Adding upstream version 5.10.209.upstream/5.10.209

Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>

10 files changed, 4233 insertions, 0 deletions

diff --git a/tools/testing/selftests/cgroup/.gitignore b/tools/testing/selftests/cgroup/.gitignore
new file mode 100644
index 000000000..84cfcabea
--- /dev/null
+++ b/tools/testing/selftests/cgroup/.gitignore
@@ -0,0 +1,5 @@
+# SPDX-License-Identifier: GPL-2.0-only
+test_memcontrol
+test_core
+test_freezer
+test_kmem
\ No newline at end of file
diff --git a/tools/testing/selftests/cgroup/Makefile b/tools/testing/selftests/cgroup/Makefile
new file mode 100644
index 000000000..f027d9335
--- /dev/null
+++ b/tools/testing/selftests/cgroup/Makefile
@@ -0,0 +1,18 @@
+# SPDX-License-Identifier: GPL-2.0
+CFLAGS += -Wall -pthread
+
+all:
+
+TEST_FILES     := with_stress.sh
+TEST_PROGS     := test_stress.sh
+TEST_GEN_PROGS = test_memcontrol
+TEST_GEN_PROGS += test_kmem
+TEST_GEN_PROGS += test_core
+TEST_GEN_PROGS += test_freezer
+
+include ../lib.mk
+
+$(OUTPUT)/test_memcontrol: cgroup_util.c ../clone3/clone3_selftests.h
+$(OUTPUT)/test_kmem: cgroup_util.c ../clone3/clone3_selftests.h
+$(OUTPUT)/test_core: cgroup_util.c ../clone3/clone3_selftests.h
+$(OUTPUT)/test_freezer: cgroup_util.c ../clone3/clone3_selftests.h
diff --git a/tools/testing/selftests/cgroup/cgroup_util.c b/tools/testing/selftests/cgroup/cgroup_util.c
new file mode 100644
index 000000000..5b16c7b0a
--- /dev/null
+++ b/tools/testing/selftests/cgroup/cgroup_util.c
@@ -0,0 +1,578 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#define _GNU_SOURCE
+
+#include <errno.h>
+#include <fcntl.h>
+#include <linux/limits.h>
+#include <signal.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include <unistd.h>
+
+#include "cgroup_util.h"
+#include "../clone3/clone3_selftests.h"
+
+static ssize_t read_text(const char *path, char *buf, size_t max_len)
+{
+	ssize_t len;
+	int fd;
+
+	fd = open(path, O_RDONLY);
+	if (fd < 0)
+		return fd;
+
+	len = read(fd, buf, max_len - 1);
+	if (len < 0)
+		goto out;
+
+	buf[len] = 0;
+out:
+	close(fd);
+	return len;
+}
+
+static ssize_t write_text(const char *path, char *buf, ssize_t len)
+{
+	int fd;
+
+	fd = open(path, O_WRONLY | O_APPEND);
+	if (fd < 0)
+		return fd;
+
+	len = write(fd, buf, len);
+	if (len < 0) {
+		close(fd);
+		return len;
+	}
+
+	close(fd);
+
+	return len;
+}
+
+char *cg_name(const char *root, const char *name)
+{
+	size_t len = strlen(root) + strlen(name) + 2;
+	char *ret = malloc(len);
+
+	snprintf(ret, len, "%s/%s", root, name);
+
+	return ret;
+}
+
+char *cg_name_indexed(const char *root, const char *name, int index)
+{
+	size_t len = strlen(root) + strlen(name) + 10;
+	char *ret = malloc(len);
+
+	snprintf(ret, len, "%s/%s_%d", root, name, index);
+
+	return ret;
+}
+
+char *cg_control(const char *cgroup, const char *control)
+{
+	size_t len = strlen(cgroup) + strlen(control) + 2;
+	char *ret = malloc(len);
+
+	snprintf(ret, len, "%s/%s", cgroup, control);
+
+	return ret;
+}
+
+int cg_read(const char *cgroup, const char *control, char *buf, size_t len)
+{
+	char path[PATH_MAX];
+
+	snprintf(path, sizeof(path), "%s/%s", cgroup, control);
+
+	if (read_text(path, buf, len) >= 0)
+		return 0;
+
+	return -1;
+}
+
+int cg_read_strcmp(const char *cgroup, const char *control,
+		   const char *expected)
+{
+	size_t size;
+	char *buf;
+	int ret;
+
+	/* Handle the case of comparing against empty string */
+	if (!expected)
+		return -1;
+	else
+		size = strlen(expected) + 1;
+
+	buf = malloc(size);
+	if (!buf)
+		return -1;
+
+	if (cg_read(cgroup, control, buf, size)) {
+		free(buf);
+		return -1;
+	}
+
+	ret = strcmp(expected, buf);
+	free(buf);
+	return ret;
+}
+
+int cg_read_strstr(const char *cgroup, const char *control, const char *needle)
+{
+	char buf[PAGE_SIZE];
+
+	if (cg_read(cgroup, control, buf, sizeof(buf)))
+		return -1;
+
+	return strstr(buf, needle) ? 0 : -1;
+}
+
+long cg_read_long(const char *cgroup, const char *control)
+{
+	char buf[128];
+
+	if (cg_read(cgroup, control, buf, sizeof(buf)))
+		return -1;
+
+	return atol(buf);
+}
+
+long cg_read_key_long(const char *cgroup, const char *control, const char *key)
+{
+	char buf[PAGE_SIZE];
+	char *ptr;
+
+	if (cg_read(cgroup, control, buf, sizeof(buf)))
+		return -1;
+
+	ptr = strstr(buf, key);
+	if (!ptr)
+		return -1;
+
+	return atol(ptr + strlen(key));
+}
+
+long cg_read_lc(const char *cgroup, const char *control)
+{
+	char buf[PAGE_SIZE];
+	const char delim[] = "\n";
+	char *line;
+	long cnt = 0;
+
+	if (cg_read(cgroup, control, buf, sizeof(buf)))
+		return -1;
+
+	for (line = strtok(buf, delim); line; line = strtok(NULL, delim))
+		cnt++;
+
+	return cnt;
+}
+
+int cg_write(const char *cgroup, const char *control, char *buf)
+{
+	char path[PATH_MAX];
+	ssize_t len = strlen(buf);
+
+	snprintf(path, sizeof(path), "%s/%s", cgroup, control);
+
+	if (write_text(path, buf, len) == len)
+		return 0;
+
+	return -1;
+}
+
+int cg_find_unified_root(char *root, size_t len)
+{
+	char buf[10 * PAGE_SIZE];
+	char *fs, *mount, *type;
+	const char delim[] = "\n\t ";
+
+	if (read_text("/proc/self/mounts", buf, sizeof(buf)) <= 0)
+		return -1;
+
+	/*
+	 * Example:
+	 * cgroup /sys/fs/cgroup cgroup2 rw,seclabel,noexec,relatime 0 0
+	 */
+	for (fs = strtok(buf, delim); fs; fs = strtok(NULL, delim)) {
+		mount = strtok(NULL, delim);
+		type = strtok(NULL, delim);
+		strtok(NULL, delim);
+		strtok(NULL, delim);
+		strtok(NULL, delim);
+
+		if (strcmp(type, "cgroup2") == 0) {
+			strncpy(root, mount, len);
+			return 0;
+		}
+	}
+
+	return -1;
+}
+
+int cg_create(const char *cgroup)
+{
+	return mkdir(cgroup, 0755);
+}
+
+int cg_wait_for_proc_count(const char *cgroup, int count)
+{
+	char buf[10 * PAGE_SIZE] = {0};
+	int attempts;
+	char *ptr;
+
+	for (attempts = 10; attempts >= 0; attempts--) {
+		int nr = 0;
+
+		if (cg_read(cgroup, "cgroup.procs", buf, sizeof(buf)))
+			break;
+
+		for (ptr = buf; *ptr; ptr++)
+			if (*ptr == '\n')
+				nr++;
+
+		if (nr >= count)
+			return 0;
+
+		usleep(100000);
+	}
+
+	return -1;
+}
+
+int cg_killall(const char *cgroup)
+{
+	char buf[PAGE_SIZE];
+	char *ptr = buf;
+
+	if (cg_read(cgroup, "cgroup.procs", buf, sizeof(buf)))
+		return -1;
+
+	while (ptr < buf + sizeof(buf)) {
+		int pid = strtol(ptr, &ptr, 10);
+
+		if (pid == 0)
+			break;
+		if (*ptr)
+			ptr++;
+		else
+			break;
+		if (kill(pid, SIGKILL))
+			return -1;
+	}
+
+	return 0;
+}
+
+int cg_destroy(const char *cgroup)
+{
+	int ret;
+
+retry:
+	ret = rmdir(cgroup);
+	if (ret && errno == EBUSY) {
+		cg_killall(cgroup);
+		usleep(100);
+		goto retry;
+	}
+
+	if (ret && errno == ENOENT)
+		ret = 0;
+
+	return ret;
+}
+
+int cg_enter(const char *cgroup, int pid)
+{
+	char pidbuf[64];
+
+	snprintf(pidbuf, sizeof(pidbuf), "%d", pid);
+	return cg_write(cgroup, "cgroup.procs", pidbuf);
+}
+
+int cg_enter_current(const char *cgroup)
+{
+	return cg_write(cgroup, "cgroup.procs", "0");
+}
+
+int cg_enter_current_thread(const char *cgroup)
+{
+	return cg_write(cgroup, "cgroup.threads", "0");
+}
+
+int cg_run(const char *cgroup,
+	   int (*fn)(const char *cgroup, void *arg),
+	   void *arg)
+{
+	int pid, retcode;
+
+	pid = fork();
+	if (pid < 0) {
+		return pid;
+	} else if (pid == 0) {
+		char buf[64];
+
+		snprintf(buf, sizeof(buf), "%d", getpid());
+		if (cg_write(cgroup, "cgroup.procs", buf))
+			exit(EXIT_FAILURE);
+		exit(fn(cgroup, arg));
+	} else {
+		waitpid(pid, &retcode, 0);
+		if (WIFEXITED(retcode))
+			return WEXITSTATUS(retcode);
+		else
+			return -1;
+	}
+}
+
+pid_t clone_into_cgroup(int cgroup_fd)
+{
+#ifdef CLONE_ARGS_SIZE_VER2
+	pid_t pid;
+
+	struct __clone_args args = {
+		.flags = CLONE_INTO_CGROUP,
+		.exit_signal = SIGCHLD,
+		.cgroup = cgroup_fd,
+	};
+
+	pid = sys_clone3(&args, sizeof(struct __clone_args));
+	/*
+	 * Verify that this is a genuine test failure:
+	 * ENOSYS -> clone3() not available
+	 * E2BIG  -> CLONE_INTO_CGROUP not available
+	 */
+	if (pid < 0 && (errno == ENOSYS || errno == E2BIG))
+		goto pretend_enosys;
+
+	return pid;
+
+pretend_enosys:
+#endif
+	errno = ENOSYS;
+	return -ENOSYS;
+}
+
+int clone_reap(pid_t pid, int options)
+{
+	int ret;
+	siginfo_t info = {
+		.si_signo = 0,
+	};
+
+again:
+	ret = waitid(P_PID, pid, &info, options | __WALL | __WNOTHREAD);
+	if (ret < 0) {
+		if (errno == EINTR)
+			goto again;
+		return -1;
+	}
+
+	if (options & WEXITED) {
+		if (WIFEXITED(info.si_status))
