From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-15.7 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id DEC35C433B4 for ; Fri, 30 Apr 2021 05:56:28 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id BA1ED6147D for ; Fri, 30 Apr 2021 05:56:28 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230229AbhD3F5P (ORCPT ); Fri, 30 Apr 2021 01:57:15 -0400 Received: from mail.kernel.org ([198.145.29.99]:49828 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229591AbhD3F5P (ORCPT ); Fri, 30 Apr 2021 01:57:15 -0400 Received: by mail.kernel.org (Postfix) with ESMTPSA id 1497B61476; Fri, 30 Apr 2021 05:56:27 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linux-foundation.org; s=korg; t=1619762187; bh=3OLzqtgjsYC5QJlX0DKbcXl84augKY2xXUtr7ESf6Fw=; h=Date:From:To:Subject:In-Reply-To:From; b=a9nV9fw8rZwKLxNjZyuMYoLRcBUXlscFD/nmsXXBkn69flAIlRZp5uGus/iuEHkoE V9xa1P8uPS4JcRTRaLe0ssnxBqrbfOa+5wqkfsDPaue6XwgoX8Ou4r9UAzYVDZE2dy v1smxlFEmVgloQgpYtcnkQaSaWVGI/yLRLz22qa0= Date: Thu, 29 Apr 2021 22:56:26 -0700 From: Andrew Morton To: akpm@linux-foundation.org, bsingharora@gmail.com, guro@fb.com, hannes@cmpxchg.org, linux-mm@kvack.org, mhocko@suse.com, mkoutny@suse.com, mm-commits@vger.kernel.org, shakeelb@google.com, tj@kernel.org, torvalds@linux-foundation.org Subject: [patch 064/178] mm: memcontrol: switch to rstat Message-ID: <20210430055626.WZK5r2afb%akpm@linux-foundation.org> In-Reply-To: <20210429225251.02b6386d21b69255b4f6c163@linux-foundation.org> User-Agent: s-nail v14.8.16 MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Precedence: bulk Reply-To: linux-kernel@vger.kernel.org List-ID: X-Mailing-List: mm-commits@vger.kernel.org =46rom: Johannes Weiner Subject: mm: memcontrol: switch to rstat Replace the memory controller's custom hierarchical stats code with the generic rstat infrastructure provided by the cgroup core. The current implementation does batched upward propagation from the write side (i.e. as stats change). The per-cpu batches introduce an error, which is multiplied by the number of subgroups in a tree. In systems with many CPUs and sizable cgroup trees, the error can be large enough to confuse users (e.g. 32 batch pages * 32 CPUs * 32 subgroups results in an error of up to 128M per stat item). This can entirely swallow allocation bursts inside a workload that the user is expecting to see reflected in the statistics. In the past, we've done read-side aggregation, where a memory.stat read would have to walk the entire subtree and add up per-cpu counts. This became problematic with lazily-freed cgroups: we could have large subtrees where most cgroups were entirely idle. Hence the switch to change-driven upward propagation. Unfortunately, it needed to trade accuracy for speed due to the write side being so hot. Rstat combines the best of both worlds: from the write side, it cheaply maintains a queue of cgroups that have pending changes, so that the read side can do selective tree aggregation. This way the reported stats will always be precise and recent as can be, while the aggregation can skip over potentially large numbers of idle cgroups. The way rstat works is that it implements a tree for tracking cgroups with pending local changes, as