The patch titled
memcg: reclaim memory from nodes in round-robin order
has been added to the -mm tree. Its filename is
memcg-reclaim-memory-from-nodes-in-round-robin-order.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
See http://userweb.kernel.org/~akpm/stuff/added-to-mm.txt to find
out what to do about this
The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/
------------------------------------------------------
Subject: memcg: reclaim memory from nodes in round-robin order
From: Ying Han <yinghan@xxxxxxxxxx>
Presently, memory cgroup's direct reclaim frees memory from the current
node. But this has some troubles. In usual, when a set of threads works
in cooperative way, they are tend to on the same node. So, if they hit
limits under memcg, it will reclaim memory from themselves, it may be
active working set.
For example, assume 2 node system which has Node 0 and Node 1 and a memcg
which has 1G limit. After some work, file cacne remains and and usages
are
Node 0: 1M
Node 1: 998M.
and run an application on Node 0, it will eats its foot before freeing
unnecessary file caches.
This patch adds round-robin for NUMA and adds equal pressure to each node.
When using cpuset's spread memory feature, this will work very well.
But yes, a better algorithm is needed.
Signed-off-by: Ying Han <yinghan@xxxxxxxxxx>
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Cc: Balbir Singh <balbir@xxxxxxxxxx>
Cc: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
Cc: Daisuke Nishimura <nishimura@xxxxxxxxxxxxxxxxx>
Cc: Mel Gorman <mel@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/memcontrol.h | 1
mm/memcontrol.c | 102 ++++++++++++++++++++++++++++++++---
mm/vmscan.c | 10 +++
3 files changed, 106 insertions(+), 7 deletions(-)
diff -puN include/linux/memcontrol.h~memcg-reclaim-memory-from-nodes-in-round-robin-order include/linux/memcontrol.h
--- a/include/linux/memcontrol.h~memcg-reclaim-memory-from-nodes-in-round-robin-order
+++ a/include/linux/memcontrol.h
@@ -108,6 +108,7 @@ extern void mem_cgroup_end_migration(str
*/
int mem_cgroup_inactive_anon_is_low(struct mem_cgroup *memcg);
int mem_cgroup_inactive_file_is_low(struct mem_cgroup *memcg);
+int mem_cgroup_select_victim_node(struct mem_cgroup *memcg);
unsigned long mem_cgroup_zone_nr_pages(struct mem_cgroup *memcg,
struct zone *zone,
enum lru_list lru);
diff -puN mm/memcontrol.c~memcg-reclaim-memory-from-nodes-in-round-robin-order mm/memcontrol.c
--- a/mm/memcontrol.c~memcg-reclaim-memory-from-nodes-in-round-robin-order
+++ a/mm/memcontrol.c
@@ -241,6 +241,11 @@ struct mem_cgroup {
* reclaimed from.
*/
int last_scanned_child;
+ int last_scanned_node;
+#if MAX_NUMNODES > 1
+ nodemask_t scan_nodes;
+ unsigned long next_scan_node_update;
+#endif
/*
* Should the accounting and control be hierarchical, per subtree?
*/
@@ -670,18 +675,27 @@ static void mem_cgroup_soft_scan(struct
val);
}
+static unsigned long
+mem_cgroup_get_zonestat_node(struct mem_cgroup *mem, int nid, enum lru_list idx)
+{
+ struct mem_cgroup_per_zone *mz;
+ u64 total;
+ int zid;
+
+ for (zid = 0; zid < MAX_NR_ZONES; zid++) {
+ mz = mem_cgroup_zoneinfo(mem, nid, zid);
+ total += MEM_CGROUP_ZSTAT(mz, idx);
+ }
+ return total;
+}
static unsigned long mem_cgroup_get_local_zonestat(struct mem_cgroup *mem,
enum lru_list idx)
{
- int nid, zid;
- struct mem_cgroup_per_zone *mz;
+ int nid;
u64 total = 0;
for_each_online_node(nid)
- for (zid = 0; zid < MAX_NR_ZONES; zid++) {
- mz = mem_cgroup_zoneinfo(mem, nid, zid);
- total += MEM_CGROUP_ZSTAT(mz, idx);
- }
+ total += mem_cgroup_get_zonestat_node(mem, nid, idx);
return total;
}
@@ -1491,6 +1505,81 @@ mem_cgroup_select_victim(struct mem_cgro
return ret;
}
+#if MAX_NUMNODES > 1
+
+/*
+ * Update nodemask always is not very good. Even if we have empty
+ * list, or wrong list here, we can start from some node and traverse all nodes
+ * based on zonelist. So, update the list loosely once in 10 secs.
