Re: [BUG] __GFP_THISNODE is not always honored

[Mel Gorman]

On (18/08/08 14:21), Christoph Lameter didst pronounce:
> Adam Litke wrote:
> >
> > So far my debugging has led me to get_page_from_freelist() inside the
> > for_each_zone_zonelist() loop.  When buffered_rmqueue() returns a page I
> > compare the value of page_to_nid(page), zone->node and the node that the
> > hugetlb code requested with __GFP_THISNODE.  These all match -- except when the
> > problem triggers.  In that case, zone->node matches the node we asked for but
> > page_to_nid() does not.
> 
> Uhhh.. A page that was just taken off the freelist? So we may have freed or
> coalesced a page to the wrong zone? Looks like there is something more
> fundamental that broke here.
> 

It's still a bit hard to tell but I don't believe we are coalescing wrong
at the moment. buffered_rmqueue() is pretty high in the call chain for the
page allocator. The problem could have been explained if the zonelist walking
for __GFP_THISNODE was screwed but the dmesg output seems to show that's
ok at least. It could also be something really wacky like the page
linkages don't match the zone->node linkages.

-- 
Mel Gorman
Part-time Phd Student                          Linux Technology Center
University of Limerick                         IBM Dublin Software Lab
--
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/