well as a flush function that walks the tree upwards. The controller then drives this by 1) telling rstat when a local cgroup stat changes (e.g. mod_memcg_state) and 2) when a flush is required to get uptodate hierarchy stats for a given subtree (e.g. when memory.stat is read). The controller also provides a flush callback that is called during the rstat flush walk for each cgroup and aggregates its local per-cpu counters and propagates them upwards. This adds a second vmstats to struct mem_cgroup (MEMCG_NR_STAT + NR_VM_EVENT_ITEMS) to track pending subtree deltas during upward aggregation. It removes 3 words from the per-cpu data. It eliminates memcg_exact_page_state(), since memcg_page_state() is now exact. [akpm@linux-foundation.org: merge fix] [hannes@cmpxchg.org: fix a sleep in atomic section problem] Link: https://lkml.kernel.org/r/20210315234100.64307-1-hannes@cmpxchg.org Link: https://lkml.kernel.org/r/20210209163304.77088-7-hannes@cmpxchg.org Signed-off-by: Johannes Weiner Reviewed-by: Roman Gushchin Acked-by: Michal Hocko Reviewed-by: Shakeel Butt Reviewed-by: Michal Koutn=C3=BD Acked-by: Balbir Singh Cc: Tejun Heo Signed-off-by: Andrew Morton --- include/linux/memcontrol.h | 67 ++++++---- mm/memcontrol.c | 218 +++++++++++++---------------------- 2 files changed, 127 insertions(+), 158 deletions(-) --- a/include/linux/memcontrol.h~mm-memcontrol-switch-to-rstat +++ a/include/linux/memcontrol.h @@ -76,10 +76,27 @@ enum mem_cgroup_events_target { }; =20 struct memcg_vmstats_percpu { - long stat[MEMCG_NR_STAT]; - unsigned long events[NR_VM_EVENT_ITEMS]; - unsigned long nr_page_events; - unsigned long targets[MEM_CGROUP_NTARGETS]; + /* Local (CPU and cgroup) page state & events */ + long state[MEMCG_NR_STAT]; + unsigned long events[NR_VM_EVENT_ITEMS]; + + /* Delta calculation for lockless upward propagation */ + long state_prev[MEMCG_NR_STAT]; + unsigned long events_prev[NR_VM_EVENT_ITEMS]; + + /* Cgroup1: threshold notifications & softlimit tree updates */ + unsigned long nr_page_events; + unsigned long targets[MEM_CGROUP_NTARGETS]; +}; + +struct memcg_vmstats { + /* Aggregated (CPU and subtree) page state & events */ + long state[MEMCG_NR_STAT]; + unsigned long events[NR_VM_EVENT_ITEMS]; + + /* Pending child counts during tree propagation */ + long state_pending[MEMCG_NR_STAT]; + unsigned long events_pending[NR_VM_EVENT_ITEMS]; }; =20 struct mem_cgroup_reclaim_iter { @@ -287,8 +304,8 @@ struct mem_cgroup { =20 MEMCG_PADDING(_pad1_); =20 - atomic_long_t vmstats[MEMCG_NR_STAT]; - atomic_long_t vmevents[NR_VM_EVENT_ITEMS]; + /* memory.stat */ + struct memcg_vmstats vmstats; =20 /* memory.events */ atomic_long_t memory_events[MEMCG_NR_MEMORY_EVENTS]; @@ -315,10 +332,6 @@ struct mem_cgroup { atomic_t moving_account; struct task_struct *move_lock_task; =20 - /* Legacy local VM stats and events */ - struct memcg_vmstats_percpu __percpu *vmstats_local; - - /* Subtree VM stats and events (batched updates) */ struct memcg_vmstats_percpu __percpu *vmstats_percpu; =20 #ifdef CONFIG_CGROUP_WRITEBACK @@ -939,10 +952,6 @@ static inline void mod_memcg_lruvec_stat local_irq_restore(flags); } =20 -unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order, - gfp_t gfp_mask, - unsigned long *total_scanned); - void __count_memcg_events(struct mem_cgroup *memcg, enum vm_event_item idx, unsigned long count); =20 @@ -1023,6 +1032,10 @@ static inline void memcg_memory_event_mm =20 void split_page_memcg(struct page *head, unsigned int nr); =20 +unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order, + gfp_t gfp_mask, + unsigned long *total_scanned); + #else /* CONFIG_MEMCG */ =20 #define MEM_CGROUP_ID_SHIFT 0 @@ -1131,6 +1144,10 @@ static inline bool lruvec_holds_page_lru return lruvec =3D=3D &pgdat->__lruvec; } =20 +static inline void lruvec_memcg_debug(struct lruvec *lruvec, struct page *= page) +{ +} + static inline struct mem_cgroup *parent_mem_cgroup(struct mem_cgroup *memc= g) { return NULL; @@ -1334,18 +1351,6 @@ static inline void mod_lruvec_kmem_state mod_node_page_state(page_pgdat(page), idx, val); } =20 -static inline -unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order, - gfp_t gfp_mask, - unsigned long *total_scanned) -{ - return 0; -} - -static inline void split_page_memcg(struct page *head, unsigned int nr) -{ -} - static inline void count_memcg_events(struct mem_cgroup *memcg, enum vm_event_item idx, unsigned long count) @@ -1368,9 +1373,17 @@ void count_memcg_event_mm(struct mm_stru { } =20 -static inline void lruvec_memcg_debug(struct lruvec *lruvec, struct page *= page) +static inline void split_page_memcg(struct page *head, unsigned int nr) { } + +static inline +unsigned long mem_cgroup_soft_limit_reclaim(pg_data_t *pgdat, int order, + gfp_t gfp_mask, + unsigned long *total_scanned) +{ + return 0; +} #endif /* CONFIG_MEMCG */ =20 static inline void __inc_lruvec_kmem_state(void *p, enum node_stat_item id= x) --- a/mm/memcontrol.c~mm-memcontrol-switch-to-rstat +++ a/mm/memcontrol.c @@ -765,37 +765,17 @@ mem_cgroup_largest_soft_limit_node(struc */ void __mod_memcg_state(struct mem_cgroup *memcg, int idx, int val) { - long x, threshold =3D MEMCG_CHARGE_BATCH; - if (mem_cgroup_disabled()) return; =20 - if (memcg_stat_item_in_bytes(idx)) - threshold <<=3D PAGE_SHIFT; - - x =3D val + __this_cpu_read(memcg->vmstats_percpu->stat[idx]); - if (unlikely(abs(x) > threshold)) { - struct mem_cgroup *mi; - - /* - * Batch local counters to keep them in sync with - * the hierarchical ones. - */ - __this_cpu_add(memcg->vmstats_local->stat[idx], x); - for (mi =3D memcg; mi; mi =3D parent_mem_cgroup(mi)) - atomic_long_add(x, &mi->vmstats[idx]); - x =3D 0; - } - __this_cpu_write(memcg->vmstats_percpu->stat[idx], x); + __this_cpu_add(memcg->vmstats_percpu->state[idx], val); + cgroup_rstat_updated(memcg->css.cgroup, smp_processor_id()); } =20 -/* - * idx can be of type enum memcg_stat_item or node_stat_item. - * Keep in sync with memcg_exact_page_state(). - */ +/* idx can be of type enum memcg_stat_item or node_stat_item. */ static unsigned long memcg_page_state(struct mem_cgroup *memcg, int idx) { - long x =3D atomic_long_read(&memcg->vmstats[idx]); + long x =3D READ_ONCE(memcg->vmstats.state[idx]); #ifdef CONFIG_SMP if (x < 0) x =3D 0; @@ -803,17 +783,14 @@ static unsigned long memcg_page_state(st return x; } =20 -/* - * idx can be of type enum memcg_stat_item or node_stat_item. - * Keep in sync with memcg_exact_page_state(). - */ +/* idx can be of type enum memcg_stat_item or node_stat_item. */ static unsigned long memcg_page_state_local(struct mem_cgroup *memcg, int = idx) { long x =3D 0; int cpu; =20 for_each_possible_cpu(cpu) - x +=3D per_cpu(memcg->vmstats_local->stat[idx], cpu); + x +=3D per_cpu(memcg->vmstats_percpu->state[idx], cpu); #ifdef CONFIG_SMP if (x < 0) x =3D 0; @@ -936,30 +913,16 @@ void __mod_lruvec_kmem_state(void *p, en void __count_memcg_events(struct mem_cgroup *memcg, enum vm_event_item idx, unsigned long count) { - unsigned long x; - if (mem_cgroup_disabled()) return; =20 - x =3D count + __this_cpu_read(memcg->vmstats_percpu->events[idx]); - if (unlikely(x > MEMCG_CHARGE_BATCH)) { - struct mem_cgroup *mi; - - /* - * Batch local counters to keep them in sync with - * the hierarchical ones. - */ - __this_cpu_add(memcg->vmstats_local->events[idx], x); - for (mi =3D memcg; mi; mi =3D parent_mem_cgroup(mi)) - atomic_long_add(x, &mi->vmevents[idx]); - x =3D 0; - } - __this_cpu_write(memcg->vmstats_percpu->events[idx], x); + __this_cpu_add(memcg->vmstats_percpu->events[idx], count); + cgroup_rstat_updated(memcg->css.cgroup, smp_processor_id()); } =20 static unsigned long memcg_events(struct mem_cgroup *memcg, int event) { - return atomic_long_read(&memcg->vmevents[event]); + return READ_ONCE(memcg->vmstats.events[event]); } =20 static unsigned long memcg_events_local(struct mem_cgroup *memcg, int even= t) @@ -968,7 +931,7 @@ static unsigned long memcg_events_local( int cpu; =20 for_each_possible_cpu(cpu) - x +=3D per_cpu(memcg->vmstats_local->events[event], cpu); + x +=3D per_cpu(memcg->vmstats_percpu->events[event], cpu); return x; } =20 @@ -1604,6 +1567,7 @@ static char *memory_stat_format(struct m * * Current memory state: */ + cgroup_rstat_flush(memcg->css.cgroup); =20 for (i =3D 0; i < ARRAY_SIZE(memory_stats); i++) { u64 size; @@ -2409,22 +2373,11 @@ static int memcg_hotplug_cpu_dead(unsign drain_stock(stock); =20 for_each_mem_cgroup(memcg) { - struct memcg_vmstats_percpu *statc; int i; =20 - statc =3D per_cpu_ptr(memcg->vmstats_percpu, cpu); - - for (i =3D 0; i < MEMCG_NR_STAT; i++) { + for (i =3D 0; i < NR_VM_NODE_STAT_ITEMS; i++) { int nid; =20 - if (statc->stat[i]) { - mod_memcg_state(memcg, i, statc->stat[i]); - statc->stat[i] =3D 0; - } - - if (i >=3D NR_VM_NODE_STAT_ITEMS) - continue; - for_each_node(nid) { struct batched_lruvec_stat *lstatc; struct mem_cgroup_per_node *pn; @@ -2443,13 +2396,6 @@ static int memcg_hotplug_cpu_dead(unsign } } } - - for (i =3D 0; i < NR_VM_EVENT_ITEMS; i++) { - if (statc->events[i]) { - count_memcg_events(memcg, i, statc->events[i]); - statc->events[i] =3D 0; - } - } } =20 return 0; @@ -3572,6 +3518,7 @@ static unsigned long mem_cgroup_usage(st unsigned long val; =20 if (mem_cgroup_is_root(memcg)) { + cgroup_rstat_flush(memcg->css.cgroup); val =3D memcg_page_state(memcg, NR_FILE_PAGES) + memcg_page_state(memcg, NR_ANON_MAPPED); if (swap) @@ -3636,26 +3583,15 @@ static u64 mem_cgroup_read_u64(struct cg } } =20 -static void memcg_flush_percpu_vmstats(struct mem_cgroup *memcg) +static void memcg_flush_lruvec_page_state(struct mem_cgroup *memcg) { - unsigned long stat[MEMCG_NR_STAT] =3D {0}; - struct mem_cgroup *mi; - int node, cpu, i; - - for_each_online_cpu(cpu) - for (i =3D 0; i < MEMCG_NR_STAT; i++) - stat[i] +=3D per_cpu(memcg->vmstats_percpu->stat[i], cpu); - - for (mi =3D memcg; mi; mi =3D parent_mem_cgroup(mi)) - for (i =3D 0; i < MEMCG_NR_STAT; i++) - atomic_long_add(stat[i], &mi->vmstats[i]); + int node; =20 for_each_node(node) { struct mem_cgroup_per_node *pn =3D memcg->nodeinfo[node]; + unsigned long stat[NR_VM_NODE_STAT_ITEMS] =3D { 0 }; struct mem_cgroup_per_node *pi; - - for (i =3D 0; i < NR_VM_NODE_STAT_ITEMS; i++) - stat[i] =3D 0; + int cpu, i; =20 for_each_online_cpu(cpu) for (i =3D 0; i < NR_VM_NODE_STAT_ITEMS; i++) @@ -3668,25 +3604,6 @@ static void memcg_flush_percpu_vmstats(s } } =20 -static void memcg_flush_percpu_vmevents(struct mem_cgroup *memcg) -{ - unsigned long events[NR_VM_EVENT_ITEMS]; - struct mem_cgroup *mi; - int cpu, i; - - for (i =3D 0; i < NR_VM_EVENT_ITEMS; i++) - events[i] =3D 0; - - for_each_online_cpu(cpu) - for (i =3D 0; i < NR_VM_EVENT_ITEMS; i++) - events[i] +=3D per_cpu(memcg->vmstats_percpu->events[i], - cpu); - - for (mi =3D memcg; mi; mi =3D parent_mem_cgroup(mi)) - for (i =3D 0; i < NR_VM_EVENT_ITEMS; i++) - atomic_long_add(events[i], &mi->vmevents[i]); -} - #ifdef CONFIG_MEMCG_KMEM static int memcg_online_kmem(struct mem_cgroup *memcg) { @@ -4003,6 +3920,8 @@ static int memcg_numa_stat_show(struct s int nid; struct mem_cgroup *memcg =3D mem_cgroup_from_seq(m); =20 + cgroup_rstat_flush(memcg->css.cgroup); + for (stat =3D stats; stat < stats + ARRAY_SIZE(stats); stat++) { seq_printf(m, "%s=3D%lu", stat->name, mem_cgroup_nr_lru_pages(memcg, stat->lru_mask, @@ -4073,6 +3992,8 @@ static int memcg_stat_show(struct seq_fi =20 BUILD_BUG_ON(ARRAY_SIZE(memcg1_stat_names) !=3D ARRAY_SIZE(memcg1_stats)); =20 + cgroup_rstat_flush(memcg->css.cgroup); + for (i =3D 0; i < ARRAY_SIZE(memcg1_stats); i++) { unsigned long nr; =20 @@ -4549,22 +4470,6 @@ struct wb_domain *mem_cgroup_wb_domain(s return &memcg->cgwb_domain; } =20 -/* - * idx can be of type enum memcg_stat_item or node_stat_item. - * Keep in sync with memcg_exact_page(). - */ -static unsigned long memcg_exact_page_state(struct mem_cgroup *memcg, int = idx) -{ - long x =3D atomic_long_read(&memcg->vmstats[idx]); - int cpu; - - for_each_online_cpu(cpu) - x +=3D per_cpu_ptr(memcg->vmstats_percpu, cpu)->stat[idx]; - if (x < 0) - x =3D 0; - return x; -} - /** * mem_cgroup_wb_stats - retrieve writeback related stats from its memcg * @wb: bdi_writeback in question @@ -4590,13 +4495,14 @@ void mem_cgroup_wb_stats(struct bdi_writ struct mem_cgroup *memcg =3D mem_cgroup_from_css(wb->memcg_css); struct mem_cgroup *parent; =20 - *pdirty =3D memcg_exact_page_state(memcg, NR_FILE_DIRTY); + cgroup_rstat_flush_irqsafe(memcg->css.cgroup); =20 - *pwriteback =3D memcg_exact_page_state(memcg, NR_WRITEBACK); - *pfilepages =3D memcg_exact_page_state(memcg, NR_INACTIVE_FILE) + - memcg_exact_page_state(memcg, NR_ACTIVE_FILE); - *pheadroom =3D PAGE_COUNTER_MAX; + *pdirty =3D memcg_page_state(memcg, NR_FILE_DIRTY); + *pwriteback =3D memcg_page_state(memcg, NR_WRITEBACK); + *pfilepages =3D memcg_page_state(memcg, NR_INACTIVE_FILE) + + memcg_page_state(memcg, NR_ACTIVE_FILE); =20 + *pheadroom =3D PAGE_COUNTER_MAX; while ((parent =3D parent_mem_cgroup(memcg))) { unsigned long ceiling =3D min(READ_ONCE(memcg->memory.max), READ_ONCE(memcg->memory.high)); @@ -5228,7 +5134,6 @@ static void __mem_cgroup_free(struct mem for_each_node(node) free_mem_cgroup_per_node_info(memcg, node); free_percpu(memcg->vmstats_percpu); - free_percpu(memcg->vmstats_local); kfree(memcg); } =20 @@ -5236,11 +5141,10 @@ static void mem_cgroup_free(struct mem_c { memcg_wb_domain_exit(memcg); /* - * Flush percpu vmstats and vmevents to guarantee the value correctness - * on parent's and all ancestor levels. + * Flush percpu lruvec stats to guarantee the value + * correctness on parent's and all ancestor levels. */ - memcg_flush_percpu_vmstats(memcg); - memcg_flush_percpu_vmevents(memcg); + memcg_flush_lruvec_page_state(memcg); __mem_cgroup_free(memcg); } =20 @@ -5267,11 +5171,6 @@ static struct mem_cgroup *mem_cgroup_all goto fail; } =20 - memcg->vmstats_local =3D alloc_percpu_gfp(struct memcg_vmstats_percpu, - GFP_KERNEL_ACCOUNT); - if (!memcg->vmstats_local) - goto fail; - memcg->vmstats_percpu =3D alloc_percpu_gfp(struct memcg_vmstats_percpu, GFP_KERNEL_ACCOUNT); if (!memcg->vmstats_percpu) @@ -5471,6 +5370,62 @@ static void mem_cgroup_css_reset(struct memcg_wb_domain_size_changed(memcg); } =20 +static void mem_cgroup_css_rstat_flush(struct cgroup_subsys_state *css, in= t cpu) +{ + struct mem_cgroup *memcg =3D mem_cgroup_from_css(css); + struct mem_cgroup *parent =3D parent_mem_cgroup(memcg); + struct memcg_vmstats_percpu *statc; + long delta, v; + int i; + + statc =3D per_cpu_ptr(memcg->vmstats_percpu, cpu); + + for (i =3D 0; i < MEMCG_NR_STAT; i++) { + /* + * Collect the aggregated propagation counts of groups + * below us. We're in a per-cpu loop here and this is + * a global counter, so the first cycle will get them. + */ + delta =3D memcg->vmstats.state_pending[i]; + if (delta) + memcg->vmstats.state_pending[i] =3D 0; + + /* Add CPU changes on this level since the last flush */ + v =3D READ_ONCE(statc->state[i]); + if (v !=3D statc->state_prev[i]) { + delta +=3D v - statc->state_prev[i]; + statc->state_prev[i] =3D v; + } + + if (!delta) + continue; + + /* Aggregate counts on this level and propagate upwards */ + memcg->vmstats.state[i] +=3D delta; + if (parent) + parent->vmstats.state_pending[i] +=3D delta; + } + + for (i =3D 0; i < NR_VM_EVENT_ITEMS; i++) { + delta =3D memcg->vmstats.events_pending[i]; + if (delta) + memcg->vmstats.events_pending[i] =3D 0; + + v =3D READ_ONCE(statc->events[i]); + if (v !=3D statc->events_prev[i]) { + delta +=3D v - statc->events_prev[i]; + statc->events_prev[i] =3D v; + } + + if (!delta) + continue; + + memcg->vmstats.events[i] +=3D delta; + if (parent) + parent->vmstats.events_pending[i] +=3D delta; + } +} + #ifdef CONFIG_MMU /* Handlers for move charge at task migration. */ static int mem_cgroup_do_precharge(unsigned long count) @@ -6524,6 +6479,7 @@ struct cgroup_subsys memory_cgrp_subsys .css_released =3D mem_cgroup_css_released, .css_free =3D mem_cgroup_css_free, .css_reset =3D mem_cgroup_css_reset, + .css_rstat_flush =3D mem_cgroup_css_rstat_flush, .can_attach =3D mem_cgroup_can_attach, .cancel_attach =3D mem_cgroup_cancel_attach, .post_attach =3D mem_cgroup_move_task, _