+			return WEXITSTATUS(info.si_status);
+	}
+
+	if (options & WSTOPPED) {
+		if (WIFSTOPPED(info.si_status))
+			return WSTOPSIG(info.si_status);
+	}
+
+	if (options & WCONTINUED) {
+		if (WIFCONTINUED(info.si_status))
+			return 0;
+	}
+
+	return -1;
+}
+
+int dirfd_open_opath(const char *dir)
+{
+	return open(dir, O_DIRECTORY | O_CLOEXEC | O_NOFOLLOW | O_PATH);
+}
+
+#define close_prot_errno(fd)                                                   \
+	if (fd >= 0) {                                                         \
+		int _e_ = errno;                                               \
+		close(fd);                                                     \
+		errno = _e_;                                                   \
+	}
+
+static int clone_into_cgroup_run_nowait(const char *cgroup,
+					int (*fn)(const char *cgroup, void *arg),
+					void *arg)
+{
+	int cgroup_fd;
+	pid_t pid;
+
+	cgroup_fd =  dirfd_open_opath(cgroup);
+	if (cgroup_fd < 0)
+		return -1;
+
+	pid = clone_into_cgroup(cgroup_fd);
+	close_prot_errno(cgroup_fd);
+	if (pid == 0)
+		exit(fn(cgroup, arg));
+
+	return pid;
+}
+
+int cg_run_nowait(const char *cgroup,
+		  int (*fn)(const char *cgroup, void *arg),
+		  void *arg)
+{
+	int pid;
+
+	pid = clone_into_cgroup_run_nowait(cgroup, fn, arg);
+	if (pid > 0)
+		return pid;
+
+	/* Genuine test failure. */
+	if (pid < 0 && errno != ENOSYS)
+		return -1;
+
+	pid = fork();
+	if (pid == 0) {
+		char buf[64];
+
+		snprintf(buf, sizeof(buf), "%d", getpid());
+		if (cg_write(cgroup, "cgroup.procs", buf))
+			exit(EXIT_FAILURE);
+		exit(fn(cgroup, arg));
+	}
+
+	return pid;
+}
+
+int get_temp_fd(void)
+{
+	return open(".", O_TMPFILE | O_RDWR | O_EXCL);
+}
+
+int alloc_pagecache(int fd, size_t size)
+{
+	char buf[PAGE_SIZE];
+	struct stat st;
+	int i;
+
+	if (fstat(fd, &st))
+		goto cleanup;
+
+	size += st.st_size;
+
+	if (ftruncate(fd, size))
+		goto cleanup;
+
+	for (i = 0; i < size; i += sizeof(buf))
+		read(fd, buf, sizeof(buf));
+
+	return 0;
+
+cleanup:
+	return -1;
+}
+
+int alloc_anon(const char *cgroup, void *arg)
+{
+	size_t size = (unsigned long)arg;
+	char *buf, *ptr;
+
+	buf = malloc(size);
+	for (ptr = buf; ptr < buf + size; ptr += PAGE_SIZE)
+		*ptr = 0;
+
+	free(buf);
+	return 0;
+}
+
+int is_swap_enabled(void)
+{
+	char buf[PAGE_SIZE];
+	const char delim[] = "\n";
+	int cnt = 0;
+	char *line;
+
+	if (read_text("/proc/swaps", buf, sizeof(buf)) <= 0)
+		return -1;
+
+	for (line = strtok(buf, delim); line; line = strtok(NULL, delim))
+		cnt++;
+
+	return cnt > 1;
+}
+
+int set_oom_adj_score(int pid, int score)
+{
+	char path[PATH_MAX];
+	int fd, len;
+
+	sprintf(path, "/proc/%d/oom_score_adj", pid);
+
+	fd = open(path, O_WRONLY | O_APPEND);
+	if (fd < 0)
+		return fd;
+
+	len = dprintf(fd, "%d", score);
+	if (len < 0) {
+		close(fd);
+		return len;
+	}
+
+	close(fd);
+	return 0;
+}
+
+ssize_t proc_read_text(int pid, bool thread, const char *item, char *buf, size_t size)
+{
+	char path[PATH_MAX];
+
+	if (!pid)
+		snprintf(path, sizeof(path), "/proc/%s/%s",
+			 thread ? "thread-self" : "self", item);
+	else
+		snprintf(path, sizeof(path), "/proc/%d/%s", pid, item);
+
+	return read_text(path, buf, size);
+}
+
+int proc_read_strstr(int pid, bool thread, const char *item, const char *needle)
+{
+	char buf[PAGE_SIZE];
+
+	if (proc_read_text(pid, thread, item, buf, sizeof(buf)) < 0)
+		return -1;
+
+	return strstr(buf, needle) ? 0 : -1;
+}
+
+int clone_into_cgroup_run_wait(const char *cgroup)
+{
+	int cgroup_fd;
+	pid_t pid;
+
+	cgroup_fd =  dirfd_open_opath(cgroup);
+	if (cgroup_fd < 0)
+		return -1;
+
+	pid = clone_into_cgroup(cgroup_fd);
+	close_prot_errno(cgroup_fd);
+	if (pid < 0)
+		return -1;
+
+	if (pid == 0)
+		exit(EXIT_SUCCESS);
+
+	/*
+	 * We don't care whether this fails. We only care whether the initial
+	 * clone succeeded.
+	 */
+	(void)clone_reap(pid, WEXITED);
+	return 0;
+}
diff --git a/tools/testing/selftests/cgroup/cgroup_util.h b/tools/testing/selftests/cgroup/cgroup_util.h
new file mode 100644
index 000000000..5a1305dd1
--- /dev/null
+++ b/tools/testing/selftests/cgroup/cgroup_util.h
@@ -0,0 +1,56 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#include <stdbool.h>
+#include <stdlib.h>
+
+#define PAGE_SIZE 4096
+
+#define ARRAY_SIZE(arr) (sizeof(arr) / sizeof((arr)[0]))
+
+#define MB(x) (x << 20)
+
+/*
+ * Checks if two given values differ by less than err% of their sum.
+ */
+static inline int values_close(long a, long b, int err)
+{
+	return abs(a - b) <= (a + b) / 100 * err;
+}
+
+extern int cg_find_unified_root(char *root, size_t len);
+extern char *cg_name(const char *root, const char *name);
+extern char *cg_name_indexed(const char *root, const char *name, int index);
+extern char *cg_control(const char *cgroup, const char *control);
+extern int cg_create(const char *cgroup);
+extern int cg_destroy(const char *cgroup);
+extern int cg_read(const char *cgroup, const char *control,
+		   char *buf, size_t len);
+extern int cg_read_strcmp(const char *cgroup, const char *control,
+			  const char *expected);
+extern int cg_read_strstr(const char *cgroup, const char *control,
+			  const char *needle);
+extern long cg_read_long(const char *cgroup, const char *control);
+long cg_read_key_long(const char *cgroup, const char *control, const char *key);
+extern long cg_read_lc(const char *cgroup, const char *control);
+extern int cg_write(const char *cgroup, const char *control, char *buf);
+extern int cg_run(const char *cgroup,
+		  int (*fn)(const char *cgroup, void *arg),
+		  void *arg);
+extern int cg_enter(const char *cgroup, int pid);
+extern int cg_enter_current(const char *cgroup);
+extern int cg_enter_current_thread(const char *cgroup);
+extern int cg_run_nowait(const char *cgroup,
+			 int (*fn)(const char *cgroup, void *arg),
+			 void *arg);
+extern int get_temp_fd(void);
+extern int alloc_pagecache(int fd, size_t size);
+extern int alloc_anon(const char *cgroup, void *arg);
+extern int is_swap_enabled(void);
+extern int set_oom_adj_score(int pid, int score);
+extern int cg_wait_for_proc_count(const char *cgroup, int count);
+extern int cg_killall(const char *cgroup);
+extern ssize_t proc_read_text(int pid, bool thread, const char *item, char *buf, size_t size);
+extern int proc_read_strstr(int pid, bool thread, const char *item, const char *needle);
+extern pid_t clone_into_cgroup(int cgroup_fd);
+extern int clone_reap(pid_t pid, int options);
+extern int clone_into_cgroup_run_wait(const char *cgroup);
+extern int dirfd_open_opath(const char *dir);
diff --git a/tools/testing/selftests/cgroup/test_core.c b/tools/testing/selftests/cgroup/test_core.c
new file mode 100644
index 000000000..600123503
--- /dev/null
+++ b/tools/testing/selftests/cgroup/test_core.c
@@ -0,0 +1,888 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+
+#define _GNU_SOURCE
+#include <linux/limits.h>
+#include <linux/sched.h>
+#include <sys/types.h>
+#include <sys/mman.h>
+#include <sys/wait.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <sched.h>
+#include <stdio.h>
+#include <errno.h>
+#include <signal.h>
+#include <string.h>
+#include <pthread.h>
+
+#include "../kselftest.h"
+#include "cgroup_util.h"
+
+static int touch_anon(char *buf, size_t size)
+{
+	int fd;
+	char *pos = buf;
+
+	fd = open("/dev/urandom", O_RDONLY);
+	if (fd < 0)
+		return -1;
+
+	while (size > 0) {
+		ssize_t ret = read(fd, pos, size);
+
+		if (ret < 0) {
+			if (errno != EINTR) {
+				close(fd);
+				return -1;
+			}
+		} else {
+			pos += ret;
+			size -= ret;
+		}
+	}
+	close(fd);
+
+	return 0;
+}
+
+static int alloc_and_touch_anon_noexit(const char *cgroup, void *arg)
+{
+	int ppid = getppid();
+	size_t size = (size_t)arg;
+	void *buf;
+
+	buf = mmap(NULL, size, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANON,
+		   0, 0);
+	if (buf == MAP_FAILED)
+		return -1;
+
+	if (touch_anon((char *)buf, size)) {
+		munmap(buf, size);
+		return -1;
+	}
+
+	while (getppid() == ppid)
+		sleep(1);
+
+	munmap(buf, size);
+	return 0;
+}
+
+/*
+ * Create a child process that allocates and touches 100MB, then waits to be
+ * killed. Wait until the child is attached to the cgroup, kill all processes
+ * in that cgroup and wait until "cgroup.procs" is empty. At this point try to
+ * destroy the empty cgroup. The test helps detect race conditions between
+ * dying processes leaving the cgroup and cgroup destruction path.
