The patch titled
Subject: mm: fix inactive list balancing between NUMA nodes and cgroups
has been added to the -mm tree. Its filename is
mm-fix-inactive-list-balancing-between-numa-nodes-and-cgroups.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/mm-fix-inactive-list-balancing-between-numa-nodes-and-cgroups.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/mm-fix-inactive-list-balancing-between-numa-nodes-and-cgroups.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/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Johannes Weiner <hannes@xxxxxxxxxxx>
Subject: mm: fix inactive list balancing between NUMA nodes and cgroups
During !CONFIG_CGROUP reclaim, we expand the inactive list size if it's
thrashing on the node that is about to be reclaimed. But when cgroups are
enabled, we suddenly ignore the node scope and use the cgroup scope only.
The result is that pressure bleeds between NUMA nodes depending on whether
cgroups are merely compiled into Linux. This behavioral difference is
unexpected and undesirable.
When the refault adaptivity of the inactive list was first introduced,
there were no statistics at the lruvec level - the intersection of node
and memcg - so it was better than nothing.
But now that we have that infrastructure, use lruvec_page_state() to make
the list balancing decision always NUMA aware.
Link: http://lkml.kernel.org/r/20190412144438.2645-1-hannes@xxxxxxxxxxx
Fixes: 2a2e48854d70 ("mm: vmscan: fix IO/refault regression in cache workingset transition")
Signed-off-by: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Shakeel Butt <shakeelb@xxxxxxxxxx>
Cc: Roman Gushchin <guro@xxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/vmscan.c | 29 +++++++++--------------------
1 file changed, 9 insertions(+), 20 deletions(-)
--- a/mm/vmscan.c~mm-fix-inactive-list-balancing-between-numa-nodes-and-cgroups
+++ a/mm/vmscan.c
@@ -2176,7 +2176,6 @@ static void shrink_active_list(unsigned
* 10TB 320 32GB
*/
static bool inactive_list_is_low(struct lruvec *lruvec, bool file,
- struct mem_cgroup *memcg,
struct scan_control *sc, bool actual_reclaim)
{
enum lru_list active_lru = file * LRU_FILE + LRU_ACTIVE;
@@ -2197,16 +2196,12 @@ static bool inactive_list_is_low(struct
inactive = lruvec_lru_size(lruvec, inactive_lru, sc->reclaim_idx);
active = lruvec_lru_size(lruvec, active_lru, sc->reclaim_idx);
- if (memcg)
- refaults = memcg_page_state(memcg, WORKINGSET_ACTIVATE);
- else
- refaults = node_page_state(pgdat, WORKINGSET_ACTIVATE);
-
/*
* When refaults are being observed, it means a new workingset
* is being established. Disable active list protection to get
* rid of the stale workingset quickly.
*/
+ refaults = lruvec_page_state(lruvec, WORKINGSET_ACTIVATE);
if (file && actual_reclaim && lruvec->refaults != refaults) {
inactive_ratio = 0;
} else {
@@ -2227,12 +2222,10 @@ static bool inactive_list_is_low(struct
}
static unsigned long shrink_list(enum lru_list lru, unsigned long nr_to_scan,
- struct lruvec *lruvec, struct mem_cgroup *memcg,
- struct scan_control *sc)
+ struct lruvec *lruvec, struct scan_control *sc)
{
if (is_active_lru(lru)) {
- if (inactive_list_is_low(lruvec, is_file_lru(lru),
- memcg, sc, true))
+ if (inactive_list_is_low(lruvec, is_file_lru(lru), sc, true))
shrink_active_list(nr_to_scan, lruvec, sc, lru);
return 0;
}
@@ -2332,7 +2325,7 @@ static void get_scan_count(struct lruvec
* anonymous pages on the LRU in eligible zones.
* Otherwise, the small LRU gets thrashed.
*/
- if (!inactive_list_is_low(lruvec, false, memcg, sc, false) &&
+ if (!inactive_list_is_low(lruvec, false, sc, false) &&
lruvec_lru_size(lruvec, LRU_INACTIVE_ANON, sc->reclaim_idx)
>> sc->priority) {
scan_balance = SCAN_ANON;
@@ -2350,7 +2343,7 @@ static void get_scan_count(struct lruvec
* lruvec even if it has plenty of old anonymous pages unless the
* system is under heavy pressure.
*/
- if (!inactive_list_is_low(lruvec, true, memcg, sc, false) &&
+ if (!inactive_list_is_low(lruvec, true, sc, false) &&
lruvec_lru_size(lruvec, LRU_INACTIVE_FILE, sc->reclaim_idx) >> sc->priority) {
scan_balance = SCAN_FILE;
goto out;
@@ -2503,7 +2496,7 @@ static void shrink_node_memcg(struct pgl
nr[lru] -= nr_to_scan;
nr_reclaimed += shrink_list(lru, nr_to_scan,
- lruvec, memcg, sc);
+ lruvec, sc);
}
}
@@ -2570,7 +2563,7 @@ static void shrink_node_memcg(struct pgl
* Even if we did not try to evict anon pages at all, we want to
* rebalance the anon lru active/inactive ratio.
*/
- if (inactive_list_is_low(lruvec, false, memcg, sc, true))
+ if (inactive_list_is_low(lruvec, false, sc, true))
shrink_active_list(SWAP_CLUSTER_MAX, lruvec,
sc, LRU_ACTIVE_ANON);
}
@@ -2969,12 +2962,8 @@ static void snapshot_refaults(struct mem
unsigned long refaults;
struct lruvec *lruvec;
- if (memcg)
- refaults = memcg_page_state(memcg, WORKINGSET_ACTIVATE);
- else
- refaults = node_page_state(pgdat, WORKINGSET_ACTIVATE);
-
lruvec = mem_cgroup_lruvec(pgdat, memcg);
+ refaults = lruvec_page_state_local(lruvec, WORKINGSET_ACTIVATE);
lruvec->refaults = refaults;
} while ((memcg = mem_cgroup_iter(root_memcg, memcg, NULL)));
}
@@ -3339,7 +3328,7 @@ static void age_active_anon(struct pglis
do {
struct lruvec *lruvec = mem_cgroup_lruvec(pgdat, memcg);
- if (inactive_list_is_low(lruvec, false, memcg, sc, true))
+ if (inactive_list_is_low(lruvec, false, sc, true))
shrink_active_list(SWAP_CLUSTER_MAX, lruvec,
sc, LRU_ACTIVE_ANON);
_
Patches currently in -mm which might be from hannes@xxxxxxxxxxx are
mm-fix-inactive-list-balancing-between-numa-nodes-and-cgroups.patch
mm-memcontrol-track-lru-counts-in-the-vmstats-array.patch
mm-memcontrol-replace-zone-summing-with-lruvec_page_state.patch
mm-memcontrol-replace-node-summing-with-memcg_page_state.patch
mm-memcontrol-push-down-mem_cgroup_node_nr_lru_pages.patch
mm-memcontrol-push-down-mem_cgroup_nr_lru_pages.patch
mm-memcontrol-quarantine-the-mem_cgroup_nr_lru_pages-api.patch
mm-fix-false-positive-overcommit_guess-failures.patch
