Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Zhongkun He <hezhongkun.hzk@bytedance.com>
To:     minchan@kernel.org, senozhatsky@chromium.org, mhocko@suse.com
Cc:     david@redhat.com, yosryahmed@google.com, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org,
        Zhongkun He <hezhongkun.hzk@bytedance.com>
Subject: [RFC PATCH v2 1/2] memcg: Add support for zram object charge
Date:   Mon, 24 Jul 2023 14:22:02 +0800
Message-Id: <20230724062202.2244188-1-hezhongkun.hzk@bytedance.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

The compressed RAM(ZRAM) is currently charged to kernel, not
to any memory cgroup, which can escape their cgroup memory
containment if the memory of a task is limited by memcgroup.
it will swap out the memory to zram swap device when the memory
is insufficient. In that case, the memory hard limit will be
invalidated. So, it should makes sense to charge the
compressed RAM to the page's memory cgroup.

As we know, zram can be used in two ways, direct and
indirect, this patchset can charge memory in both cases.
Direct zram usage by process within a cgroup will fail
to charge if there is no memory. Indirect zram usage by
process within a cgroup via swap in PF_MEMALLOC context,
will charge successfully.

Signed-off-by: Zhongkun He <hezhongkun.hzk@bytedance.com>
---
 include/linux/memcontrol.h | 12 ++++++++++++
 mm/memcontrol.c            | 24 ++++++++++++++++++++++++
 2 files changed, 36 insertions(+)

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 5818af8eca5a..24bac877bc83 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -1819,6 +1819,9 @@ static inline void count_objcg_event(struct obj_cgroup *objcg,
 	rcu_read_unlock();
 }
 
+int obj_cgroup_charge_zram(struct obj_cgroup *objcg, gfp_t gfp,
+					size_t size);
+void obj_cgroup_uncharge_zram(struct obj_cgroup *objcg, size_t size);
 #else
 static inline bool mem_cgroup_kmem_disabled(void)
 {
@@ -1880,6 +1883,15 @@ static inline void count_objcg_event(struct obj_cgroup *objcg,
 {
 }
 
+int obj_cgroup_charge_zram(struct obj_cgroup *objcg, gfp_t gfp,
+					size_t size)
+{
+	return 0;
+}
+
+void obj_cgroup_uncharge_zram(struct obj_cgroup *objcg, size_t size)
+{
+}
 #endif /* CONFIG_MEMCG_KMEM */
 
 #if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_ZSWAP)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index e8ca4bdcb03c..118544acf895 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3059,6 +3059,7 @@ struct obj_cgroup *get_obj_cgroup_from_page(struct page *page)
 	}
 	return objcg;
 }
+EXPORT_SYMBOL(get_obj_cgroup_from_page);
 
 static void memcg_account_kmem(struct mem_cgroup *memcg, int nr_pages)
 {
@@ -3409,6 +3410,29 @@ void obj_cgroup_uncharge(struct obj_cgroup *objcg, size_t size)
 	refill_obj_stock(objcg, size, true);
 }
 
+int obj_cgroup_charge_zram(struct obj_cgroup *objcg, gfp_t gfp,
+							size_t size)
+{
+	if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))
+		return 0;
+
+	/*
+	 * Indirect zram usage in PF_MEMALLOC, charging must succeed.
+	 * Direct zram usage, charging  may failed.
+	 */
+	return obj_cgroup_charge(objcg, gfp, size);
+}
+EXPORT_SYMBOL(obj_cgroup_charge_zram);
+
+void obj_cgroup_uncharge_zram(struct obj_cgroup *objcg, size_t size)
+{
+	if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))
+		return;
+
+	obj_cgroup_uncharge(objcg, size);
+}
+EXPORT_SYMBOL(obj_cgroup_uncharge_zram);
+
 #endif /* CONFIG_MEMCG_KMEM */
 
 /*
-- 
2.25.1