+ */
+static int test_cgcore_destroy(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *cg_test = NULL;
+	int child_pid;
+	char buf[PAGE_SIZE];
+
+	cg_test = cg_name(root, "cg_test");
+
+	if (!cg_test)
+		goto cleanup;
+
+	for (int i = 0; i < 10; i++) {
+		if (cg_create(cg_test))
+			goto cleanup;
+
+		child_pid = cg_run_nowait(cg_test, alloc_and_touch_anon_noexit,
+					  (void *) MB(100));
+
+		if (child_pid < 0)
+			goto cleanup;
+
+		/* wait for the child to enter cgroup */
+		if (cg_wait_for_proc_count(cg_test, 1))
+			goto cleanup;
+
+		if (cg_killall(cg_test))
+			goto cleanup;
+
+		/* wait for cgroup to be empty */
+		while (1) {
+			if (cg_read(cg_test, "cgroup.procs", buf, sizeof(buf)))
+				goto cleanup;
+			if (buf[0] == '\0')
+				break;
+			usleep(1000);
+		}
+
+		if (rmdir(cg_test))
+			goto cleanup;
+
+		if (waitpid(child_pid, NULL, 0) < 0)
+			goto cleanup;
+	}
+	ret = KSFT_PASS;
+cleanup:
+	if (cg_test)
+		cg_destroy(cg_test);
+	free(cg_test);
+	return ret;
+}
+
+/*
+ * A(0) - B(0) - C(1)
+ *        \ D(0)
+ *
+ * A, B and C's "populated" fields would be 1 while D's 0.
+ * test that after the one process in C is moved to root,
+ * A,B and C's "populated" fields would flip to "0" and file
+ * modified events will be generated on the
+ * "cgroup.events" files of both cgroups.
+ */
+static int test_cgcore_populated(const char *root)
+{
+	int ret = KSFT_FAIL;
+	int err;
+	char *cg_test_a = NULL, *cg_test_b = NULL;
+	char *cg_test_c = NULL, *cg_test_d = NULL;
+	int cgroup_fd = -EBADF;
+	pid_t pid;
+
+	cg_test_a = cg_name(root, "cg_test_a");
+	cg_test_b = cg_name(root, "cg_test_a/cg_test_b");
+	cg_test_c = cg_name(root, "cg_test_a/cg_test_b/cg_test_c");
+	cg_test_d = cg_name(root, "cg_test_a/cg_test_b/cg_test_d");
+
+	if (!cg_test_a || !cg_test_b || !cg_test_c || !cg_test_d)
+		goto cleanup;
+
+	if (cg_create(cg_test_a))
+		goto cleanup;
+
+	if (cg_create(cg_test_b))
+		goto cleanup;
+
+	if (cg_create(cg_test_c))
+		goto cleanup;
+
+	if (cg_create(cg_test_d))
+		goto cleanup;
+
+	if (cg_enter_current(cg_test_c))
+		goto cleanup;
+
+	if (cg_read_strcmp(cg_test_a, "cgroup.events", "populated 1\n"))
+		goto cleanup;
+
+	if (cg_read_strcmp(cg_test_b, "cgroup.events", "populated 1\n"))
+		goto cleanup;
+
+	if (cg_read_strcmp(cg_test_c, "cgroup.events", "populated 1\n"))
+		goto cleanup;
+
+	if (cg_read_strcmp(cg_test_d, "cgroup.events", "populated 0\n"))
+		goto cleanup;
+
+	if (cg_enter_current(root))
+		goto cleanup;
+
+	if (cg_read_strcmp(cg_test_a, "cgroup.events", "populated 0\n"))
+		goto cleanup;
+
+	if (cg_read_strcmp(cg_test_b, "cgroup.events", "populated 0\n"))
+		goto cleanup;
+
+	if (cg_read_strcmp(cg_test_c, "cgroup.events", "populated 0\n"))
+		goto cleanup;
+
+	if (cg_read_strcmp(cg_test_d, "cgroup.events", "populated 0\n"))
+		goto cleanup;
+
+	/* Test that we can directly clone into a new cgroup. */
+	cgroup_fd = dirfd_open_opath(cg_test_d);
+	if (cgroup_fd < 0)
+		goto cleanup;
+
+	pid = clone_into_cgroup(cgroup_fd);
+	if (pid < 0) {
+		if (errno == ENOSYS)
+			goto cleanup_pass;
+		goto cleanup;
+	}
+
+	if (pid == 0) {
+		if (raise(SIGSTOP))
+			exit(EXIT_FAILURE);
+		exit(EXIT_SUCCESS);
+	}
+
+	err = cg_read_strcmp(cg_test_d, "cgroup.events", "populated 1\n");
+
+	(void)clone_reap(pid, WSTOPPED);
+	(void)kill(pid, SIGCONT);
+	(void)clone_reap(pid, WEXITED);
+
+	if (err)
+		goto cleanup;
+
+	if (cg_read_strcmp(cg_test_d, "cgroup.events", "populated 0\n"))
+		goto cleanup;
+
+	/* Remove cgroup. */
+	if (cg_test_d) {
+		cg_destroy(cg_test_d);
+		free(cg_test_d);
+		cg_test_d = NULL;
+	}
+
+	pid = clone_into_cgroup(cgroup_fd);
+	if (pid < 0)
+		goto cleanup_pass;
+	if (pid == 0)
+		exit(EXIT_SUCCESS);
+	(void)clone_reap(pid, WEXITED);
+	goto cleanup;
+
+cleanup_pass:
+	ret = KSFT_PASS;
+
+cleanup:
+	if (cg_test_d)
+		cg_destroy(cg_test_d);
+	if (cg_test_c)
+		cg_destroy(cg_test_c);
+	if (cg_test_b)
+		cg_destroy(cg_test_b);
+	if (cg_test_a)
+		cg_destroy(cg_test_a);
+	free(cg_test_d);
+	free(cg_test_c);
+	free(cg_test_b);
+	free(cg_test_a);
+	if (cgroup_fd >= 0)
+		close(cgroup_fd);
+	return ret;
+}
+
+/*
+ * A (domain threaded) - B (threaded) - C (domain)
+ *
+ * test that C can't be used until it is turned into a
+ * threaded cgroup.  "cgroup.type" file will report "domain (invalid)" in
+ * these cases. Operations which fail due to invalid topology use
+ * EOPNOTSUPP as the errno.
+ */
+static int test_cgcore_invalid_domain(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *grandparent = NULL, *parent = NULL, *child = NULL;
+
+	grandparent = cg_name(root, "cg_test_grandparent");
+	parent = cg_name(root, "cg_test_grandparent/cg_test_parent");
+	child = cg_name(root, "cg_test_grandparent/cg_test_parent/cg_test_child");
+	if (!parent || !child || !grandparent)
+		goto cleanup;
+
+	if (cg_create(grandparent))
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_create(child))
+		goto cleanup;
+
+	if (cg_write(parent, "cgroup.type", "threaded"))
+		goto cleanup;
+
+	if (cg_read_strcmp(child, "cgroup.type", "domain invalid\n"))
+		goto cleanup;
+
+	if (!cg_enter_current(child))
+		goto cleanup;
+
+	if (errno != EOPNOTSUPP)
+		goto cleanup;
+
+	if (!clone_into_cgroup_run_wait(child))
+		goto cleanup;
+
+	if (errno == ENOSYS)
+		goto cleanup_pass;
+
+	if (errno != EOPNOTSUPP)
+		goto cleanup;
+
+cleanup_pass:
+	ret = KSFT_PASS;
+
+cleanup:
+	cg_enter_current(root);
+	if (child)
+		cg_destroy(child);
+	if (parent)
+		cg_destroy(parent);
+	if (grandparent)
+		cg_destroy(grandparent);
+	free(child);
+	free(parent);
+	free(grandparent);
+	return ret;
+}
+
+/*
+ * Test that when a child becomes threaded
+ * the parent type becomes domain threaded.
+ */
+static int test_cgcore_parent_becomes_threaded(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent = NULL, *child = NULL;
+
+	parent = cg_name(root, "cg_test_parent");
+	child = cg_name(root, "cg_test_parent/cg_test_child");
+	if (!parent || !child)
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_create(child))
+		goto cleanup;
+
+	if (cg_write(child, "cgroup.type", "threaded"))
+		goto cleanup;
+
+	if (cg_read_strcmp(parent, "cgroup.type", "domain threaded\n"))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (child)
+		cg_destroy(child);
+	if (parent)
+		cg_destroy(parent);
+	free(child);
+	free(parent);
+	return ret;
+
+}
+
+/*
+ * Test that there's no internal process constrain on threaded cgroups.
+ * You can add threads/processes on a parent with a controller enabled.
+ */
+static int test_cgcore_no_internal_process_constraint_on_threads(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent = NULL, *child = NULL;
+
+	if (cg_read_strstr(root, "cgroup.controllers", "cpu") ||
+	    cg_write(root, "cgroup.subtree_control", "+cpu")) {
+		ret = KSFT_SKIP;
+		goto cleanup;
+	}
+
+	parent = cg_name(root, "cg_test_parent");
+	child = cg_name(root, "cg_test_parent/cg_test_child");
+	if (!parent || !child)
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_create(child))
+		goto cleanup;
+
+	if (cg_write(parent, "cgroup.type", "threaded"))
+		goto cleanup;
+
+	if (cg_write(child, "cgroup.type", "threaded"))
+		goto cleanup;
+
+	if (cg_write(parent, "cgroup.subtree_control", "+cpu"))
+		goto cleanup;
+
+	if (cg_enter_current(parent))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	cg_enter_current(root);
+	cg_enter_current(root);
+	if (child)
+		cg_destroy(child);
+	if (parent)
+		cg_destroy(parent);
+	free(child);
+	free(parent);
+	return ret;
+}
+
+/*
+ * Test that you can't enable a controller on a child if it's not enabled
+ * on the parent.
