Move remote node draining out of slab allocators

Currently the slab allocators contain callbacks into the page allocator to
perform the draining of pagesets on remote nodes.  This requires SLUB to have
a whole subsystem in order to be compatible with SLAB.  Moving node draining
out of the slab allocators avoids a section of code in SLUB.

Move the node draining so that is is done when the vm statistics are updated.
At that point we are already touching all the cachelines with the pagesets of
a processor.

Add a expire counter there.  If we have to update per zone or global vm
statistics then assume that the pageset will require subsequent draining.

The expire counter will be decremented on each vm stats update pass until it
reaches zero.  Then we will drain one batch from the pageset.  The draining
will cause vm counter updates which will then cause another expiration until
the pcp is empty.  So we will drain a batch every 3 seconds.

Note that remote node draining is a somewhat esoteric feature that is required
on large NUMA systems because otherwise significant portions of system memory
can become trapped in pcp queues.  The number of pcp is determined by the
number of processors and nodes in a system.  A system with 4 processors and 2
nodes has 8 pcps which is okay.  But a system with 1024 processors and 512
nodes has 512k pcps with a high potential for large amount of memory being
caught in them.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 97a36c3..0d2ef0b 100644 (file)
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -176,10 +176,6 @@ extern void FASTCALL(free_cold_page(struct page *page));
 #define free_page(addr) free_pages((addr),0)
 
 void page_alloc_init(void);
-#ifdef CONFIG_NUMA
-void drain_node_pages(int node);
-#else
-static inline void drain_node_pages(int node) { };
-#endif
+void drain_zone_pages(struct zone *zone, struct per_cpu_pages *pcp);
 
 #endif /* __LINUX_GFP_H */

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 2f1544e..d09b134 100644 (file)
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -83,6 +83,9 @@ struct per_cpu_pages {
 
 struct per_cpu_pageset {
        struct per_cpu_pages pcp[2];    /* 0: hot.  1: cold */
+#ifdef CONFIG_NUMA
+       s8 expire;
+#endif
 #ifdef CONFIG_SMP
        s8 stat_threshold;
        s8 vm_stat_diff[NR_VM_ZONE_STAT_ITEMS];

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index d53cbf8..f9b5d6d 100644 (file)
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -691,43 +691,26 @@ static void __init setup_nr_node_ids(void) {}
 
 #ifdef CONFIG_NUMA
 /*
- * Called from the slab reaper to drain pagesets on a particular node that
- * belongs to the currently executing processor.
+ * Called from the vmstat counter updater to drain pagesets of this
+ * currently executing processor on remote nodes after they have
+ * expired.
+ *
  * Note that this function must be called with the thread pinned to
  * a single processor.
  */
-void drain_node_pages(int nodeid)
+void drain_zone_pages(struct zone *zone, struct per_cpu_pages *pcp)
 {
-       int i;
-       enum zone_type z;
        unsigned long flags;
+       int to_drain;
 
-       for (z = 0; z < MAX_NR_ZONES; z++) {
-               struct zone *zone = NODE_DATA(nodeid)->node_zones + z;
-               struct per_cpu_pageset *pset;
-
-               if (!populated_zone(zone))
-                       continue;
-
-               pset = zone_pcp(zone, smp_processor_id());
-               for (i = 0; i < ARRAY_SIZE(pset->pcp); i++) {
-                       struct per_cpu_pages *pcp;
-
-                       pcp = &pset->pcp[i];
-                       if (pcp->count) {
-                               int to_drain;
-
-                               local_irq_save(flags);
-                               if (pcp->count >= pcp->batch)
-                                       to_drain = pcp->batch;
-                               else
-                                       to_drain = pcp->count;
-                               free_pages_bulk(zone, to_drain, &pcp->list, 0);
-                               pcp->count -= to_drain;
-                               local_irq_restore(flags);
-                       }
-               }
-       }
+       local_irq_save(flags);
+       if (pcp->count >= pcp->batch)
+               to_drain = pcp->batch;
+       else
+               to_drain = pcp->count;
+       free_pages_bulk(zone, to_drain, &pcp->list, 0);
+       pcp->count -= to_drain;
+       local_irq_restore(flags);
 }
 #endif
 

diff --git a/mm/slab.c b/mm/slab.c
index e50908b..944b205 100644 (file)
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -928,12 +928,6 @@ static void next_reap_node(void)
 {
        int node = __get_cpu_var(reap_node);
 
-       /*
-        * Also drain per cpu pages on remote zones
-        */
-       if (node != numa_node_id())
-               drain_node_pages(node);
-
        node = next_node(node, node_online_map);
        if (unlikely(node >= MAX_NUMNODES))
                node = first_node(node_online_map);

diff --cc mm/slub.c
Simple merge

diff --cc mm/vmstat.c
Simple merge