Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Muchun Song <songmuchun@bytedance.com>
To:     hannes@cmpxchg.org, mhocko@kernel.org, vdavydov.dev@gmail.com,
        akpm@linux-foundation.org
Cc:     cgroups@vger.kernel.org, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org,
        Muchun Song <songmuchun@bytedance.com>
Subject: [PATCH 2/4] mm: memcontrol: add missing memcg_oom_recover() when uncharge slab page
Date:   Sat, 13 Feb 2021 01:01:57 +0800
Message-Id: <20210212170159.32153-2-songmuchun@bytedance.com>
In-Reply-To: <20210212170159.32153-1-songmuchun@bytedance.com>
References: <20210212170159.32153-1-songmuchun@bytedance.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

When we uncharge a page, we wake up oom victims when the memcg oom
handling is outsourced to the userspace. The uncharge_batch do that
for normal and kmem pages but not slab pages. It is likely an
omission. So add the missing memcg_oom_recover() to
__memcg_kmem_uncharge(). And the function of memory.oom_control
is only suitable for cgroup v1. So guard this test (memcg->under_oom)
by the cgroup_subsys_on_dfl(memory_cgrp_subsys).

Signed-off-by: Muchun Song <songmuchun@bytedance.com>
---
 mm/memcontrol.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 7afca9677693..a3f26522765a 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -3126,8 +3126,10 @@ static int __memcg_kmem_charge(struct mem_cgroup *memcg, gfp_t gfp,
  */
 static void __memcg_kmem_uncharge(struct mem_cgroup *memcg, unsigned int nr_pages)
 {
-	if (!cgroup_subsys_on_dfl(memory_cgrp_subsys))
+	if (!cgroup_subsys_on_dfl(memory_cgrp_subsys)) {
 		page_counter_uncharge(&memcg->kmem, nr_pages);
+		memcg_oom_recover(memcg);
+	}
 
 	refill_stock(memcg, nr_pages);
 }
@@ -6806,11 +6808,15 @@ static void uncharge_batch(const struct uncharge_gather *ug)
 
 	if (!mem_cgroup_is_root(ug->memcg)) {
 		page_counter_uncharge(&ug->memcg->memory, ug->nr_pages);
-		if (do_memsw_account())
-			page_counter_uncharge(&ug->memcg->memsw, ug->nr_pages);
-		if (!cgroup_subsys_on_dfl(memory_cgrp_subsys) && ug->nr_kmem)
-			page_counter_uncharge(&ug->memcg->kmem, ug->nr_kmem);
-		memcg_oom_recover(ug->memcg);
+		if (!cgroup_subsys_on_dfl(memory_cgrp_subsys)) {
+			if (!cgroup_memory_noswap)
+				page_counter_uncharge(&ug->memcg->memsw,
+						      ug->nr_pages);
+			if (ug->nr_kmem)
+				page_counter_uncharge(&ug->memcg->kmem,
+						      ug->nr_kmem);
+			memcg_oom_recover(ug->memcg);
+		}
 	}
 
 	local_irq_save(flags);
-- 
2.11.0