Messages in this thread | | | Date | Mon, 23 Feb 2009 16:24:32 +0000 | From | Mel Gorman <> | Subject | Re: [PATCH 03/20] Do not check NUMA node ID when the caller knows the node is valid |
| |
On Mon, Feb 23, 2009 at 10:01:35AM -0500, Christoph Lameter wrote: > On Sun, 22 Feb 2009, Mel Gorman wrote: > > > diff --git a/mm/vmalloc.c b/mm/vmalloc.c > > index 75f49d3..6566c9e 100644 > > --- a/mm/vmalloc.c > > +++ b/mm/vmalloc.c > > @@ -1318,11 +1318,7 @@ static void *__vmalloc_area_node(struct vm_struct *area, gfp_t gfp_mask, > > for (i = 0; i < area->nr_pages; i++) { > > struct page *page; > > > > - if (node < 0) > > - page = alloc_page(gfp_mask); > > - else > > - page = alloc_pages_node(node, gfp_mask, 0); > > - > > + page = alloc_pages_node(node, gfp_mask, 0); > > if (unlikely(!page)) { > > /* Successfully allocated i pages, free them in __vunmap() */ > > area->nr_pages = i; > > > > That wont work. alloc_pages() obeys memory policies. alloc_pages_node() > does not. >
Correct, I failed to take policies into account. The same comment applied for the slub modification. I've reverted this part. Good spot.
-- Mel Gorman Part-time Phd Student Linux Technology Center University of Limerick IBM Dublin Software Lab
| |