+ */
+static int test_cgcore_top_down_constraint_enable(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent = NULL, *child = NULL;
+
+	parent = cg_name(root, "cg_test_parent");
+	child = cg_name(root, "cg_test_parent/cg_test_child");
+	if (!parent || !child)
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_create(child))
+		goto cleanup;
+
+	if (!cg_write(child, "cgroup.subtree_control", "+memory"))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (child)
+		cg_destroy(child);
+	if (parent)
+		cg_destroy(parent);
+	free(child);
+	free(parent);
+	return ret;
+}
+
+/*
+ * Test that you can't disable a controller on a parent
+ * if it's enabled in a child.
+ */
+static int test_cgcore_top_down_constraint_disable(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent = NULL, *child = NULL;
+
+	parent = cg_name(root, "cg_test_parent");
+	child = cg_name(root, "cg_test_parent/cg_test_child");
+	if (!parent || !child)
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_create(child))
+		goto cleanup;
+
+	if (cg_write(parent, "cgroup.subtree_control", "+memory"))
+		goto cleanup;
+
+	if (cg_write(child, "cgroup.subtree_control", "+memory"))
+		goto cleanup;
+
+	if (!cg_write(parent, "cgroup.subtree_control", "-memory"))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (child)
+		cg_destroy(child);
+	if (parent)
+		cg_destroy(parent);
+	free(child);
+	free(parent);
+	return ret;
+}
+
+/*
+ * Test internal process constraint.
+ * You can't add a pid to a domain parent if a controller is enabled.
+ */
+static int test_cgcore_internal_process_constraint(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent = NULL, *child = NULL;
+
+	parent = cg_name(root, "cg_test_parent");
+	child = cg_name(root, "cg_test_parent/cg_test_child");
+	if (!parent || !child)
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_create(child))
+		goto cleanup;
+
+	if (cg_write(parent, "cgroup.subtree_control", "+memory"))
+		goto cleanup;
+
+	if (!cg_enter_current(parent))
+		goto cleanup;
+
+	if (!clone_into_cgroup_run_wait(parent))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (child)
+		cg_destroy(child);
+	if (parent)
+		cg_destroy(parent);
+	free(child);
+	free(parent);
+	return ret;
+}
+
+static void *dummy_thread_fn(void *arg)
+{
+	return (void *)(size_t)pause();
+}
+
+/*
+ * Test threadgroup migration.
+ * All threads of a process are migrated together.
+ */
+static int test_cgcore_proc_migration(const char *root)
+{
+	int ret = KSFT_FAIL;
+	int t, c_threads = 0, n_threads = 13;
+	char *src = NULL, *dst = NULL;
+	pthread_t threads[n_threads];
+
+	src = cg_name(root, "cg_src");
+	dst = cg_name(root, "cg_dst");
+	if (!src || !dst)
+		goto cleanup;
+
+	if (cg_create(src))
+		goto cleanup;
+	if (cg_create(dst))
+		goto cleanup;
+
+	if (cg_enter_current(src))
+		goto cleanup;
+
+	for (c_threads = 0; c_threads < n_threads; ++c_threads) {
+		if (pthread_create(&threads[c_threads], NULL, dummy_thread_fn, NULL))
+			goto cleanup;
+	}
+
+	cg_enter_current(dst);
+	if (cg_read_lc(dst, "cgroup.threads") != n_threads + 1)
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	for (t = 0; t < c_threads; ++t) {
+		pthread_cancel(threads[t]);
+	}
+
+	for (t = 0; t < c_threads; ++t) {
+		pthread_join(threads[t], NULL);
+	}
+
+	cg_enter_current(root);
+
+	if (dst)
+		cg_destroy(dst);
+	if (src)
+		cg_destroy(src);
+	free(dst);
+	free(src);
+	return ret;
+}
+
+static void *migrating_thread_fn(void *arg)
+{
+	int g, i, n_iterations = 1000;
+	char **grps = arg;
+	char lines[3][PATH_MAX];
+
+	for (g = 1; g < 3; ++g)
+		snprintf(lines[g], sizeof(lines[g]), "0::%s", grps[g] + strlen(grps[0]));
+
+	for (i = 0; i < n_iterations; ++i) {
+		cg_enter_current_thread(grps[(i % 2) + 1]);
+
+		if (proc_read_strstr(0, 1, "cgroup", lines[(i % 2) + 1]))
+			return (void *)-1;
+	}
+	return NULL;
+}
+
+/*
+ * Test single thread migration.
+ * Threaded cgroups allow successful migration of a thread.
+ */
+static int test_cgcore_thread_migration(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *dom = NULL;
+	char line[PATH_MAX];
+	char *grps[3] = { (char *)root, NULL, NULL };
+	pthread_t thr;
+	void *retval;
+
+	dom = cg_name(root, "cg_dom");
+	grps[1] = cg_name(root, "cg_dom/cg_src");
+	grps[2] = cg_name(root, "cg_dom/cg_dst");
+	if (!grps[1] || !grps[2] || !dom)
+		goto cleanup;
+
+	if (cg_create(dom))
+		goto cleanup;
+	if (cg_create(grps[1]))
+		goto cleanup;
+	if (cg_create(grps[2]))
+		goto cleanup;
+
+	if (cg_write(grps[1], "cgroup.type", "threaded"))
+		goto cleanup;
+	if (cg_write(grps[2], "cgroup.type", "threaded"))
+		goto cleanup;
+
+	if (cg_enter_current(grps[1]))
+		goto cleanup;
+
+	if (pthread_create(&thr, NULL, migrating_thread_fn, grps))
+		goto cleanup;
+
+	if (pthread_join(thr, &retval))
+		goto cleanup;
+
+	if (retval)
+		goto cleanup;
+
+	snprintf(line, sizeof(line), "0::%s", grps[1] + strlen(grps[0]));
+	if (proc_read_strstr(0, 1, "cgroup", line))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	cg_enter_current(root);
+	if (grps[2])
+		cg_destroy(grps[2]);
+	if (grps[1])
+		cg_destroy(grps[1]);
+	if (dom)
+		cg_destroy(dom);
+	free(grps[2]);
+	free(grps[1]);
+	free(dom);
+	return ret;
+}
+
+/*
+ * cgroup migration permission check should be performed based on the
+ * credentials at the time of open instead of write.
+ */
+static int test_cgcore_lesser_euid_open(const char *root)
+{
+	const uid_t test_euid = 65534;	/* usually nobody, any !root is fine */
+	int ret = KSFT_FAIL;
+	char *cg_test_a = NULL, *cg_test_b = NULL;
+	char *cg_test_a_procs = NULL, *cg_test_b_procs = NULL;
+	int cg_test_b_procs_fd = -1;
+	uid_t saved_uid;
+
+	cg_test_a = cg_name(root, "cg_test_a");
+	cg_test_b = cg_name(root, "cg_test_b");
+
+	if (!cg_test_a || !cg_test_b)
+		goto cleanup;
+
+	cg_test_a_procs = cg_name(cg_test_a, "cgroup.procs");
+	cg_test_b_procs = cg_name(cg_test_b, "cgroup.procs");
+
+	if (!cg_test_a_procs || !cg_test_b_procs)
+		goto cleanup;
+
+	if (cg_create(cg_test_a) || cg_create(cg_test_b))
+		goto cleanup;
+
+	if (cg_enter_current(cg_test_a))
+		goto cleanup;
+
+	if (chown(cg_test_a_procs, test_euid, -1) ||
+	    chown(cg_test_b_procs, test_euid, -1))
+		goto cleanup;
+
+	saved_uid = geteuid();
+	if (seteuid(test_euid))
+		goto cleanup;
+
+	cg_test_b_procs_fd = open(cg_test_b_procs, O_RDWR);
+
+	if (seteuid(saved_uid))
+		goto cleanup;
+
+	if (cg_test_b_procs_fd < 0)
+		goto cleanup;
+
+	if (write(cg_test_b_procs_fd, "0", 1) >= 0 || errno != EACCES)
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	cg_enter_current(root);
+	if (cg_test_b_procs_fd >= 0)
+		close(cg_test_b_procs_fd);
+	if (cg_test_b)
+		cg_destroy(cg_test_b);
+	if (cg_test_a)
+		cg_destroy(cg_test_a);
+	free(cg_test_b_procs);
+	free(cg_test_a_procs);
+	free(cg_test_b);
+	free(cg_test_a);
+	return ret;
+}
+
+struct lesser_ns_open_thread_arg {
+	const char	*path;
+	int		fd;
+	int		err;
+};
+
+static int lesser_ns_open_thread_fn(void *arg)
+{
+	struct lesser_ns_open_thread_arg *targ = arg;
+
+	targ->fd = open(targ->path, O_RDWR);
+	targ->err = errno;
+	return 0;
+}
+
+/*
+ * cgroup migration permission check should be performed based on the cgroup
+ * namespace at the time of open instead of write.
