linux-kernel - [PATCH v2 3/5] mm/memcontrol: Reorder memcg_stock

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20230125073502.743446-4-leobras@redhat.com>
Date:   Wed, 25 Jan 2023 04:35:00 -0300
From:   Leonardo Bras <leobras@...hat.com>
To:     Johannes Weiner <hannes@...xchg.org>,
        Michal Hocko <mhocko@...nel.org>,
        Roman Gushchin <roman.gushchin@...ux.dev>,
        Shakeel Butt <shakeelb@...gle.com>,
        Muchun Song <muchun.song@...ux.dev>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Marcelo Tosatti <mtosatti@...hat.com>
Cc:     Leonardo Bras <leobras@...hat.com>, cgroups@...r.kernel.org,
        linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: [PATCH v2 3/5] mm/memcontrol: Reorder memcg_stock_pcp members to avoid holes

In 64-bit architectures, the current layout of memcg_stock_pcp should look
like this:

struct memcg_stock_pcp {
        spinlock_t                 stock_lock;           /*     0     4 */
        /* 4 bytes hole */
        struct mem_cgroup *        cached;               /*     8     8 */
        unsigned int               nr_pages;             /*    16     4 */
        /* 4 bytes hole */
	[...]
};

This happens because pointers will have 8 bytes (64-bit) and ints will have
4 bytes.

Those 2 holes are avoided if we reorder nr_pages and cached,
effectivelly putting nr_pages in the first hole, and saving 8 bytes.

Signed-off-by: Leonardo Bras <leobras@...hat.com>
---
 mm/memcontrol.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 1d5c108413c83..373fa78c4d881 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -2173,8 +2173,8 @@ void unlock_page_memcg(struct page *page)
 
 struct memcg_stock_pcp {
 	spinlock_t stock_lock; /* Protects the percpu struct */
-	struct mem_cgroup *cached; /* this never be root cgroup */
 	unsigned int nr_pages;
+	struct mem_cgroup *cached; /* this never be root cgroup */
 
 #ifdef CONFIG_MEMCG_KMEM
 	struct obj_cgroup *cached_objcg;
-- 
2.39.1