Patchwork slub numa: Fix rare allocation from unexpected node

login
register
mail settings
Submitter Christoph Lameter
Date July 26, 2010, 3:41 p.m.
Message ID <alpine.DEB.2.00.1007261040430.5438@router.home>
Download mbox | patch
Permalink /patch/59925/
State Not Applicable
Delegated to: David Miller
Headers show

Comments

Christoph Lameter - July 26, 2010, 3:41 p.m.
Subject: slub numa: Fix rare allocation from unexpected node

The network developers have seen sporadic allocations resulting in objects
coming from unexpected NUMA nodes despite asking for objects from a
specific node.

This is due to get_partial() calling get_any_partial() if partial
slabs are exhausted for a node even if a node was specified and therefore
one would expect allocations only from the specified node.

get_any_partial() sporadically may return a slab from a foreign
node to gradually reduce the size of partial lists on remote nodes
and thereby reduce total memory use for a slab cache.

The behavior is controlled by the remote_defrag_ratio of each cache.

Strictly speaking this is permitted behavior since __GFP_THISNODE was
not specified for the allocation but it is certain surprising.

This patch makes sure that the remote defrag behavior only occurs
if no node was specified.

Signed-off-by: Christoph Lameter <cl@linux-foundation.org>

---
 mm/slub.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Pekka Enberg - July 29, 2010, 10 a.m.
Christoph Lameter wrote:
> Subject: slub numa: Fix rare allocation from unexpected node
> 
> The network developers have seen sporadic allocations resulting in objects
> coming from unexpected NUMA nodes despite asking for objects from a
> specific node.
> 
> This is due to get_partial() calling get_any_partial() if partial
> slabs are exhausted for a node even if a node was specified and therefore
> one would expect allocations only from the specified node.
> 
> get_any_partial() sporadically may return a slab from a foreign
> node to gradually reduce the size of partial lists on remote nodes
> and thereby reduce total memory use for a slab cache.
> 
> The behavior is controlled by the remote_defrag_ratio of each cache.
> 
> Strictly speaking this is permitted behavior since __GFP_THISNODE was
> not specified for the allocation but it is certain surprising.
> 
> This patch makes sure that the remote defrag behavior only occurs
> if no node was specified.
> 
> Signed-off-by: Christoph Lameter <cl@linux-foundation.org>
> 
> ---
>  mm/slub.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> Index: linux-2.6/mm/slub.c
> ===================================================================
> --- linux-2.6.orig/mm/slub.c	2010-07-23 09:24:11.000000000 -0500
> +++ linux-2.6/mm/slub.c	2010-07-23 09:25:15.000000000 -0500
> @@ -1390,7 +1390,7 @@ static struct page *get_partial(struct k
>  	int searchnode = (node == -1) ? numa_node_id() : node;
> 
>  	page = get_partial_node(get_node(s, searchnode));
> -	if (page || (flags & __GFP_THISNODE))
> +	if (page || node != -1)
>  		return page;
> 
>  	return get_any_partial(s, flags);

Applied, thanks!
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

Index: linux-2.6/mm/slub.c
===================================================================
--- linux-2.6.orig/mm/slub.c	2010-07-23 09:24:11.000000000 -0500
+++ linux-2.6/mm/slub.c	2010-07-23 09:25:15.000000000 -0500
@@ -1390,7 +1390,7 @@  static struct page *get_partial(struct k
 	int searchnode = (node == -1) ? numa_node_id() : node;

 	page = get_partial_node(get_node(s, searchnode));
-	if (page || (flags & __GFP_THISNODE))
+	if (page || node != -1)
 		return page;

 	return get_any_partial(s, flags);