+ */
+static int test_cgcore_lesser_ns_open(const char *root)
+{
+	static char stack[65536];
+	const uid_t test_euid = 65534;	/* usually nobody, any !root is fine */
+	int ret = KSFT_FAIL;
+	char *cg_test_a = NULL, *cg_test_b = NULL;
+	char *cg_test_a_procs = NULL, *cg_test_b_procs = NULL;
+	int cg_test_b_procs_fd = -1;
+	struct lesser_ns_open_thread_arg targ = { .fd = -1 };
+	pid_t pid;
+	int status;
+
+	cg_test_a = cg_name(root, "cg_test_a");
+	cg_test_b = cg_name(root, "cg_test_b");
+
+	if (!cg_test_a || !cg_test_b)
+		goto cleanup;
+
+	cg_test_a_procs = cg_name(cg_test_a, "cgroup.procs");
+	cg_test_b_procs = cg_name(cg_test_b, "cgroup.procs");
+
+	if (!cg_test_a_procs || !cg_test_b_procs)
+		goto cleanup;
+
+	if (cg_create(cg_test_a) || cg_create(cg_test_b))
+		goto cleanup;
+
+	if (cg_enter_current(cg_test_b))
+		goto cleanup;
+
+	if (chown(cg_test_a_procs, test_euid, -1) ||
+	    chown(cg_test_b_procs, test_euid, -1))
+		goto cleanup;
+
+	targ.path = cg_test_b_procs;
+	pid = clone(lesser_ns_open_thread_fn, stack + sizeof(stack),
+		    CLONE_NEWCGROUP | CLONE_FILES | CLONE_VM | SIGCHLD,
+		    &targ);
+	if (pid < 0)
+		goto cleanup;
+
+	if (waitpid(pid, &status, 0) < 0)
+		goto cleanup;
+
+	if (!WIFEXITED(status))
+		goto cleanup;
+
+	cg_test_b_procs_fd = targ.fd;
+	if (cg_test_b_procs_fd < 0)
+		goto cleanup;
+
+	if (cg_enter_current(cg_test_a))
+		goto cleanup;
+
+	if ((status = write(cg_test_b_procs_fd, "0", 1)) >= 0 || errno != ENOENT)
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	cg_enter_current(root);
+	if (cg_test_b_procs_fd >= 0)
+		close(cg_test_b_procs_fd);
+	if (cg_test_b)
+		cg_destroy(cg_test_b);
+	if (cg_test_a)
+		cg_destroy(cg_test_a);
+	free(cg_test_b_procs);
+	free(cg_test_a_procs);
+	free(cg_test_b);
+	free(cg_test_a);
+	return ret;
+}
+
+#define T(x) { x, #x }
+struct corecg_test {
+	int (*fn)(const char *root);
+	const char *name;
+} tests[] = {
+	T(test_cgcore_internal_process_constraint),
+	T(test_cgcore_top_down_constraint_enable),
+	T(test_cgcore_top_down_constraint_disable),
+	T(test_cgcore_no_internal_process_constraint_on_threads),
+	T(test_cgcore_parent_becomes_threaded),
+	T(test_cgcore_invalid_domain),
+	T(test_cgcore_populated),
+	T(test_cgcore_proc_migration),
+	T(test_cgcore_thread_migration),
+	T(test_cgcore_destroy),
+	T(test_cgcore_lesser_euid_open),
+	T(test_cgcore_lesser_ns_open),
+};
+#undef T
+
+int main(int argc, char *argv[])
+{
+	char root[PATH_MAX];
+	int i, ret = EXIT_SUCCESS;
+
+	if (cg_find_unified_root(root, sizeof(root)))
+		ksft_exit_skip("cgroup v2 isn't mounted\n");
+
+	if (cg_read_strstr(root, "cgroup.subtree_control", "memory"))
+		if (cg_write(root, "cgroup.subtree_control", "+memory"))
+			ksft_exit_skip("Failed to set memory controller\n");
+
+	for (i = 0; i < ARRAY_SIZE(tests); i++) {
+		switch (tests[i].fn(root)) {
+		case KSFT_PASS:
+			ksft_test_result_pass("%s\n", tests[i].name);
+			break;
+		case KSFT_SKIP:
+			ksft_test_result_skip("%s\n", tests[i].name);
+			break;
+		default:
+			ret = EXIT_FAILURE;
+			ksft_test_result_fail("%s\n", tests[i].name);
+			break;
+		}
+	}
+
+	return ret;
+}
diff --git a/tools/testing/selftests/cgroup/test_freezer.c b/tools/testing/selftests/cgroup/test_freezer.c
new file mode 100644
index 000000000..23d8fa4a3
--- /dev/null
+++ b/tools/testing/selftests/cgroup/test_freezer.c
@@ -0,0 +1,905 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#include <stdbool.h>
+#include <linux/limits.h>
+#include <sys/ptrace.h>
+#include <sys/types.h>
+#include <sys/mman.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <errno.h>
+#include <poll.h>
+#include <stdlib.h>
+#include <sys/inotify.h>
+#include <string.h>
+#include <sys/wait.h>
+
+#include "../kselftest.h"
+#include "cgroup_util.h"
+
+#define DEBUG
+#ifdef DEBUG
+#define debug(args...) fprintf(stderr, args)
+#else
+#define debug(args...)
+#endif
+
+/*
+ * Check if the cgroup is frozen by looking at the cgroup.events::frozen value.
+ */
+static int cg_check_frozen(const char *cgroup, bool frozen)
+{
+	if (frozen) {
+		if (cg_read_strstr(cgroup, "cgroup.events", "frozen 1") != 0) {
+			debug("Cgroup %s isn't frozen\n", cgroup);
+			return -1;
+		}
+	} else {
+		/*
+		 * Check the cgroup.events::frozen value.
+		 */
+		if (cg_read_strstr(cgroup, "cgroup.events", "frozen 0") != 0) {
+			debug("Cgroup %s is frozen\n", cgroup);
+			return -1;
+		}
+	}
+
+	return 0;
+}
+
+/*
+ * Freeze the given cgroup.
+ */
+static int cg_freeze_nowait(const char *cgroup, bool freeze)
+{
+	return cg_write(cgroup, "cgroup.freeze", freeze ? "1" : "0");
+}
+
+/*
+ * Prepare for waiting on cgroup.events file.
+ */
+static int cg_prepare_for_wait(const char *cgroup)
+{
+	int fd, ret = -1;
+
+	fd = inotify_init1(0);
+	if (fd == -1) {
+		debug("Error: inotify_init1() failed\n");
+		return fd;
+	}
+
+	ret = inotify_add_watch(fd, cg_control(cgroup, "cgroup.events"),
+				IN_MODIFY);
+	if (ret == -1) {
+		debug("Error: inotify_add_watch() failed\n");
+		close(fd);
+		fd = -1;
+	}
+
+	return fd;
+}
+
+/*
+ * Wait for an event. If there are no events for 10 seconds,
+ * treat this an error.
+ */
+static int cg_wait_for(int fd)
+{
+	int ret = -1;
+	struct pollfd fds = {
+		.fd = fd,
+		.events = POLLIN,
+	};
+
+	while (true) {
+		ret = poll(&fds, 1, 10000);
+
+		if (ret == -1) {
+			if (errno == EINTR)
+				continue;
+			debug("Error: poll() failed\n");
+			break;
+		}
+
+		if (ret > 0 && fds.revents & POLLIN) {
+			ret = 0;
+			break;
+		}
+	}
+
+	return ret;
+}
+
+/*
+ * Attach a task to the given cgroup and wait for a cgroup frozen event.
+ * All transient events (e.g. populated) are ignored.
+ */
+static int cg_enter_and_wait_for_frozen(const char *cgroup, int pid,
+					bool frozen)
+{
+	int fd, ret = -1;
+	int attempts;
+
+	fd = cg_prepare_for_wait(cgroup);
+	if (fd < 0)
+		return fd;
+
+	ret = cg_enter(cgroup, pid);
+	if (ret)
+		goto out;
+
+	for (attempts = 0; attempts < 10; attempts++) {
+		ret = cg_wait_for(fd);
+		if (ret)
+			break;
+
+		ret = cg_check_frozen(cgroup, frozen);
+		if (ret)
+			continue;
+	}
+
+out:
+	close(fd);
+	return ret;
+}
+
+/*
+ * Freeze the given cgroup and wait for the inotify signal.
+ * If there are no events in 10 seconds, treat this as an error.
+ * Then check that the cgroup is in the desired state.
+ */
+static int cg_freeze_wait(const char *cgroup, bool freeze)
+{
+	int fd, ret = -1;
+
+	fd = cg_prepare_for_wait(cgroup);
+	if (fd < 0)
+		return fd;
+
+	ret = cg_freeze_nowait(cgroup, freeze);
+	if (ret) {
+		debug("Error: cg_freeze_nowait() failed\n");
+		goto out;
+	}
+
+	ret = cg_wait_for(fd);
+	if (ret)
+		goto out;
+
+	ret = cg_check_frozen(cgroup, freeze);
+out:
+	close(fd);
+	return ret;
+}
+
+/*
+ * A simple process running in a sleep loop until being
+ * re-parented.
+ */
+static int child_fn(const char *cgroup, void *arg)
+{
+	int ppid = getppid();
+
+	while (getppid() == ppid)
+		usleep(1000);
+
+	return getppid() == ppid;
+}
+
+/*
+ * A simple test for the cgroup freezer: populated the cgroup with 100
+ * running processes and freeze it. Then unfreeze it. Then it kills all
+ * processes and destroys the cgroup.
+ */
+static int test_cgfreezer_simple(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *cgroup = NULL;
+	int i;
+
+	cgroup = cg_name(root, "cg_test_simple");
+	if (!cgroup)
+		goto cleanup;
+
+	if (cg_create(cgroup))
+		goto cleanup;
+
+	for (i = 0; i < 100; i++)
+		cg_run_nowait(cgroup, child_fn, NULL);
+
+	if (cg_wait_for_proc_count(cgroup, 100))
+		goto cleanup;
+
+	if (cg_check_frozen(cgroup, false))
+		goto cleanup;
+
+	if (cg_freeze_wait(cgroup, true))
+		goto cleanup;
+
+	if (cg_freeze_wait(cgroup, false))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (cgroup)
+		cg_destroy(cgroup);
+	free(cgroup);
+	return ret;
+}
+
+/*
+ * The test creates the following hierarchy:
+ *       A
+ *    / / \ \
+ *   B  E  I K
+ *  /\  |
+ * C  D F
+ *      |
+ *      G
+ *      |
+ *      H
+ *
+ * with a process in C, H and 3 processes in K.
+ * Then it tries to freeze and unfreeze the whole tree.
