On Sun, 20 Mar 2011, Andrea Arcangeli wrote: > > Interesting that slab allocates with order > 0 an object that is <4096 > bytes. Is this related to slab_break_gfp_order? No, it's SLUB I'm using (partly for its excellent debugging, partly to trigger issues like this). Remember, that's SLUB's great weakness, that for optimal efficiency it relies upon higher order pages than you'd expect. It's much better since Christoph put in the ORDER_FALLBACK, but still makes a first attempt for a higher order page, which is liable to stir up page_alloc more than we'd like. Hugh -- 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/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>