On Fri, 31 Jul 2015, Joonsoo Kim wrote: > I don't think that this fix is right. > Just "kmalloc_size(INDEX_NODE) * 2" looks insane because it means 192 * 2 > = 384 on his platform. Why we need to check size is larger than 384? Its an arbitrary boundary. Making it large ensures that the smaller caches stay operational and do not fall back to page sized allocations. > I'm wondering what's the meaning of this check "size >= > kmalloc_size(INDEX_NODE + 1)". This is pretty old code. IMHO The check is if it fits in the kmem_cache used for INDEX_NODE. If not then fall back to a page sized allocation for the cache. Looks like DEBUG_PAGEALLOC wants one page per object. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx"> email@xxxxxxxxx </a>