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:   zangchunxin@bytedance.com
To:     hannes@cmpxchg.org, mhocko@kernel.org, vdavydov.dev@gmail.com,
        akpm@linux-foundation.org, tj@kernel.org, lizefan@huawei.com,
        corbet@lwn.net
Cc:     ast@kernel.org, daniel@iogearbox.net, kafai@fb.com,
        songliubraving@fb.com, yhs@fb.com, andriin@fb.com,
        john.fastabend@gmail.com, kpsingh@chromium.org,
        cgroups@vger.kernel.org, linux-doc@vger.kernel.org,
        linux-mm@kvack.org, linux-kernel@vger.kernel.org,
        netdev@vger.kernel.org, bpf@vger.kernel.org,
        Chunxin Zang <zangchunxin@bytedance.com>
Subject: [PATCH] mm/memcontrol: Add the drop_cache interface for cgroup v2
Date:   Mon, 21 Sep 2020 16:02:55 +0800
Message-Id: <20200921080255.15505-1-zangchunxin@bytedance.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

From: Chunxin Zang <zangchunxin@bytedance.com>

In the cgroup v1, we have 'force_mepty' interface. This is very
useful for userspace to actively release memory. But the cgroup
v2 does not.

This patch reuse cgroup v1's function, but have a new name for
the interface. Because I think 'drop_cache' may be is easier to
understand :)

Signed-off-by: Chunxin Zang <zangchunxin@bytedance.com>
---
 Documentation/admin-guide/cgroup-v2.rst | 11 +++++++++++
 mm/memcontrol.c                         |  5 +++++
 2 files changed, 16 insertions(+)

diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst
index ce3e05e41724..fbff959c8116 100644
--- a/Documentation/admin-guide/cgroup-v2.rst
+++ b/Documentation/admin-guide/cgroup-v2.rst
@@ -1181,6 +1181,17 @@ PAGE_SIZE multiple when read back.
 	high limit is used and monitored properly, this limit's
 	utility is limited to providing the final safety net.
 
+  memory.drop_cache
+    A write-only single value file which exists on non-root
+    cgroups.
+
+    Provide a mechanism for users to actively trigger memory
+    reclaim. The cgroup will be reclaimed and as many pages
+    reclaimed as possible.
+
+    It will broke low boundary. Because it tries to reclaim the
+    memory many times, until the memory drops to a certain level.
+
   memory.oom.group
 	A read-write single value file which exists on non-root
 	cgroups.  The default value is "0".
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 0b38b6ad547d..98646484efff 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -6226,6 +6226,11 @@ static struct cftype memory_files[] = {
 		.write = memory_max_write,
 	},
 	{
+		.name = "drop_cache",
+		.flags = CFTYPE_NOT_ON_ROOT,
+		.write = mem_cgroup_force_empty_write,
+	},
+	{
 		.name = "events",
 		.flags = CFTYPE_NOT_ON_ROOT,
 		.file_offset = offsetof(struct mem_cgroup, events_file),
-- 
2.11.0