The patch titled
memcg: fix race in file_mapped accounting
has been removed from the -mm tree. Its filename was
memcg-fix-race-in-file_mapped-accounting.patch
This patch was dropped because it was merged into mainline or a subsystem tree
The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/
------------------------------------------------------
Subject: memcg: fix race in file_mapped accounting
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Presently, memcg's FILE_MAPPED accounting has following race with
move_account (happens at rmdir()).
increment page->mapcount (rmap.c)
mem_cgroup_update_file_mapped() move_account()
lock_page_cgroup()
check page_mapped() if
page_mapped(page)>1 {
FILE_MAPPED -1 from old memcg
FILE_MAPPED +1 to old memcg
}
.....
overwrite pc->mem_cgroup
unlock_page_cgroup()
lock_page_cgroup()
FILE_MAPPED + 1 to pc->mem_cgroup
unlock_page_cgroup()
Then,
old memcg (-1 file mapped)
new memcg (+2 file mapped)
This happens because move_account see page_mapped() which is not guarded
by lock_page_cgroup(). This patch adds FILE_MAPPED flag to page_cgroup
and move account information based on it. Now, all checks are synchronous
with lock_page_cgroup().
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Reviewed-by: Balbir Singh <balbir@xxxxxxxxxx>
Reviewed-by: Daisuke Nishimura <nishimura@xxxxxxxxxxxxxxxxx>
Cc: Andrea Righi <arighi@xxxxxxxxxxx>
Cc: Andrea Arcangeli <aarcange@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/page_cgroup.h | 6 ++++++
mm/memcontrol.c | 18 +++++++++---------
2 files changed, 15 insertions(+), 9 deletions(-)
diff -puN include/linux/page_cgroup.h~memcg-fix-race-in-file_mapped-accounting include/linux/page_cgroup.h
--- a/include/linux/page_cgroup.h~memcg-fix-race-in-file_mapped-accounting
+++ a/include/linux/page_cgroup.h
@@ -39,6 +39,7 @@ enum {
PCG_CACHE, /* charged as cache */
PCG_USED, /* this object is in use. */
PCG_ACCT_LRU, /* page has been accounted for */
+ PCG_FILE_MAPPED, /* page is accounted as "mapped" */
};
#define TESTPCGFLAG(uname, lname) \
@@ -73,6 +74,11 @@ CLEARPCGFLAG(AcctLRU, ACCT_LRU)
TESTPCGFLAG(AcctLRU, ACCT_LRU)
TESTCLEARPCGFLAG(AcctLRU, ACCT_LRU)
+
+SETPCGFLAG(FileMapped, FILE_MAPPED)
+CLEARPCGFLAG(FileMapped, FILE_MAPPED)
+TESTPCGFLAG(FileMapped, FILE_MAPPED)
+
static inline int page_cgroup_nid(struct page_cgroup *pc)
{
return page_to_nid(pc->page);
diff -puN mm/memcontrol.c~memcg-fix-race-in-file_mapped-accounting mm/memcontrol.c
--- a/mm/memcontrol.c~memcg-fix-race-in-file_mapped-accounting
+++ a/mm/memcontrol.c
@@ -1359,16 +1359,19 @@ void mem_cgroup_update_file_mapped(struc
lock_page_cgroup(pc);
mem = pc->mem_cgroup;
- if (!mem)
- goto done;
-
- if (!PageCgroupUsed(pc))
+ if (!mem || !PageCgroupUsed(pc))
goto done;
/*
* Preemption is already disabled. We can use __this_cpu_xxx
*/
- __this_cpu_add(mem->stat->count[MEM_CGROUP_STAT_FILE_MAPPED], val);
+ if (val > 0) {
+ __this_cpu_inc(mem->stat->count[MEM_CGROUP_STAT_FILE_MAPPED]);
+ SetPageCgroupFileMapped(pc);
+ } else {
+ __this_cpu_dec(mem->stat->count[MEM_CGROUP_STAT_FILE_MAPPED]);
