From 19d3046bc5d682d2d478c95cf73f8e394e304d14 Mon Sep 17 00:00:00 2001
From: Huan Yang
Date: Fri, 25 Apr 2025 11:19:24 +0800
Subject: [PATCH] BACKPORT: mm/memcg: use kmem_cache when alloc memcg
When tracing mem_cgroup_alloc() with kmalloc ftrace, we observe:
kmalloc: call_site=mem_cgroup_css_alloc+0xd8/0x5b4 ptr=000000003e4c3799
bytes_req=2312 bytes_alloc=4096 gfp_flags=GFP_KERNEL|__GFP_ZERO node=-1
accounted=false
The output indicates that while allocating mem_cgroup struct (2312 bytes),
the slab allocator actually provides 4096-byte chunks. This occurs because:
1. The slab allocator predefines bucket sizes from 64B to 8096B
2. The mem_cgroup allocation size (2312B) falls between the 2KB and 4KB
slabs
3. The allocator rounds up to the nearest larger slab (4KB), resulting in
~1KB wasted memory per allocation
This patch introduces a dedicated kmem_cache for mem_cgroup structs,
achieving precise memory allocation. Post-patch ftrace verification shows:
kmem_cache_alloc: call_site=mem_cgroup_css_alloc+0xbc/0x5d4
ptr=00000000695c1806 bytes_req=2312 bytes_alloc=2368
gfp_flags=GFP_KERNEL|__GFP_ZERO node=-1 accounted=false
Each memcg alloc offer 2368bytes(include hw cacheline align), compare to
4096, avoid waste.
Link: https://lkml.kernel.org/r/20250425031935.76411-3-link@vivo.com
Signed-off-by: Huan Yang
Acked-by: Shakeel Butt
Acked-by: Johannes Weiner
Cc: Francesco Valla
Cc: guoweikang
Cc: Huang Shijie
Cc: KP Singh
Cc: Michal Hocko
Cc: Muchun Song
Cc: "Paul E . McKenney"
Cc: Petr Mladek
Cc: Rasmus Villemoes
Cc: Raul E Rangel
Cc: Roman Gushchin
Cc: "Uladzislau Rezki (Sony)"
Cc: Vlastimil Babka
Cc: Matthew Wilcox
Signed-off-by: Andrew Morton
(cherry picked from commit 97e4fc4b35dc1b98f28671fda35bd37d4f401bca)
[TJ: struct_size_t not available. Open code it with struct_size.]
Change-Id: Ib5c73e7af6eaf2b3066088566690cd0c6f4c796c
Bug: 417296244
Signed-off-by: T.J. Mercier
---
mm/memcontrol.c | 9 ++++++++-
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 434391a31387..b395871dd23c 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -90,6 +90,8 @@ static bool cgroup_memory_nosocket __ro_after_init;
/* Kernel memory accounting disabled? */
static bool cgroup_memory_nokmem __ro_after_init;
+static struct kmem_cache *memcg_cachep;
+
#ifdef CONFIG_CGROUP_WRITEBACK
static DECLARE_WAIT_QUEUE_HEAD(memcg_cgwb_frn_waitq);
#endif
@@ -5320,7 +5322,7 @@ static struct mem_cgroup *mem_cgroup_alloc(void)
int __maybe_unused i;
long error = -ENOMEM;
- memcg = kzalloc(struct_size(memcg, nodeinfo, nr_node_ids), GFP_KERNEL);
+ memcg = kmem_cache_zalloc(memcg_cachep, GFP_KERNEL);
if (!memcg)
return ERR_PTR(error);
@@ -7345,6 +7347,7 @@ __setup("cgroup.memory=", cgroup_memory);
*/
int __init mem_cgroup_init(void)
{
+ unsigned int memcg_size;
int cpu, node;
/*
@@ -7362,6 +7365,10 @@ int __init mem_cgroup_init(void)
INIT_WORK(&per_cpu_ptr(&memcg_stock, cpu)->work,
drain_local_stock);
+ memcg_size = struct_size((struct mem_cgroup *)NULL, nodeinfo, nr_node_ids);
+ memcg_cachep = kmem_cache_create("mem_cgroup", memcg_size, 0,
+ SLAB_PANIC | SLAB_HWCACHE_ALIGN, NULL);
+
for_each_node(node) {
struct mem_cgroup_tree_per_node *rtpn;