On Wed, Aug 25, 2010 at 09:29:41AM +0800, Wu Fengguang wrote:
> > +static struct page *alloc_buddy_huge_page_node(struct hstate *h, int nid)
> > +{
> > + struct page *page = __alloc_huge_page_node(h, nid);
> > if (page) {
> > - if (arch_prepare_hugepage(page)) {
> > - __free_pages(page, huge_page_order(h));
> > + set_compound_page_dtor(page, free_huge_page);
> > + spin_lock(&hugetlb_lock);
> > + h->nr_huge_pages++;
> > + h->nr_huge_pages_node[nid]++;
> > + spin_unlock(&hugetlb_lock);
> > + put_page_testzero(page);
> > + }
> > + return page;
> > +}
>
> One would expect the alloc_buddy_huge_page_node() to only differ with
> alloc_buddy_huge_page() in the alloc_pages/alloc_pages_exact_node
> calls. However you implement alloc_buddy_huge_page_node() in a quite
> different way. Can the two functions be unified at all?
Yes. I did it by adding argument @nid to alloc_buddy_huge_page().
Code gets cleaner and work without problems.
Thanks,
Naoya
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxxx For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>