The commit 2d146aa3aa84 ("mm: memcontrol: switch to rstat") switched
memcg stats to rstat infrastructure but skipped the conversion of the
lruvec stats as such stats are read in the performance critical code
paths and flushing stats may have impacted the performances of the
applications. This patch converts the lruvec stats to rstat and later
patches add mechanisms to keep the performance impact to minimum.
The rstat conversion comes with the price i.e. memory cost. Effectively
this patch reverts the savings done by the commit f3344adf38bd ("mm:
memcontrol: optimize per-lruvec stats counter memory usage"). However
this cost is justified due to negative impact of the inaccurate lruvec
stats on many heuristics. One such case is reported in [1].
The memory reclaim code is filled with plethora of heuristics and many
of those heuristics reads the lruvec stats. So, inaccurate stats can
make such heuristics ineffective. [1] reports the impact of inaccurate
lruvec stats on the "cache trim mode" heuristic. Inaccurate lruvec stats
can impact the deactivation and aging anon heuristics as well.
[1] https://lore.kernel.org/linux-mm/[email protected]/
---
Changes since v2:
- updated commit message
Changes since v1:
- None
include/linux/memcontrol.h | 42 +++++++-------
mm/memcontrol.c | 114 +++++++++++++------------------------
2 files changed, 58 insertions(+), 98 deletions(-)
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index bfe5c486f4ad..0bfa0409af22 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -105,14 +105,6 @@ struct mem_cgroup_reclaim_iter {
unsigned int generation;
};
-struct lruvec_stat {
- long count[NR_VM_NODE_STAT_ITEMS];
-};
-
-struct batched_lruvec_stat {
- s32 count[NR_VM_NODE_STAT_ITEMS];
-};
-
/*
* Bitmap and deferred work of shrinker::id corresponding to memcg-aware
* shrinkers, which have elements charged to this memcg.
@@ -123,24 +115,30 @@ struct shrinker_info {
unsigned long *map;
};
+struct lruvec_stats_percpu {
+ /* Local (CPU and cgroup) state */
+ long state[NR_VM_NODE_STAT_ITEMS];
+
+ /* Delta calculation for lockless upward propagation */
+ long state_prev[NR_VM_NODE_STAT_ITEMS];
+};
+
+struct lruvec_stats {
+ /* Aggregated (CPU and subtree) state */
+ long state[NR_VM_NODE_STAT_ITEMS];
+
+ /* Pending child counts during tree propagation */
+ long state_pending[NR_VM_NODE_STAT_ITEMS];
+};
+
/*
* per-node information in memory controller.
*/
struct mem_cgroup_per_node {
struct lruvec lruvec;
- /*
- * Legacy local VM stats. This should be struct lruvec_stat and
- * cannot be optimized to struct batched_lruvec_stat. Because
- * the threshold of the lruvec_stat_cpu can be as big as
- * MEMCG_CHARGE_BATCH * PAGE_SIZE. It can fit into s32. But this
- * filed has no upper limit.
- */
- struct lruvec_stat __percpu *lruvec_stat_local;
-
- /* Subtree VM stats (batched updates) */
- struct batched_lruvec_stat __percpu *lruvec_stat_cpu;
- atomic_long_t lruvec_stat[NR_VM_NODE_STAT_ITEMS];
+ struct lruvec_stats_percpu __percpu *lruvec_stats_percpu;
+ struct lruvec_stats lruvec_stats;
unsigned long lru_zone_size[MAX_NR_ZONES][NR_LRU_LISTS];
@@ -965,7 +963,7 @@ static inline unsigned long lruvec_page_state(struct lruvec *lruvec,
return node_page_state(lruvec_pgdat(lruvec), idx);
pn = container_of(lruvec, struct mem_cgroup_per_node, lruvec);
- x = atomic_long_read(&pn->lruvec_stat[idx]);
+ x = READ_ONCE(pn->lruvec_stats.state[idx]);
#ifdef CONFIG_SMP
if (x < 0)
x = 0;
@@ -985,7 +983,7 @@ static inline unsigned long lruvec_page_state_local(struct lruvec *lruvec,
pn = container_of(lruvec, struct mem_cgroup_per_node, lruvec);
for_each_possible_cpu(cpu)
- x += per_cpu(pn->lruvec_stat_local->count[idx], cpu);
+ x += per_cpu(pn->lruvec_stats_percpu->state[idx], cpu);
#ifdef CONFIG_SMP
if (x < 0)
x = 0;
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index ae1f5d0cb581..848d711bf576 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -671,23 +671,11 @@ static unsigned long memcg_page_state_local(struct mem_cgroup *memcg, int idx)
return x;
}
-static struct mem_cgroup_per_node *
-parent_nodeinfo(struct mem_cgroup_per_node *pn, int nid)
-{
- struct mem_cgroup *parent;
-
- parent = parent_mem_cgroup(pn->memcg);
- if (!parent)
- return NULL;
- return parent->nodeinfo[nid];
-}
-
void __mod_memcg_lruvec_state(struct lruvec *lruvec, enum node_stat_item idx,
int val)
{
struct mem_cgroup_per_node *pn;
struct mem_cgroup *memcg;
- long x, threshold = MEMCG_CHARGE_BATCH;
pn = container_of(lruvec, struct mem_cgroup_per_node, lruvec);
memcg = pn->memcg;
@@ -696,21 +684,7 @@ void __mod_memcg_lruvec_state(struct lruvec *lruvec, enum node_stat_item idx,
__mod_memcg_state(memcg, idx, val);
/* Update lruvec */
- __this_cpu_add(pn->lruvec_stat_local->count[idx], val);
-
- if (vmstat_item_in_bytes(idx))
- threshold <<= PAGE_SHIFT;
-
- x = val + __this_cpu_read(pn->lruvec_stat_cpu->count[idx]);
- if (unlikely(abs(x) > threshold)) {
- pg_data_t *pgdat = lruvec_pgdat(lruvec);
- struct mem_cgroup_per_node *pi;
-
- for (pi = pn; pi; pi = parent_nodeinfo(pi, pgdat->node_id))
- atomic_long_add(x, &pi->lruvec_stat[idx]);
- x = 0;
- }
- __this_cpu_write(pn->lruvec_stat_cpu->count[idx], x);
+ __this_cpu_add(pn->lruvec_stats_percpu->state[idx], val);
}
/**
@@ -2289,40 +2263,13 @@ static void drain_all_stock(struct mem_cgroup *root_memcg)
mutex_unlock(&percpu_charge_mutex);
}
-static void memcg_flush_lruvec_page_state(struct mem_cgroup *memcg, int cpu)
-{
- int nid;
-
- for_each_node(nid) {
- struct mem_cgroup_per_node *pn = memcg->nodeinfo[nid];
- unsigned long stat[NR_VM_NODE_STAT_ITEMS];
- struct batched_lruvec_stat *lstatc;
- int i;
-
- lstatc = per_cpu_ptr(pn->lruvec_stat_cpu, cpu);
- for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) {
- stat[i] = lstatc->count[i];
- lstatc->count[i] = 0;
- }
-
- do {
- for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++)
- atomic_long_add(stat[i], &pn->lruvec_stat[i]);
- } while ((pn = parent_nodeinfo(pn, nid)));
- }
-}
-
static int memcg_hotplug_cpu_dead(unsigned int cpu)
{
struct memcg_stock_pcp *stock;
- struct mem_cgroup *memcg;
stock = &per_cpu(memcg_stock, cpu);
drain_stock(stock);
- for_each_mem_cgroup(memcg)
- memcg_flush_lruvec_page_state(memcg, cpu);
-
return 0;
}
@@ -5126,17 +5073,9 @@ static int alloc_mem_cgroup_per_node_info(struct mem_cgroup *memcg, int node)
if (!pn)
return 1;
- pn->lruvec_stat_local = alloc_percpu_gfp(struct lruvec_stat,
- GFP_KERNEL_ACCOUNT);
- if (!pn->lruvec_stat_local) {
- kfree(pn);
- return 1;
- }
-
- pn->lruvec_stat_cpu = alloc_percpu_gfp(struct batched_lruvec_stat,
- GFP_KERNEL_ACCOUNT);
- if (!pn->lruvec_stat_cpu) {
- free_percpu(pn->lruvec_stat_local);
+ pn->lruvec_stats_percpu = alloc_percpu_gfp(struct lruvec_stats_percpu,
+ GFP_KERNEL_ACCOUNT);
+ if (!pn->lruvec_stats_percpu) {
kfree(pn);
return 1;
}
@@ -5157,8 +5096,7 @@ static void free_mem_cgroup_per_node_info(struct mem_cgroup *memcg, int node)
if (!pn)
return;
- free_percpu(pn->lruvec_stat_cpu);
- free_percpu(pn->lruvec_stat_local);
+ free_percpu(pn->lruvec_stats_percpu);
kfree(pn);
}
@@ -5174,15 +5112,7 @@ static void __mem_cgroup_free(struct mem_cgroup *memcg)
static void mem_cgroup_free(struct mem_cgroup *memcg)
{
- int cpu;
-
memcg_wb_domain_exit(memcg);
- /*
- * Flush percpu lruvec stats to guarantee the value
- * correctness on parent's and all ancestor levels.
