From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Wed, 23 Sep 2009 22:56:33 +0000 (-0700)
Subject: memcg: add comments explaining memory barriers
X-Git-Url: https://git.karo-electronics.de/?a=commitdiff_plain;h=261fb61a8bf6d3bd964ae6f1e6af49585d30db51;p=mv-sheeva.git

memcg: add comments explaining memory barriers

Add comments for the reason of smp_wmb() in mem_cgroup_commit_charge().

[akpm@linux-foundation.org: coding-style fixes]
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index b0757660663..eb9571815f0 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1134,6 +1134,13 @@ static void __mem_cgroup_commit_charge(struct mem_cgroup *mem,
 	}
 
 	pc->mem_cgroup = mem;
+	/*
+	 * We access a page_cgroup asynchronously without lock_page_cgroup().
+	 * Especially when a page_cgroup is taken from a page, pc->mem_cgroup
+	 * is accessed after testing USED bit. To make pc->mem_cgroup visible
+	 * before USED bit, we need memory barrier here.
+	 * See mem_cgroup_add_lru_list(), etc.
+ 	 */
 	smp_wmb();
 	switch (ctype) {
 	case MEM_CGROUP_CHARGE_TYPE_CACHE: