:37:11 addict kernel: [ 3010.243323]     LR = .cpu_idle+0x10c/0x1cc
Mar 28 00:37:11 addict kernel: [ 3010.252811] [c0000000007efdd0] [c0000000000150f4] .cpu_idle+0x64/0x1cc (unreliable)
Mar 28 00:37:11 addict kernel: [ 3010.262134] [c0000000007efe60] [c00000000000ae54] .rest_init+0x84/0x90
Mar 28 00:37:11 addict kernel: [ 3010.266863] [c0000000007efee0] [c000000000676918] .start_kernel+0x454/0x45c
Mar 28 00:37:11 addict kernel: [ 3010.271492] [c0000000007eff90] [c0000000000096e8] .start_here_common+0x1c/0x34
Mar 28 00:37:11 addict kernel: [ 3010.280281] Mem-Info:
Mar 28 00:37:11 addict kernel: [ 3010.284602] Node 0 DMA per-cpu:
Mar 28 00:37:11 addict kernel: [ 3010.288923] CPU    0: hi:  186, btch:  31 usd:  94
Mar 28 00:37:11 addict kernel: [ 3010.293292] active_anon:3232 inactive_anon:2031 isolated_anon:0
Mar 28 00:37:11 addict kernel: [ 3010.293296]  active_file:29875 inactive_file:28307 isolated_file:0
Mar 28 00:37:11 addict kernel: [ 3010.293299]  unevictable:0 dirty:0 writeback:0 unstable:0
Mar 28 00:37:11 addict kernel: [ 3010.293301]  free:4180 slab_reclaimable:210472 slab_unreclaimable:38368
Mar 28 00:37:11 addict kernel: [ 3010.293304]  mapped:2564 shmem:83 pagetables:360 bounce:0
Mar 28 00:37:11 addict kernel: [ 3010.315010] Node 0 DMA free:16720kB min:5752kB low:7188kB high:8628kB active_anon:12928kB inactive_anon:8124kB active_file:119500kB inactive_file:113228kB unevictable:0kB isolated(anon):0kB isolated(file):0kB present:2068480kB mlocked:0kB dirty:0kB writeback:0kB mapped:10256kB shmem:332kB slab_reclaimable:841888kB slab_unreclaimable:153472kB kernel_stack:1792kB pagetables:1440kB unstable:0kB bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no
Mar 28 00:37:11 addict kernel: [ 3010.341624] lowmem_reserve[]: 0 0 0
Mar 28 00:37:11 addict kernel: [ 3010.346100] Node 0 DMA: 4180*4kB 0*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB 0*8192kB 0*16384kB = 16720kB
Mar 28 00:37:11 addict kernel: [ 3010.355529] 58769 total pagecache pages
Mar 28 00:37:11 addict kernel: [ 3010.360256] 500 pages in swap cache
Mar 28 00:37:11 addict kernel: [ 3010.364980] Swap cache stats: add 2284, delete 1784, find 338/418
Mar 28 00:37:11 addict kernel: [ 3010.369828] Free swap  = 2091148kB
Mar 28 00:37:11 addict kernel: [ 3010.374520] Total swap = 2097148kB
Mar 28 00:37:11 addict kernel: [ 3010.394630] 524288 pages RAM
Mar 28 00:37:11 addict kernel: [ 3010.399171] 22937 pages reserved
Mar 28 00:37:11 addict kernel: [ 3010.403595] 31296 pages shared
Mar 28 00:37:11 addict kernel: [ 3010.407858] 471269 pages non-shared


So installed 3.2.13-1 and since then everything seems to be fine. I'll keep an eye on it.

Mar 30 22:56:53 addict kernel: [    0.000000] Linux version 3.2.0-2-powerpc64 (Debian 3.2.13-1) (debian-kernel@lists.debian.org) (gcc version 4.6.2 (Debian 4.6.2-11) ) #1 SMP Thu Mar 29 01:04:30 UTC 2012

BTW it's PowerMac7,2 with 2 GB RAM

28. 3. 2012 v 1:26, Jonathan Nieder:

> reopen 666021
> tags 666021 + moreinfo
> quit
> 
> Ben Hutchings wrote:
>> On Tue, Mar 27, 2012 at 04:33:10PM -0500, Jonathan Nieder wrote:
> 
>>> This is probably one of the tcp-related bugs fixed in 3.2.12-1.
>> 
>> I don't think so.  I think it's likely to be a case of running too
>> many network services in too little memory.
> 
> Sure, that's what an allocation failure usually would suggest.
> Reopening, but I'm not sure what to suggest to track this down
> further.
> 
> Petr, did you only get the page allocation failure once, or did it
> happen on multiple occasions?




Reply to: