From 2a7106f2cb0768d00fe8c1eb42a754a7d8518f08 Mon Sep 17 00:00:00 2001 From: Greg Thelen Date: Thu, 13 Jan 2011 15:47:37 -0800 Subject: memcg: create extensible page stat update routines Replace usage of the mem_cgroup_update_file_mapped() memcg statistic update routine with two new routines: * mem_cgroup_inc_page_stat() * mem_cgroup_dec_page_stat() As before, only the file_mapped statistic is managed. However, these more general interfaces allow for new statistics to be more easily added. New statistics are added with memcg dirty page accounting. Signed-off-by: Greg Thelen Signed-off-by: Andrea Righi Acked-by: KAMEZAWA Hiroyuki Acked-by: Daisuke Nishimura Cc: Balbir Singh Cc: Minchan Kim Cc: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 31 ++++++++++++++++++++++++++++--- 1 file changed, 28 insertions(+), 3 deletions(-) (limited to 'include/linux/memcontrol.h') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 159a0762aea..067115ce6b3 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -25,6 +25,11 @@ struct page_cgroup; struct page; struct mm_struct; +/* Stats that can be updated by kernel. */ +enum mem_cgroup_page_stat_item { + MEMCG_NR_FILE_MAPPED, /* # of pages charged as file rss */ +}; + extern unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan, struct list_head *dst, unsigned long *scanned, int order, @@ -121,7 +126,22 @@ static inline bool mem_cgroup_disabled(void) return false; } -void mem_cgroup_update_file_mapped(struct page *page, int val); +void mem_cgroup_update_page_stat(struct page *page, + enum mem_cgroup_page_stat_item idx, + int val); + +static inline void mem_cgroup_inc_page_stat(struct page *page, + enum mem_cgroup_page_stat_item idx) +{ + mem_cgroup_update_page_stat(page, idx, 1); +} + +static inline void mem_cgroup_dec_page_stat(struct page *page, + enum mem_cgroup_page_stat_item idx) +{ + mem_cgroup_update_page_stat(page, idx, -1); +} + unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order, gfp_t gfp_mask); u64 mem_cgroup_get_limit(struct mem_cgroup *mem); @@ -293,8 +313,13 @@ mem_cgroup_print_oom_info(struct mem_cgroup *memcg, struct task_struct *p) { } -static inline void mem_cgroup_update_file_mapped(struct page *page, - int val) +static inline void mem_cgroup_inc_page_stat(struct page *page, + enum mem_cgroup_page_stat_item idx) +{ +} + +static inline void mem_cgroup_dec_page_stat(struct page *page, + enum mem_cgroup_page_stat_item idx) { } -- cgit v1.2.3-70-g09d2 From 50de1dd967d4ba3b8a90ebe7a4f5feca24191317 Mon Sep 17 00:00:00 2001 From: Daisuke Nishimura Date: Thu, 13 Jan 2011 15:47:43 -0800 Subject: memcg: fix memory migration of shmem swapcache In the current implementation mem_cgroup_end_migration() decides whether the page migration has succeeded or not by checking "oldpage->mapping". But if we are tring to migrate a shmem swapcache, the page->mapping of it is NULL from the begining, so the check would be invalid. As a result, mem_cgroup_end_migration() assumes the migration has succeeded even if it's not, so "newpage" would be freed while it's not uncharged. This patch fixes it by passing mem_cgroup_end_migration() the result of the page migration. Signed-off-by: Daisuke Nishimura Reviewed-by: Minchan Kim Acked-by: KAMEZAWA Hiroyuki Acked-by: Balbir Singh Cc: Minchan Kim Reviewed-by: Johannes Weiner Cc: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 5 ++--- mm/memcontrol.c | 5 ++--- mm/migrate.c | 2 +- 3 files changed, 5 insertions(+), 7 deletions(-) (limited to 'include/linux/memcontrol.h') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 067115ce6b3..6a576f98943 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -98,7 +98,7 @@ extern int mem_cgroup_prepare_migration(struct page *page, struct page *newpage, struct mem_cgroup **ptr); extern void mem_cgroup_end_migration(struct mem_cgroup *mem, - struct page *oldpage, struct page *newpage); + struct page *oldpage, struct page *newpage, bool migration_ok); /* * For memory reclaim. @@ -251,8 +251,7 @@ mem_cgroup_prepare_migration(struct page *page, struct page *newpage, } static inline void mem_cgroup_end_migration(struct mem_cgroup *mem, - struct page *oldpage, - struct page *newpage) + struct page *oldpage, struct page *newpage, bool migration_ok) { } diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 6424ba0fce8..8ab84103143 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -2896,7 +2896,7 @@ int mem_cgroup_prepare_migration(struct page *page, /* remove redundant charge if migration failed*/ void mem_cgroup_end_migration(struct mem_cgroup *mem, - struct page *oldpage, struct page *newpage) + struct page *oldpage, struct page *newpage, bool migration_ok) { struct page *used, *unused; struct page_cgroup *pc; @@ -2905,8 +2905,7 @@ void mem_cgroup_end_migration(struct mem_cgroup *mem, return; /* blocks rmdir() */ cgroup_exclude_rmdir(&mem->css); - /* at migration success, oldpage->mapping is NULL. */ - if (oldpage->mapping) { + if (!migration_ok) { used = oldpage; unused = newpage; } else { diff --git a/mm/migrate.c b/mm/migrate.c index 5b7d1fd2962..46fe8cc13d6 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -768,7 +768,7 @@ skip_unmap: uncharge: if (!charge) - mem_cgroup_end_migration(mem, page, newpage); + mem_cgroup_end_migration(mem, page, newpage, rc == 0); unlock: unlock_page(page); -- cgit v1.2.3-70-g09d2 From ca3e021417eed30ec2b64ce88eb0acf64aa9bc29 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Thu, 20 Jan 2011 14:44:24 -0800 Subject: memcg: fix USED bit handling at uncharge in THP Now, under THP: at charge: - PageCgroupUsed bit is set to all page_cgroup on a hugepage. ....set to 512 pages. at uncharge - PageCgroupUsed bit is unset on the head page. So, some pages will remain with "Used" bit. This patch fixes that Used bit is set only to the head page. Used bits for tail pages will be set at splitting if necessary. This patch adds this lock order: compound_lock() -> page_cgroup_move_lock(). [akpm@linux-foundation.org: fix warning] Signed-off-by: KAMEZAWA Hiroyuki Cc: Daisuke Nishimura Cc: Balbir Singh Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 9 +++++ mm/huge_memory.c | 2 + mm/memcontrol.c | 91 ++++++++++++++++++++++++++-------------------- 3 files changed, 62 insertions(+), 40 deletions(-) (limited to 'include/linux/memcontrol.h') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 6a576f98943..f512e189be5 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -146,6 +146,10 @@ unsigned long mem_cgroup_soft_limit_reclaim(struct zone *zone, int order, gfp_t gfp_mask); u64 mem_cgroup_get_limit(struct mem_cgroup *mem); +#ifdef CONFIG_TRANSPARENT_HUGEPAGE +void mem_cgroup_split_huge_fixup(struct page *head, struct page *tail); +#endif + #else /* CONFIG_CGROUP_MEM_RES_CTLR */ struct mem_cgroup; @@ -335,6 +339,11 @@ u64 mem_cgroup_get_limit(struct mem_cgroup *mem) return 0; } +static inline void mem_cgroup_split_huge_fixup(struct page *head, + struct page *tail) +{ +} + #endif /* CONFIG_CGROUP_MEM_CONT */ #endif /* _LINUX_MEMCONTROL_H */ diff --git a/mm/huge_memory.c b/mm/huge_memory.c index c4f634b3a48..e187454d82f 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1203,6 +1203,8 @@ static void __split_huge_page_refcount(struct page *page) BUG_ON(!PageDirty(page_tail)); BUG_ON(!PageSwapBacked(page_tail)); + mem_cgroup_split_huge_fixup(page, page_tail); + lru_add_page_tail(zone, page, page_tail); } diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 6d59a2bd520..848b42195e5 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1614,7 +1614,7 @@ void mem_cgroup_update_page_stat(struct page *page, if (unlikely(!mem || !PageCgroupUsed(pc))) goto out; /* pc->mem_cgroup