+ ClearPageCgroupFileMapped(pc);
+ }
done:
unlock_page_cgroup(pc);
@@ -1801,16 +1804,13 @@ static void __mem_cgroup_commit_charge(s
static void __mem_cgroup_move_account(struct page_cgroup *pc,
struct mem_cgroup *from, struct mem_cgroup *to, bool uncharge)
{
- struct page *page;
-
VM_BUG_ON(from == to);
VM_BUG_ON(PageLRU(pc->page));
VM_BUG_ON(!PageCgroupLocked(pc));
VM_BUG_ON(!PageCgroupUsed(pc));
VM_BUG_ON(pc->mem_cgroup != from);
- page = pc->page;
- if (page_mapped(page) && !PageAnon(page)) {
+ if (PageCgroupFileMapped(pc)) {
/* Update mapped_file data for mem_cgroup */
preempt_disable();
__this_cpu_dec(from->stat->count[MEM_CGROUP_STAT_FILE_MAPPED]);
_
Patches currently in -mm which might be from kamezawa.hiroyu@xxxxxxxxxxxxxx are
origin.patch
linux-next.patch
vfs-introduce-fmode_neg_offset-for-allowing-negative-f_pos.patch
mm-remove-return-value-of-putback_lru_pages.patch
oom-filter-tasks-not-sharing-the-same-cpuset.patch
oom-sacrifice-child-with-highest-badness-score-for-parent.patch
oom-select-task-from-tasklist-for-mempolicy-ooms.patch
oom-remove-special-handling-for-pagefault-ooms.patch
oom-badness-heuristic-rewrite.patch
oom-deprecate-oom_adj-tunable.patch
oom-replace-sysctls-with-quick-mode.patch
oom-avoid-oom-killer-for-lowmem-allocations.patch
oom-remove-unnecessary-code-and-cleanup.patch
oom-default-to-killing-current-for-pagefault-ooms.patch
oom-avoid-race-for-oom-killed-tasks-detaching-mm-prior-to-exit.patch
oom-hold-tasklist_lock-when-dumping-tasks.patch
oom-give-current-access-to-memory-reserves-if-it-has-been-killed.patch
oom-avoid-sending-exiting-tasks-a-sigkill.patch
oom-clean-up-oom_kill_task.patch
oom-clean-up-oom_badness.patch
mm-default-to-node-zonelist-ordering-when-nodes-have-only-lowmem.patch
mm-migration-take-a-reference-to-the-anon_vma-before-migrating.patch
mm-migration-do-not-try-to-migrate-unmapped-anonymous-pages.patch
mm-share-the-anon_vma-ref-counts-between-ksm-and-page-migration.patch
mm-allow-config_migration-to-be-set-without-config_numa-or-memory-hot-remove.patch
mm-export-unusable-free-space-index-via-proc-unusable_index.patch
mm-export-fragmentation-index-via-proc-extfrag_index.patch
mm-move-definition-for-lru-isolation-modes-to-a-header.patch
mm-compaction-memory-compaction-core.patch
mm-compaction-add-proc-trigger-for-memory-compaction.patch
mm-compaction-add-sys-trigger-for-per-node-memory-compaction.patch
mm-compaction-direct-compact-when-a-high-order-allocation-fails.patch
mm-compaction-add-a-tunable-that-decides-when-memory-should-be-compacted-and-when-it-should-be-reclaimed.patch
mm-compaction-do-not-compact-within-a-preferred-zone-after-a-compaction-failure.patch
mm-migration-allow-the-migration-of-pageswapcache-pages.patch
memcg-oom-wakeup-filter.patch
memcg-oom-wakeup-filter-update.patch
memcg-oom-notifier.patch
memcg-oom-notifier-update.patch
memcg-oom-kill-disable-and-oom-status.patch
memcg-oom-kill-disable-and-oom-status-update.patch
memcg-oom-kill-disable-and-oom-status-update-checkpatch-fixes.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
