Re: __alloc_pages: 2-order allocation failed

[Mike Edwards <sauron-debian-sparc@psychology.rutgers.edu>]

This is slightly different (1-order rather than 2-order), but it does
indeed look like SA is involved.  I'm also noting the one complaint
about imapd.


__alloc_pages: 1-order allocation failed (gfp=0x20/0)
__alloc_pages: task(spamc) pid(15858) caller(000000000046ed98)
__alloc_pages: 1-order allocation failed (gfp=0x20/0)
__alloc_pages: task(spamc) pid(15873) caller(000000000046ed98)
__alloc_pages: 1-order allocation failed (gfp=0x20/0)
__alloc_pages: task(spamc) pid(16043) caller(000000000046ed98)
__alloc_pages: 1-order allocation failed (gfp=0x20/0)
__alloc_pages: task(spamc) pid(16043) caller(000000000046ed98)
__alloc_pages: 1-order allocation failed (gfp=0x20/0)
__alloc_pages: task(imapd) pid(14534) caller(000000000046ed98)


On Thu, Mar 31, 2005 at 11:49:36AM -0800, David S. Miller said:
> On Thu, 31 Mar 2005 14:42:45 -0500
> Mike Edwards <sauron-debian-sparc@psychology.rutgers.edu> wrote:
> 
> > I can try the patch if you still feel it'll help - but it does look
> > like SA was at fault in this particular case.
> 
> Just knowing it is SA is insufficient information, we need to know
> what inside of the kernel, as part of whatever SA is doing, asks
> for such a large allocation.
> 
> Please apply the patch and obtain the debugging information, as I have
> asked you to.

-- 
Mike Edwards <sauron-debian-sparc@psychology.rutgers.edu>
System Administrator
Psychology Department, Rutgers University, Newark campus