+ *
+ */
+static void mem_cgroup_may_update_nodemask(struct mem_cgroup *mem)
+{
+ int nid;
+
+ if (time_before(mem->next_scan_node_update, jiffies))
+ return;
+
+ mem->next_scan_node_update = jiffies + 10*HZ;
+ /* make a nodemask where this memcg uses memory from */
+ mem->scan_nodes = node_states[N_HIGH_MEMORY];
+
+ for_each_node_mask(nid, node_states[N_HIGH_MEMORY]) {
+
+ if (mem_cgroup_get_zonestat_node(mem, nid, LRU_INACTIVE_FILE) ||
+ mem_cgroup_get_zonestat_node(mem, nid, LRU_ACTIVE_FILE))
+ continue;
+
+ if (total_swap_pages &&
+ (mem_cgroup_get_zonestat_node(mem, nid, LRU_INACTIVE_ANON) ||
+ mem_cgroup_get_zonestat_node(mem, nid, LRU_ACTIVE_ANON)))
+ continue;
+ node_clear(nid, mem->scan_nodes);
+ }
+}
+
+/*
+ * Selecting a node where we start reclaim from. Because what we need is just
+ * reducing usage counter, start from anywhere is O,K. Considering
+ * memory reclaim from current node, there are pros. and cons.
+ *
+ * Freeing memory from current node means freeing memory from a node which
+ * we'll use or we've used. So, it may make LRU bad. And if several threads
+ * hit limits, it will see a contention on a node. But freeing from remote
+ * node means more costs for memory reclaim because of memory latency.
+ *
+ * Now, we use round-robin. Better algorithm is welcomed.
+ */
+int mem_cgroup_select_victim_node(struct mem_cgroup *mem)
+{
+ int node;
+
+ mem_cgroup_may_update_nodemask(mem);
+ node = mem->last_scanned_node;
+
+ node = next_node(node, mem->scan_nodes);
+ if (node == MAX_NUMNODES)
+ node = first_node(mem->scan_nodes);
+ /*
+ * We call this when we hit limit, not when pages are added to LRU.
+ * No LRU may hold pages because all pages are UNEVICTABLE or
+ * memcg is too small and all pages are not on LRU. In that case,
+ * we use curret node.
+ */
+ if (unlikely(node == MAX_NUMNODES))
+ node = numa_node_id();
+
+ mem->last_scanned_node = node;
+ return node;
+}
+
+#else
+int mem_cgroup_select_victim_node(struct mem_cgroup *mem)
+{
+ return 0;
+}
+#endif
+
/*
* Scan the hierarchy if needed to reclaim memory. We remember the last child
* we reclaimed from, so that we don't end up penalizing one child extensively
@@ -4708,6 +4797,7 @@ mem_cgroup_create(struct cgroup_subsys *
res_counter_init(&mem->memsw, NULL);
}
mem->last_scanned_child = 0;
+ mem->last_scanned_node = MAX_NUMNODES;
INIT_LIST_HEAD(&mem->oom_notify);
if (parent)
diff -puN mm/vmscan.c~memcg-reclaim-memory-from-nodes-in-round-robin-order mm/vmscan.c
--- a/mm/vmscan.c~memcg-reclaim-memory-from-nodes-in-round-robin-order
+++ a/mm/vmscan.c
@@ -2216,6 +2216,7 @@ unsigned long try_to_free_mem_cgroup_pag
{
struct zonelist *zonelist;
unsigned long nr_reclaimed;
+ int nid;
struct scan_control sc = {
.may_writepage = !laptop_mode,
.may_unmap = 1,
@@ -2232,7 +2233,14 @@ unsigned long try_to_free_mem_cgroup_pag
.gfp_mask = sc.gfp_mask,
};
- zonelist = NODE_DATA(numa_node_id())->node_zonelists;
+ /*
+ * Unlike direct reclaim via alloc_pages(), memcg's reclaim doesn't
+ * take care of from where we get pages. So the node where we start the
+ * scan does not need to be the current node.
+ */
+ nid = mem_cgroup_select_victim_node(mem_cont);
+
+ zonelist = NODE_DATA(nid)->node_zonelists;
trace_mm_vmscan_memcg_reclaim_begin(0,
sc.may_writepage,
_
Patches currently in -mm which might be from yinghan@xxxxxxxxxx are
mm-check-pageunevictable-in-lru_deactivate_fn.patch
vmscan-change-shrink_slab-interfaces-by-passing-shrink_control.patch
vmscan-change-shrink_slab-interfaces-by-passing-shrink_control-fix.patch
vmscan-change-shrink_slab-interfaces-by-passing-shrink_control-fix-2.patch
vmscan-change-shrinker-api-by-passing-shrink_control-struct.patch
vmscan-change-shrinker-api-by-passing-shrink_control-struct-fix.patch
vmscan-change-shrinker-api-by-passing-shrink_control-struct-fix-2.patch
mm-move-enum-vm_event_item-into-a-standalone-header-file.patch
memcg-count-the-soft_limit-reclaim-in-global-background-reclaim.patch
memcg-add-the-soft_limit-reclaim-in-global-direct-reclaim.patch
memcg-add-stats-to-monitor-soft_limit-reclaim.patch
memcg-add-stats-to-monitor-soft_limit-reclaim-v2.patch
add-the-pagefault-count-into-memcg-stats.patch
add-the-pagefault-count-into-memcg-stats-fix.patch
memcg-reclaim-memory-from-nodes-in-round-robin-order.patch
memcg-reclaim-memory-from-nodes-in-round-robin-fix.patch
memcg-fix-get_scan_count-for-small-targets.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