is unstable ? */ - if (unlikely(mem_cgroup_stealed(mem))) { + if (unlikely(mem_cgroup_stealed(mem)) || PageTransHuge(page)) { /* take a lock against to access pc->mem_cgroup */ move_lock_page_cgroup(pc, &flags); need_unlock = true; @@ -2083,14 +2083,27 @@ struct mem_cgroup *try_get_mem_cgroup_from_page(struct page *page) return mem; } -/* - * commit a charge got by __mem_cgroup_try_charge() and makes page_cgroup to be - * USED state. If already USED, uncharge and return. - */ -static void ____mem_cgroup_commit_charge(struct mem_cgroup *mem, - struct page_cgroup *pc, - enum charge_type ctype) +static void __mem_cgroup_commit_charge(struct mem_cgroup *mem, + struct page_cgroup *pc, + enum charge_type ctype, + int page_size) { + int nr_pages = page_size >> PAGE_SHIFT; + + /* try_charge() can return NULL to *memcg, taking care of it. */ + if (!mem) + return; + + lock_page_cgroup(pc); + if (unlikely(PageCgroupUsed(pc))) { + unlock_page_cgroup(pc); + mem_cgroup_cancel_charge(mem, page_size); + return; + } + /* + * we don't need page_cgroup_lock about tail pages, becase they are not + * accessed by any other context at this point. + */ pc->mem_cgroup = mem; /* * We access a page_cgroup asynchronously without lock_page_cgroup(). @@ -2114,35 +2127,7 @@ static void ____mem_cgroup_commit_charge(struct mem_cgroup *mem, break; } - mem_cgroup_charge_statistics(mem, PageCgroupCache(pc), 1); -} - -static void __mem_cgroup_commit_charge(struct mem_cgroup *mem, - struct page_cgroup *pc, - enum charge_type ctype, - int page_size) -{ - int i; - int count = page_size >> PAGE_SHIFT; - - /* try_charge() can return NULL to *memcg, taking care of it. */ - if (!mem) - return; - - lock_page_cgroup(pc); - if (unlikely(PageCgroupUsed(pc))) { - unlock_page_cgroup(pc); - mem_cgroup_cancel_charge(mem, page_size); - return; - } - - /* - * we don't need page_cgroup_lock about tail pages, becase they are not - * accessed by any other context at this point. - */ - for (i = 0; i < count; i++) - ____mem_cgroup_commit_charge(mem, pc + i, ctype); - + mem_cgroup_charge_statistics(mem, PageCgroupCache(pc), nr_pages); unlock_page_cgroup(pc); /* * "charge_statistics" updated event counter. Then, check it. @@ -2152,6 +2137,34 @@ static void __mem_cgroup_commit_charge(struct mem_cgroup *mem, memcg_check_events(mem, pc->page); } +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + +#define PCGF_NOCOPY_AT_SPLIT ((1 << PCG_LOCK) | (1 << PCG_MOVE_LOCK) |\ + (1 << PCG_ACCT_LRU) | (1 << PCG_MIGRATION)) +/* + * Because tail pages are not marked as "used", set it. We're under + * zone->lru_lock, 'splitting on pmd' and compund_lock. + */ +void mem_cgroup_split_huge_fixup(struct page *head, struct page *tail) +{ + struct page_cgroup *head_pc = lookup_page_cgroup(head); + struct page_cgroup *tail_pc = lookup_page_cgroup(tail); + unsigned long flags; + + /* + * We have no races witch charge/uncharge but will have races with + * page state accounting. + */ + move_lock_page_cgroup(head_pc, &flags); + + tail_pc->mem_cgroup = head_pc->mem_cgroup; + smp_wmb(); /* see __commit_charge() */ + /* we don't need to copy all flags...*/ + tail_pc->flags = head_pc->flags & ~PCGF_NOCOPY_AT_SPLIT; + move_unlock_page_cgroup(head_pc, &flags); +} +#endif + /** * __mem_cgroup_move_account - move account of the page * @pc: page_cgroup of the page. @@ -2545,7 +2558,6 @@ direct_uncharge: static struct mem_cgroup * __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype) { - int i; int count; struct page_cgroup *pc; struct mem_cgroup *mem = NULL; @@ -2595,8 +2607,7 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype) break; } - for (i = 0; i < count; i++) - mem_cgroup_charge_statistics(mem, PageCgroupCache(pc), -1); + mem_cgroup_charge_statistics(mem, PageCgroupCache(pc), -count); ClearPageCgroupUsed(pc); /* -- cgit v1.2.3-70-g09d2