- */
- for_each_online_cpu(cpu)
- memcg_flush_lruvec_page_state(memcg, cpu);
__mem_cgroup_free(memcg);
}
@@ -5415,7 +5345,7 @@ static void mem_cgroup_css_rstat_flush(struct cgroup_subsys_state *css, int cpu)
struct mem_cgroup *parent = parent_mem_cgroup(memcg);
struct memcg_vmstats_percpu *statc;
long delta, v;
- int i;
+ int i, nid;
statc = per_cpu_ptr(memcg->vmstats_percpu, cpu);
@@ -5463,6 +5393,36 @@ static void mem_cgroup_css_rstat_flush(struct cgroup_subsys_state *css, int cpu)
if (parent)
parent->vmstats.events_pending[i] += delta;
}
+
+ for_each_node_state(nid, N_MEMORY) {
+ struct mem_cgroup_per_node *pn = memcg->nodeinfo[nid];
+ struct mem_cgroup_per_node *ppn = NULL;
+ struct lruvec_stats_percpu *lstatc;
+
+ if (parent)
+ ppn = parent->nodeinfo[nid];
+
+ lstatc = per_cpu_ptr(pn->lruvec_stats_percpu, cpu);
+
+ for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) {
+ delta = pn->lruvec_stats.state_pending[i];
+ if (delta)
+ pn->lruvec_stats.state_pending[i] = 0;
+
+ v = READ_ONCE(lstatc->state[i]);
+ if (v != lstatc->state_prev[i]) {
+ delta += v - lstatc->state_prev[i];
+ lstatc->state_prev[i] = v;
+ }
+
+ if (!delta)
+ continue;
+
+ pn->lruvec_stats.state[i] += delta;
+ if (ppn)
+ ppn->lruvec_stats.state_pending[i] += delta;
+ }
+ }
}
#ifdef CONFIG_MMU
@@ -6396,6 +6356,8 @@ static int memory_numa_stat_show(struct seq_file *m, void *v)
int i;
struct mem_cgroup *memcg = mem_cgroup_from_seq(m);
+ cgroup_rstat_flush(memcg->css.cgroup);
+
for (i = 0; i < ARRAY_SIZE(memory_stats); i++) {
int nid;
--
2.32.0.93.g670b81a890-goog
At the moment memcg stats are read in four contexts:
1. memcg stat user interfaces
2. dirty throttling
3. page fault
4. memory reclaim
Currently the kernel flushes the stats for first two cases. Flushing the
stats for remaining two casese may have performance impact. Always
flushing the memcg stats on the page fault code path may negatively
impacts the performance of the applications. In addition flushing in the
memory reclaim code path, though treated as slowpath, can become the
source of contention for the global lock taken for stat flushing because
when system or memcg is under memory pressure, many tasks may enter the
reclaim path.
This patch uses following mechanisms to solve these challenges:
1. Periodically flush the stats from root memcg every 2 seconds. This
will time limit the out of sync stats.
2. Asynchronously flush the stats after fixed number of stat updates.
In the worst case the stat can be out of sync by O(nr_cpus * BATCH) for
2 seconds.
3. For avoiding thundering herd to flush the stats particularly from the
memory reclaim context, introduce memcg local spinlock and let only one
flusher active at a time. This could have been done through
cgroup_rstat_lock lock but that lock is used by other subsystem and for
userspace reading memcg stats. So, it is better to keep flushers
introduced by this patch decoupled from cgroup_rstat_lock.
---
Changes since v2:
- Changed the subject of the patch
- Added mechanism to bound errors to nr_cpus instead of nr_cgroups
- memcg local lock to let one active flusher
Changes since v1:
- use system_unbound_wq for flushing the memcg stats
include/linux/memcontrol.h | 6 ++++++
mm/memcontrol.c | 34 ++++++++++++++++++++++++++++++++++
mm/vmscan.c | 6 ++++++
3 files changed, 46 insertions(+)
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 0bfa0409af22..fa095a94ae56 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -991,6 +991,8 @@ static inline unsigned long lruvec_page_state_local(struct lruvec *lruvec,
return x;
}
+void mem_cgroup_flush_stats(void);
+
void __mod_memcg_lruvec_state(struct lruvec *lruvec, enum node_stat_item idx,
int val);
void __mod_lruvec_kmem_state(void *p, enum node_stat_item idx, int val);
@@ -1400,6 +1402,10 @@ static inline unsigned long lruvec_page_state_local(struct lruvec *lruvec,
return node_page_state(lruvec_pgdat(lruvec), idx);
}
+static inline void mem_cgroup_flush_stats(void)
+{
+}
+
static inline void __mod_memcg_lruvec_state(struct lruvec *lruvec,
enum node_stat_item idx, int val)
{
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 848d711bf576..39a00991fc80 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -103,6 +103,14 @@ static bool do_memsw_account(void)