+ */
+static int test_cgfreezer_tree(const char *root)
+{
+	char *cgroup[10] = {0};
+	int ret = KSFT_FAIL;
+	int i;
+
+	cgroup[0] = cg_name(root, "cg_test_tree_A");
+	if (!cgroup[0])
+		goto cleanup;
+
+	cgroup[1] = cg_name(cgroup[0], "B");
+	if (!cgroup[1])
+		goto cleanup;
+
+	cgroup[2] = cg_name(cgroup[1], "C");
+	if (!cgroup[2])
+		goto cleanup;
+
+	cgroup[3] = cg_name(cgroup[1], "D");
+	if (!cgroup[3])
+		goto cleanup;
+
+	cgroup[4] = cg_name(cgroup[0], "E");
+	if (!cgroup[4])
+		goto cleanup;
+
+	cgroup[5] = cg_name(cgroup[4], "F");
+	if (!cgroup[5])
+		goto cleanup;
+
+	cgroup[6] = cg_name(cgroup[5], "G");
+	if (!cgroup[6])
+		goto cleanup;
+
+	cgroup[7] = cg_name(cgroup[6], "H");
+	if (!cgroup[7])
+		goto cleanup;
+
+	cgroup[8] = cg_name(cgroup[0], "I");
+	if (!cgroup[8])
+		goto cleanup;
+
+	cgroup[9] = cg_name(cgroup[0], "K");
+	if (!cgroup[9])
+		goto cleanup;
+
+	for (i = 0; i < 10; i++)
+		if (cg_create(cgroup[i]))
+			goto cleanup;
+
+	cg_run_nowait(cgroup[2], child_fn, NULL);
+	cg_run_nowait(cgroup[7], child_fn, NULL);
+	cg_run_nowait(cgroup[9], child_fn, NULL);
+	cg_run_nowait(cgroup[9], child_fn, NULL);
+	cg_run_nowait(cgroup[9], child_fn, NULL);
+
+	/*
+	 * Wait until all child processes will enter
+	 * corresponding cgroups.
+	 */
+
+	if (cg_wait_for_proc_count(cgroup[2], 1) ||
+	    cg_wait_for_proc_count(cgroup[7], 1) ||
+	    cg_wait_for_proc_count(cgroup[9], 3))
+		goto cleanup;
+
+	/*
+	 * Freeze B.
+	 */
+	if (cg_freeze_wait(cgroup[1], true))
+		goto cleanup;
+
+	/*
+	 * Freeze F.
+	 */
+	if (cg_freeze_wait(cgroup[5], true))
+		goto cleanup;
+
+	/*
+	 * Freeze G.
+	 */
+	if (cg_freeze_wait(cgroup[6], true))
+		goto cleanup;
+
+	/*
+	 * Check that A and E are not frozen.
+	 */
+	if (cg_check_frozen(cgroup[0], false))
+		goto cleanup;
+
+	if (cg_check_frozen(cgroup[4], false))
+		goto cleanup;
+
+	/*
+	 * Freeze A. Check that A, B and E are frozen.
+	 */
+	if (cg_freeze_wait(cgroup[0], true))
+		goto cleanup;
+
+	if (cg_check_frozen(cgroup[1], true))
+		goto cleanup;
+
+	if (cg_check_frozen(cgroup[4], true))
+		goto cleanup;
+
+	/*
+	 * Unfreeze B, F and G
+	 */
+	if (cg_freeze_nowait(cgroup[1], false))
+		goto cleanup;
+
+	if (cg_freeze_nowait(cgroup[5], false))
+		goto cleanup;
+
+	if (cg_freeze_nowait(cgroup[6], false))
+		goto cleanup;
+
+	/*
+	 * Check that C and H are still frozen.
+	 */
+	if (cg_check_frozen(cgroup[2], true))
+		goto cleanup;
+
+	if (cg_check_frozen(cgroup[7], true))
+		goto cleanup;
+
+	/*
+	 * Unfreeze A. Check that A, C and K are not frozen.
+	 */
+	if (cg_freeze_wait(cgroup[0], false))
+		goto cleanup;
+
+	if (cg_check_frozen(cgroup[2], false))
+		goto cleanup;
+
+	if (cg_check_frozen(cgroup[9], false))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	for (i = 9; i >= 0 && cgroup[i]; i--) {
+		cg_destroy(cgroup[i]);
+		free(cgroup[i]);
+	}
+
+	return ret;
+}
+
+/*
+ * A fork bomb emulator.
+ */
+static int forkbomb_fn(const char *cgroup, void *arg)
+{
+	int ppid;
+
+	fork();
+	fork();
+
+	ppid = getppid();
+
+	while (getppid() == ppid)
+		usleep(1000);
+
+	return getppid() == ppid;
+}
+
+/*
+ * The test runs a fork bomb in a cgroup and tries to freeze it.
+ * Then it kills all processes and checks that cgroup isn't populated
+ * anymore.
+ */
+static int test_cgfreezer_forkbomb(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *cgroup = NULL;
+
+	cgroup = cg_name(root, "cg_forkbomb_test");
+	if (!cgroup)
+		goto cleanup;
+
+	if (cg_create(cgroup))
+		goto cleanup;
+
+	cg_run_nowait(cgroup, forkbomb_fn, NULL);
+
+	usleep(100000);
+
+	if (cg_freeze_wait(cgroup, true))
+		goto cleanup;
+
+	if (cg_killall(cgroup))
+		goto cleanup;
+
+	if (cg_wait_for_proc_count(cgroup, 0))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (cgroup)
+		cg_destroy(cgroup);
+	free(cgroup);
+	return ret;
+}
+
+/*
+ * The test creates a cgroups and freezes it. Then it creates a child cgroup
+ * and populates it with a task. After that it checks that the child cgroup
+ * is frozen and the parent cgroup remains frozen too.
+ */
+static int test_cgfreezer_mkdir(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent, *child = NULL;
+	int pid;
+
+	parent = cg_name(root, "cg_test_mkdir_A");
+	if (!parent)
+		goto cleanup;
+
+	child = cg_name(parent, "cg_test_mkdir_B");
+	if (!child)
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_freeze_wait(parent, true))
+		goto cleanup;
+
+	if (cg_create(child))
+		goto cleanup;
+
+	pid = cg_run_nowait(child, child_fn, NULL);
+	if (pid < 0)
+		goto cleanup;
+
+	if (cg_wait_for_proc_count(child, 1))
+		goto cleanup;
+
+	if (cg_check_frozen(child, true))
+		goto cleanup;
+
+	if (cg_check_frozen(parent, true))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (child)
+		cg_destroy(child);
+	free(child);
+	if (parent)
+		cg_destroy(parent);
+	free(parent);
+	return ret;
+}
+
+/*
+ * The test creates two nested cgroups, freezes the parent
+ * and removes the child. Then it checks that the parent cgroup
+ * remains frozen and it's possible to create a new child
+ * without unfreezing. The new child is frozen too.
+ */
+static int test_cgfreezer_rmdir(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent, *child = NULL;
+
+	parent = cg_name(root, "cg_test_rmdir_A");
+	if (!parent)
+		goto cleanup;
+
+	child = cg_name(parent, "cg_test_rmdir_B");
+	if (!child)
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_create(child))
+		goto cleanup;
+
+	if (cg_freeze_wait(parent, true))
+		goto cleanup;
+
+	if (cg_destroy(child))
+		goto cleanup;
+
+	if (cg_check_frozen(parent, true))
+		goto cleanup;
+
+	if (cg_create(child))
+		goto cleanup;
+
+	if (cg_check_frozen(child, true))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (child)
+		cg_destroy(child);
+	free(child);
+	if (parent)
+		cg_destroy(parent);
+	free(parent);
+	return ret;
+}
+
+/*
+ * The test creates two cgroups: A and B, runs a process in A
+ * and performs several migrations:
+ * 1) A (running) -> B (frozen)
+ * 2) B (frozen) -> A (running)
+ * 3) A (frozen) -> B (frozen)
+ *
+ * On each step it checks the actual state of both cgroups.
+ */
+static int test_cgfreezer_migrate(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *cgroup[2] = {0};
+	int pid;
+
+	cgroup[0] = cg_name(root, "cg_test_migrate_A");
+	if (!cgroup[0])
+		goto cleanup;
+
+	cgroup[1] = cg_name(root, "cg_test_migrate_B");
+	if (!cgroup[1])
+		goto cleanup;
+
+	if (cg_create(cgroup[0]))
+		goto cleanup;
+
+	if (cg_create(cgroup[1]))
+		goto cleanup;
+
+	pid = cg_run_nowait(cgroup[0], child_fn, NULL);
+	if (pid < 0)
+		goto cleanup;
+
+	if (cg_wait_for_proc_count(cgroup[0], 1))
+		goto cleanup;
+
+	/*
+	 * Migrate from A (running) to B (frozen)
+	 */
+	if (cg_freeze_wait(cgroup[1], true))
+		goto cleanup;
+
+	if (cg_enter_and_wait_for_frozen(cgroup[1], pid, true))
+		goto cleanup;
+
+	if (cg_check_frozen(cgroup[0], false))
+		goto cleanup;
+
+	/*
+	 * Migrate from B (frozen) to A (running)
+	 */
+	if (cg_enter_and_wait_for_frozen(cgroup[0], pid, false))
+		goto cleanup;
+
+	if (cg_check_frozen(cgroup[1], true))
+		goto cleanup;
+
+	/*
+	 * Migrate from A (frozen) to B (frozen)
+	 */
+	if (cg_freeze_wait(cgroup[0], true))
+		goto cleanup;
+
+	if (cg_enter_and_wait_for_frozen(cgroup[1], pid, true))
+		goto cleanup;
+
+	if (cg_check_frozen(cgroup[0], true))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (cgroup[0])
+		cg_destroy(cgroup[0]);
+	free(cgroup[0]);
+	if (cgroup[1])
+		cg_destroy(cgroup[1]);
+	free(cgroup[1]);
+	return ret;
+}
+
+/*
+ * The test checks that ptrace works with a tracing process in a frozen cgroup.
+ */
+static int test_cgfreezer_ptrace(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *cgroup = NULL;
+	siginfo_t siginfo;
+	int pid;
+
+	cgroup = cg_name(root, "cg_test_ptrace");
+	if (!cgroup)
+		goto cleanup;
+
+	if (cg_create(cgroup))
+		goto cleanup;
+
+	pid = cg_run_nowait(cgroup, child_fn, NULL);
+	if (pid < 0)
+		goto cleanup;
+
+	if (cg_wait_for_proc_count(cgroup, 1))
+		goto cleanup;
+
+	if (cg_freeze_wait(cgroup, true))
+		goto cleanup;
+
+	if (ptrace(PTRACE_SEIZE, pid, NULL, NULL))
+		goto cleanup;
+
+	if (ptrace(PTRACE_INTERRUPT, pid, NULL, NULL))
+		goto cleanup;
+
+	waitpid(pid, NULL, 0);
+
+	/*
+	 * Cgroup has to remain frozen, however the test task
+	 * is in traced state.
