2.6.39.2: page allocation failure question(__alloc_pages_nodemask+0x626/0x860)

[Justin Piszcz]

Hi,

[41624.229468] irq/52-eth0: page allocation failure. order:1, mode:0x20
[41624.229472] Pid: 1980, comm: irq/52-eth0 Not tainted 2.6.39.2 #1
[41624.229474] Call Trace:
[41624.229475]  <IRQ>  [<ffffffff81084916>] ? __alloc_pages_nodemask+0x626/0x860
[41624.229489]  [<ffffffff81601101>] ? cache_alloc_refill+0x25e/0x4c9
[41624.229493]  [<ffffffff810ad3aa>] ? kmem_cache_alloc+0x7a/0xa0
[41624.229496]  [<ffffffff814efd5c>] ? sk_prot_alloc.isra.41+0x3c/0x180
[41624.229499]  [<ffffffff814eff30>] ? sk_clone+0x10/0x2b0
[41624.229502]  [<ffffffff815436db>] ? inet_csk_clone+0xb/0xa0
[41624.229506]  [<ffffffff8155b0a1>] ? tcp_create_openreq_child+0x21/0x4e0
[41624.229509]  [<ffffffff81558233>] ? tcp_v4_syn_recv_sock+0x53/0x250
[41624.229511]  [<ffffffff8155aec4>] ? tcp_check_req+0x2d4/0x490
[41624.229513]  [<ffffffff8155810f>] ? tcp_v4_do_rcv+0x1bf/0x290

I do have enough swap available:

Mem:  16434572k total, 16165796k used,   268776k free,     9068k buffers
Swap: 31246388k total,    18296k used, 31228092k free, 11210636k cached

Perhaps I am just noticing this due to using the threadirq boot option
but is it normal to have page allocation failures on kernel processes if 
physical memory use is high/full and swap is available?

Justin.

--
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/