return !cgroup_subsys_on_dfl(memory_cgrp_subsys) && !cgroup_memory_noswap;
}
+/* memcg and lruvec stats flushing */
+static void flush_memcg_stats_dwork(struct work_struct *w);
+static DECLARE_DEFERRABLE_WORK(stats_flush_dwork, flush_memcg_stats_dwork);
+static void flush_memcg_stats_work(struct work_struct *w);
+static DECLARE_WORK(stats_flush_work, flush_memcg_stats_work);
+static DEFINE_PER_CPU(unsigned int, stats_flush_threshold);
+static DEFINE_SPINLOCK(stats_flush_lock);
+
#define THRESHOLDS_EVENTS_TARGET 128
#define SOFTLIMIT_EVENTS_TARGET 1024
@@ -685,6 +693,8 @@ void __mod_memcg_lruvec_state(struct lruvec *lruvec, enum node_stat_item idx,
/* Update lruvec */
__this_cpu_add(pn->lruvec_stats_percpu->state[idx], val);
+ if (!(__this_cpu_inc_return(stats_flush_threshold) % MEMCG_CHARGE_BATCH))
+ queue_work(system_unbound_wq, &stats_flush_work);
}
/**
@@ -5248,6 +5258,10 @@ static int mem_cgroup_css_online(struct cgroup_subsys_state *css)
/* Online state pins memcg ID, memcg ID pins CSS */
refcount_set(&memcg->id.ref, 1);
css_get(css);
+
+ if (unlikely(mem_cgroup_is_root(memcg)))
+ queue_delayed_work(system_unbound_wq, &stats_flush_dwork,
+ 2UL*HZ);
return 0;
}
@@ -5339,6 +5353,26 @@ static void mem_cgroup_css_reset(struct cgroup_subsys_state *css)
memcg_wb_domain_size_changed(memcg);
}
+void mem_cgroup_flush_stats(void)
+{
+ if (!spin_trylock(&stats_flush_lock))
+ return;
+
+ cgroup_rstat_flush(root_mem_cgroup->css.cgroup);
+ spin_unlock(&stats_flush_lock);
+}
+
+static void flush_memcg_stats_dwork(struct work_struct *w)
+{
+ mem_cgroup_flush_stats();
+ queue_delayed_work(system_unbound_wq, &stats_flush_dwork, 2UL*HZ);
+}
+
+static void flush_memcg_stats_work(struct work_struct *w)
+{
+ mem_cgroup_flush_stats();
+}
+
static void mem_cgroup_css_rstat_flush(struct cgroup_subsys_state *css, int cpu)
{
struct mem_cgroup *memcg = mem_cgroup_from_css(css);
diff --git a/mm/vmscan.c b/mm/vmscan.c
index a7602f71ec04..1cc05ab8ca15 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2893,6 +2893,12 @@ static void shrink_node(pg_data_t *pgdat, struct scan_control *sc)
target_lruvec = mem_cgroup_lruvec(sc->target_mem_cgroup, pgdat);
again:
+ /*
+ * Flush the memory cgroup stats, so that we read accurate per-memcg
+ * lruvec stats for heuristics.
+ */
+ mem_cgroup_flush_stats();
+
memset(&sc->nr, 0, sizeof(sc->nr));
nr_reclaimed = sc->nr_reclaimed;
--
2.32.0.93.g670b81a890-goog
On Tue, Jul 13, 2021 at 1:24 PM Shakeel Butt <[email protected]> wrote:
>
> At the moment memcg stats are read in four contexts:
>
> 1. memcg stat user interfaces
> 2. dirty throttling
> 3. page fault
> 4. memory reclaim
>
> Currently the kernel flushes the stats for first two cases. Flushing the
> stats for remaining two casese may have performance impact. Always
> flushing the memcg stats on the page fault code path may negatively
> impacts the performance of the applications. In addition flushing in the
> memory reclaim code path, though treated as slowpath, can become the
> source of contention for the global lock taken for stat flushing because
> when system or memcg is under memory pressure, many tasks may enter the
> reclaim path.
>
> This patch uses following mechanisms to solve these challenges:
>
> 1. Periodically flush the stats from root memcg every 2 seconds. This
> will time limit the out of sync stats.
>
> 2. Asynchronously flush the stats after fixed number of stat updates.
> In the worst case the stat can be out of sync by O(nr_cpus * BATCH) for
> 2 seconds.
>
> 3. For avoiding thundering herd to flush the stats particularly from the
> memory reclaim context, introduce memcg local spinlock and let only one
> flusher active at a time. This could have been done through
> cgroup_rstat_lock lock but that lock is used by other subsystem and for
> userspace reading memcg stats. So, it is better to keep flushers
> introduced by this patch decoupled from cgroup_rstat_lock.
> ---
> Changes since v2:
> - Changed the subject of the patch
> - Added mechanism to bound errors to nr_cpus instead of nr_cgroups
> - memcg local lock to let one active flusher
>
> Changes since v1:
> - use system_unbound_wq for flushing the memcg stats
>
Forgot to add v3 in the subject for this patch.