Messages in this thread Patch in this message | | | Date | Fri, 06 May 2011 13:05:42 -0500 | From | Christoph Lameter <> | Subject | [slubllv4 01/16] slub: Per object NUMA support |
| |
Currently slub applies NUMA policies per allocated slab page. Change that to apply memory policies for each individual object allocated.
F.e. before this patch MPOL_INTERLEAVE would return objects from the same slab page until a new slab page was allocated. Now an object from a different page is taken for each allocation.
This increases the overhead of the fastpath under NUMA.
Signed-off-by: Christoph Lameter <cl@linux.com>
--- mm/slub.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+)
Index: linux-2.6/mm/slub.c =================================================================== --- linux-2.6.orig/mm/slub.c 2011-05-05 15:21:51.000000000 -0500 +++ linux-2.6/mm/slub.c 2011-05-05 15:28:33.000000000 -0500 @@ -1873,6 +1873,21 @@ debug: goto unlock_out; } +static __always_inline int alternate_slab_node(struct kmem_cache *s, + gfp_t flags, int node) +{ +#ifdef CONFIG_NUMA + if (unlikely(node == NUMA_NO_NODE && + !(flags & __GFP_THISNODE) && + !in_interrupt())) { + if ((s->flags & SLAB_MEM_SPREAD) && cpuset_do_slab_mem_spread()) + node = cpuset_slab_spread_node(); + else if (current->mempolicy) + node = slab_node(current->mempolicy); + } +#endif + return node; +} /* * Inlined fastpath so that allocation functions (kmalloc, kmem_cache_alloc) * have the fastpath folded into their functions. So no function call @@ -1893,6 +1908,8 @@ static __always_inline void *slab_alloc( if (slab_pre_alloc_hook(s, gfpflags)) return NULL; + node = alternate_slab_node(s, gfpflags, node); + redo: /*
| |