Re: [2.6.25.8] "kswapd0: page allocation failure. order:3" intcp_collapse

[Andrew Morton]

On Wed, 02 Jul 2008 00:30:30 +0200 Tilman Schmidt <tilman@xxxxxxx> wrote:

> My Intel Pentium D940 system, Intel DQ965GF mainboard, 2 GB RAM, running an
> x86_32 build of kernel release 2.6.25.8 with openSUSE 10.3 userland, produced
> the message in the subject accompanied by a stack backtrace, seven times in a
> row within a few milliseconds, during the nightly backup of a Windows system
> through Bacula over the LAN. The backup did not report an error, though, and
> everything else seems to be running normally, too.
> 
> The kernel is built with CONFIG_SLUB=y.
> 
> ...
> 
> Should I worry? File a bug? Post more details? 

e1000 driver has a lovely habit of allocating 32k of GFP_ATOMIC memory for
packets which are a quarter that size.  AFAIK it is unfixable.  It is a
well-known and oft-reported problem.  The system should recover OK.

> Just ignore it?

Learn to enjoy it ;)
--
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/