On Tue, Apr 13, 2010 at 1:43 AM, Mel Gorman <mel@xxxxxxxxx> wrote: > On Sat, Apr 10, 2010 at 07:49:32PM +0800, Bob Liu wrote: >> Since alloc_pages_exact_node() is not for allocate page from >> exact node but just for removing check of node's valid, >> rename it to alloc_pages_from_valid_node(). Else will make >> people misunderstanding. >> > > I don't know about this change either but as I introduced the original > function name, I am biased. My reading of it is - allocate me pages and > I know exactly which node I need. I see how it it could be read as > "allocate me pages from exactly this node" but I don't feel the new > naming is that much clearer either. Tend to agree. Then, don't change function name but add some comment? /* * allow pages from fallback if page allocator can't find free page in your nid. * If you want to allocate page from exact node, please use __GFP_THISNODE flags with * gfp_mask. */ static inline struct page *alloc_pages_exact_node(.... -- Kind regards, Minchan Kim -- 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>