+	 */
+	if (cg_check_frozen(cgroup, true))
+		goto cleanup;
+
+	if (ptrace(PTRACE_GETSIGINFO, pid, NULL, &siginfo))
+		goto cleanup;
+
+	if (ptrace(PTRACE_DETACH, pid, NULL, NULL))
+		goto cleanup;
+
+	if (cg_check_frozen(cgroup, true))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (cgroup)
+		cg_destroy(cgroup);
+	free(cgroup);
+	return ret;
+}
+
+/*
+ * Check if the process is stopped.
+ */
+static int proc_check_stopped(int pid)
+{
+	char buf[PAGE_SIZE];
+	int len;
+
+	len = proc_read_text(pid, 0, "stat", buf, sizeof(buf));
+	if (len == -1) {
+		debug("Can't get %d stat\n", pid);
+		return -1;
+	}
+
+	if (strstr(buf, "(test_freezer) T ") == NULL) {
+		debug("Process %d in the unexpected state: %s\n", pid, buf);
+		return -1;
+	}
+
+	return 0;
+}
+
+/*
+ * Test that it's possible to freeze a cgroup with a stopped process.
+ */
+static int test_cgfreezer_stopped(const char *root)
+{
+	int pid, ret = KSFT_FAIL;
+	char *cgroup = NULL;
+
+	cgroup = cg_name(root, "cg_test_stopped");
+	if (!cgroup)
+		goto cleanup;
+
+	if (cg_create(cgroup))
+		goto cleanup;
+
+	pid = cg_run_nowait(cgroup, child_fn, NULL);
+
+	if (cg_wait_for_proc_count(cgroup, 1))
+		goto cleanup;
+
+	if (kill(pid, SIGSTOP))
+		goto cleanup;
+
+	if (cg_check_frozen(cgroup, false))
+		goto cleanup;
+
+	if (cg_freeze_wait(cgroup, true))
+		goto cleanup;
+
+	if (cg_freeze_wait(cgroup, false))
+		goto cleanup;
+
+	if (proc_check_stopped(pid))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (cgroup)
+		cg_destroy(cgroup);
+	free(cgroup);
+	return ret;
+}
+
+/*
+ * Test that it's possible to freeze a cgroup with a ptraced process.
+ */
+static int test_cgfreezer_ptraced(const char *root)
+{
+	int pid, ret = KSFT_FAIL;
+	char *cgroup = NULL;
+	siginfo_t siginfo;
+
+	cgroup = cg_name(root, "cg_test_ptraced");
+	if (!cgroup)
+		goto cleanup;
+
+	if (cg_create(cgroup))
+		goto cleanup;
+
+	pid = cg_run_nowait(cgroup, child_fn, NULL);
+
+	if (cg_wait_for_proc_count(cgroup, 1))
+		goto cleanup;
+
+	if (ptrace(PTRACE_SEIZE, pid, NULL, NULL))
+		goto cleanup;
+
+	if (ptrace(PTRACE_INTERRUPT, pid, NULL, NULL))
+		goto cleanup;
+
+	waitpid(pid, NULL, 0);
+
+	if (cg_check_frozen(cgroup, false))
+		goto cleanup;
+
+	if (cg_freeze_wait(cgroup, true))
+		goto cleanup;
+
+	/*
+	 * cg_check_frozen(cgroup, true) will fail here,
+	 * because the task in in the TRACEd state.
+	 */
+	if (cg_freeze_wait(cgroup, false))
+		goto cleanup;
+
+	if (ptrace(PTRACE_GETSIGINFO, pid, NULL, &siginfo))
+		goto cleanup;
+
+	if (ptrace(PTRACE_DETACH, pid, NULL, NULL))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (cgroup)
+		cg_destroy(cgroup);
+	free(cgroup);
+	return ret;
+}
+
+static int vfork_fn(const char *cgroup, void *arg)
+{
+	int pid = vfork();
+
+	if (pid == 0)
+		while (true)
+			sleep(1);
+
+	return pid;
+}
+
+/*
+ * Test that it's possible to freeze a cgroup with a process,
+ * which called vfork() and is waiting for a child.
+ */
+static int test_cgfreezer_vfork(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *cgroup = NULL;
+
+	cgroup = cg_name(root, "cg_test_vfork");
+	if (!cgroup)
+		goto cleanup;
+
+	if (cg_create(cgroup))
+		goto cleanup;
+
+	cg_run_nowait(cgroup, vfork_fn, NULL);
+
+	if (cg_wait_for_proc_count(cgroup, 2))
+		goto cleanup;
+
+	if (cg_freeze_wait(cgroup, true))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (cgroup)
+		cg_destroy(cgroup);
+	free(cgroup);
+	return ret;
+}
+
+#define T(x) { x, #x }
+struct cgfreezer_test {
+	int (*fn)(const char *root);
+	const char *name;
+} tests[] = {
+	T(test_cgfreezer_simple),
+	T(test_cgfreezer_tree),
+	T(test_cgfreezer_forkbomb),
+	T(test_cgfreezer_mkdir),
+	T(test_cgfreezer_rmdir),
+	T(test_cgfreezer_migrate),
+	T(test_cgfreezer_ptrace),
+	T(test_cgfreezer_stopped),
+	T(test_cgfreezer_ptraced),
+	T(test_cgfreezer_vfork),
+};
+#undef T
+
+int main(int argc, char *argv[])
+{
+	char root[PATH_MAX];
+	int i, ret = EXIT_SUCCESS;
+
+	if (cg_find_unified_root(root, sizeof(root)))
+		ksft_exit_skip("cgroup v2 isn't mounted\n");
+	for (i = 0; i < ARRAY_SIZE(tests); i++) {
+		switch (tests[i].fn(root)) {
+		case KSFT_PASS:
+			ksft_test_result_pass("%s\n", tests[i].name);
+			break;
+		case KSFT_SKIP:
+			ksft_test_result_skip("%s\n", tests[i].name);
+			break;
+		default:
+			ret = EXIT_FAILURE;
+			ksft_test_result_fail("%s\n", tests[i].name);
+			break;
+		}
+	}
+
+	return ret;
+}
diff --git a/tools/testing/selftests/cgroup/test_kmem.c b/tools/testing/selftests/cgroup/test_kmem.c
new file mode 100644
index 000000000..0941aa161
--- /dev/null
+++ b/tools/testing/selftests/cgroup/test_kmem.c
@@ -0,0 +1,450 @@
+// SPDX-License-Identifier: GPL-2.0
+#define _GNU_SOURCE
+
+#include <linux/limits.h>
+#include <fcntl.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <unistd.h>
+#include <sys/wait.h>
+#include <errno.h>
+#include <sys/sysinfo.h>
+#include <pthread.h>
+
+#include "../kselftest.h"
+#include "cgroup_util.h"
+
+
+/*
+ * Memory cgroup charging and vmstat data aggregation is performed using
+ * percpu batches 32 pages big (look at MEMCG_CHARGE_BATCH). So the maximum
+ * discrepancy between charge and vmstat entries is number of cpus multiplied
+ * by 32 pages multiplied by 2.
+ */
+#define MAX_VMSTAT_ERROR (4096 * 32 * 2 * get_nprocs())
+
+
+static int alloc_dcache(const char *cgroup, void *arg)
+{
+	unsigned long i;
+	struct stat st;
+	char buf[128];
+
+	for (i = 0; i < (unsigned long)arg; i++) {
+		snprintf(buf, sizeof(buf),
+			"/something-non-existent-with-a-long-name-%64lu-%d",
+			 i, getpid());
+		stat(buf, &st);
+	}
+
+	return 0;
+}
+
+/*
+ * This test allocates 100000 of negative dentries with long names.
+ * Then it checks that "slab" in memory.stat is larger than 1M.
+ * Then it sets memory.high to 1M and checks that at least 1/2
+ * of slab memory has been reclaimed.
+ */
+static int test_kmem_basic(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *cg = NULL;
+	long slab0, slab1, current;
+
+	cg = cg_name(root, "kmem_basic_test");
+	if (!cg)
+		goto cleanup;
+
+	if (cg_create(cg))
+		goto cleanup;
+
+	if (cg_run(cg, alloc_dcache, (void *)100000))
+		goto cleanup;
+
+	slab0 = cg_read_key_long(cg, "memory.stat", "slab ");
+	if (slab0 < (1 << 20))
+		goto cleanup;
+
+	cg_write(cg, "memory.high", "1M");
+	slab1 = cg_read_key_long(cg, "memory.stat", "slab ");
+	if (slab1 <= 0)
+		goto cleanup;
+
+	current = cg_read_long(cg, "memory.current");
+	if (current <= 0)
+		goto cleanup;
+
+	if (slab1 < slab0 / 2 && current < slab0 / 2)
+		ret = KSFT_PASS;
+cleanup:
+	cg_destroy(cg);
+	free(cg);
+
+	return ret;
+}
+
+static void *alloc_kmem_fn(void *arg)
+{
+	alloc_dcache(NULL, (void *)100);
+	return NULL;
+}
+
+static int alloc_kmem_smp(const char *cgroup, void *arg)
+{
+	int nr_threads = 2 * get_nprocs();
+	pthread_t *tinfo;
+	unsigned long i;
+	int ret = -1;
+
+	tinfo = calloc(nr_threads, sizeof(pthread_t));
+	if (tinfo == NULL)
+		return -1;
+
+	for (i = 0; i < nr_threads; i++) {
+		if (pthread_create(&tinfo[i], NULL, &alloc_kmem_fn,
+				   (void *)i)) {
+			free(tinfo);
+			return -1;
+		}
+	}
+
+	for (i = 0; i < nr_threads; i++) {
+		ret = pthread_join(tinfo[i], NULL);
+		if (ret)
+			break;
+	}
+
+	free(tinfo);
+	return ret;
+}
+
+static int cg_run_in_subcgroups(const char *parent,
+				int (*fn)(const char *cgroup, void *arg),
+				void *arg, int times)
+{
+	char *child;
+	int i;
+
+	for (i = 0; i < times; i++) {
+		child = cg_name_indexed(parent, "child", i);
+		if (!child)
+			return -1;
+
+		if (cg_create(child)) {
+			cg_destroy(child);
+			free(child);
+			return -1;
+		}
+
+		if (cg_run(child, fn, NULL)) {
+			cg_destroy(child);
+			free(child);
+			return -1;
+		}
+
+		cg_destroy(child);
+		free(child);
+	}
+
+	return 0;
+}
+
+/*
+ * The test creates and destroys a large number of cgroups. In each cgroup it
+ * allocates some slab memory (mostly negative dentries) using 2 * NR_CPUS
+ * threads. Then it checks the sanity of numbers on the parent level:
+ * the total size of the cgroups should be roughly equal to
+ * anon + file + slab + kernel_stack.
+ */
+static int test_kmem_memcg_deletion(const char *root)
+{
+	long current, slab, anon, file, kernel_stack, sum;
+	int ret = KSFT_FAIL;
+	char *parent;
+
+	parent = cg_name(root, "kmem_memcg_deletion_test");
+	if (!parent)
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_write(parent, "cgroup.subtree_control", "+memory"))
+		goto cleanup;
+
+	if (cg_run_in_subcgroups(parent, alloc_kmem_smp, NULL, 100))
+		goto cleanup;
+
+	current = cg_read_long(parent, "memory.current");
+	slab = cg_read_key_long(parent, "memory.stat", "slab ");
+	anon = cg_read_key_long(parent, "memory.stat", "anon ");
+	file = cg_read_key_long(parent, "memory.stat", "file ");
+	kernel_stack = cg_read_key_long(parent, "memory.stat", "kernel_stack ");
+	if (current < 0 || slab < 0 || anon < 0 || file < 0 ||
+	    kernel_stack < 0)
+		goto cleanup;
+
+	sum = slab + anon + file + kernel_stack;
+	if (abs(sum - current) < MAX_VMSTAT_ERROR) {
+		ret = KSFT_PASS;
+	} else {
+		printf("memory.current = %ld\n", current);
+		printf("slab + anon + file + kernel_stack = %ld\n", sum);
+		printf("slab = %ld\n", slab);
+		printf("anon = %ld\n", anon);
+		printf("file = %ld\n", file);
+		printf("kernel_stack = %ld\n", kernel_stack);
+	}
+
+cleanup:
+	cg_destroy(parent);
+	free(parent);
+
+	return ret;
+}
+
+/*
+ * The test reads the entire /proc/kpagecgroup. If the operation went
+ * successfully (and the kernel didn't panic), the test is treated as passed.
+ */
+static int test_kmem_proc_kpagecgroup(const char *root)
+{
+	unsigned long buf[128];
+	int ret = KSFT_FAIL;
+	ssize_t len;
+	int fd;
+
+	fd = open("/proc/kpagecgroup", O_RDONLY);
+	if (fd < 0)
+		return ret;
+
+	do {
+		len = read(fd, buf, sizeof(buf));
+	} while (len > 0);
+
+	if (len == 0)
+		ret = KSFT_PASS;
+
+	close(fd);
+	return ret;
+}
+
+static void *pthread_wait_fn(void *arg)
+{
+	sleep(100);
+	return NULL;
+}
+
+static int spawn_1000_threads(const char *cgroup, void *arg)
+{
+	int nr_threads = 1000;
+	pthread_t *tinfo;
+	unsigned long i;
+	long stack;
+	int ret = -1;
+
+	tinfo = calloc(nr_threads, sizeof(pthread_t));
+	if (tinfo == NULL)
+		return -1;
+
+	for (i = 0; i < nr_threads; i++) {
+		if (pthread_create(&tinfo[i], NULL, &pthread_wait_fn,
+				   (void *)i)) {
+			free(tinfo);
+			return(-1);
+		}
+	}
+
+	stack = cg_read_key_long(cgroup, "memory.stat", "kernel_stack ");
+	if (stack >= 4096 * 1000)
+		ret = 0;
+
+	free(tinfo);
+	return ret;
+}
+
+/*
+ * The test spawns a process, which spawns 1000 threads. Then it checks
+ * that memory.stat's kernel_stack is at least 1000 pages large.
+ */
+static int test_kmem_kernel_stacks(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *cg = NULL;
+
+	cg = cg_name(root, "kmem_kernel_stacks_test");
+	if (!cg)
+		goto cleanup;
+
+	if (cg_create(cg))
+		goto cleanup;
+
+	if (cg_run(cg, spawn_1000_threads, NULL))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+cleanup:
+	cg_destroy(cg);
+	free(cg);
+
+	return ret;
+}
+
+/*
+ * This test sequentionally creates 30 child cgroups, allocates some
+ * kernel memory in each of them, and deletes them. Then it checks
+ * that the number of dying cgroups on the parent level is 0.
+ */
+static int test_kmem_dead_cgroups(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent;
+	long dead;
+	int i;
+
+	parent = cg_name(root, "kmem_dead_cgroups_test");
+	if (!parent)
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_write(parent, "cgroup.subtree_control", "+memory"))
+		goto cleanup;
+
+	if (cg_run_in_subcgroups(parent, alloc_dcache, (void *)100, 30))
+		goto cleanup;
+
+	for (i = 0; i < 5; i++) {
+		dead = cg_read_key_long(parent, "cgroup.stat",
+					"nr_dying_descendants ");
+		if (dead == 0) {
+			ret = KSFT_PASS;
+			break;
+		}
+		/*
+		 * Reclaiming cgroups might take some time,
+		 * let's wait a bit and repeat.
+		 */
+		sleep(1);
+	}
+
+cleanup:
+	cg_destroy(parent);
+	free(parent);
+
+	return ret;
+}
+
+/*
+ * This test creates a sub-tree with 1000 memory cgroups.
+ * Then it checks that the memory.current on the parent level
+ * is greater than 0 and approximates matches the percpu value
+ * from memory.stat.
+ */
+static int test_percpu_basic(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent, *child;
+	long current, percpu;
+	int i;
+
+	parent = cg_name(root, "percpu_basic_test");
+	if (!parent)
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_write(parent, "cgroup.subtree_control", "+memory"))
+		goto cleanup;
+
+	for (i = 0; i < 1000; i++) {
+		child = cg_name_indexed(parent, "child", i);
+		if (!child)
+			return -1;
+
+		if (cg_create(child))
+			goto cleanup_children;
+
+		free(child);
+	}
+
+	current = cg_read_long(parent, "memory.current");
+	percpu = cg_read_key_long(parent, "memory.stat", "percpu ");
+
+	if (current > 0 && percpu > 0 && abs(current - percpu) <
+	    MAX_VMSTAT_ERROR)
+		ret = KSFT_PASS;
+	else
+		printf("memory.current %ld\npercpu %ld\n",
+		       current, percpu);
+
+cleanup_children:
+	for (i = 0; i < 1000; i++) {
+		child = cg_name_indexed(parent, "child", i);
+		cg_destroy(child);
+		free(child);
+	}
+
+cleanup:
+	cg_destroy(parent);
+	free(parent);
+
+	return ret;
+}
+
+#define T(x) { x, #x }
+struct kmem_test {
+	int (*fn)(const char *root);
+	const char *name;
+} tests[] = {
+	T(test_kmem_basic),
+	T(test_kmem_memcg_deletion),
+	T(test_kmem_proc_kpagecgroup),
+	T(test_kmem_kernel_stacks),
+	T(test_kmem_dead_cgroups),
+	T(test_percpu_basic),
+};
+#undef T
+
+int main(int argc, char **argv)
+{
+	char root[PATH_MAX];
+	int i, ret = EXIT_SUCCESS;
+
+	if (cg_find_unified_root(root, sizeof(root)))
+		ksft_exit_skip("cgroup v2 isn't mounted\n");
+
+	/*
+	 * Check that memory controller is available:
+	 * memory is listed in cgroup.controllers
+	 */
+	if (cg_read_strstr(root, "cgroup.controllers", "memory"))
+		ksft_exit_skip("memory controller isn't available\n");
+
+	if (cg_read_strstr(root, "cgroup.subtree_control", "memory"))
+		if (cg_write(root, "cgroup.subtree_control", "+memory"))
+			ksft_exit_skip("Failed to set memory controller\n");
+
+	for (i = 0; i < ARRAY_SIZE(tests); i++) {
+		switch (tests[i].fn(root)) {
+		case KSFT_PASS:
+			ksft_test_result_pass("%s\n", tests[i].name);
+			break;
+		case KSFT_SKIP:
+			ksft_test_result_skip("%s\n", tests[i].name);
+			break;
+		default:
+			ret = EXIT_FAILURE;
+			ksft_test_result_fail("%s\n", tests[i].name);
+			break;
+		}
+	}
+
+	return ret;
+}
diff --git a/tools/testing/selftests/cgroup/test_memcontrol.c b/tools/testing/selftests/cgroup/test_memcontrol.c
new file mode 100644
index 000000000..c19a97dd0
--- /dev/null
+++ b/tools/testing/selftests/cgroup/test_memcontrol.c
@@ -0,0 +1,1228 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#define _GNU_SOURCE
+
+#include <linux/limits.h>
+#include <linux/oom.h>
+#include <fcntl.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/stat.h>
+#include <sys/types.h>
+#include <unistd.h>
+#include <sys/socket.h>
+#include <sys/wait.h>
+#include <arpa/inet.h>
+#include <netinet/in.h>
+#include <netdb.h>
+#include <errno.h>
+
+#include "../kselftest.h"
+#include "cgroup_util.h"
+
+/*
+ * This test creates two nested cgroups with and without enabling
+ * the memory controller.
+ */
+static int test_memcg_subtree_control(const char *root)
+{
+	char *parent, *child, *parent2 = NULL, *child2 = NULL;
+	int ret = KSFT_FAIL;
+	char buf[PAGE_SIZE];
+
+	/* Create two nested cgroups with the memory controller enabled */
+	parent = cg_name(root, "memcg_test_0");
+	child = cg_name(root, "memcg_test_0/memcg_test_1");
+	if (!parent || !child)
+		goto cleanup_free;
+
+	if (cg_create(parent))
+		goto cleanup_free;
+
+	if (cg_write(parent, "cgroup.subtree_control", "+memory"))
+		goto cleanup_parent;
+
+	if (cg_create(child))
+		goto cleanup_parent;
+
+	if (cg_read_strstr(child, "cgroup.controllers", "memory"))
+		goto cleanup_child;
+
+	/* Create two nested cgroups without enabling memory controller */
+	parent2 = cg_name(root, "memcg_test_1");
+	child2 = cg_name(root, "memcg_test_1/memcg_test_1");
+	if (!parent2 || !child2)
+		goto cleanup_free2;
+
+	if (cg_create(parent2))
+		goto cleanup_free2;
+
+	if (cg_create(child2))
+		goto cleanup_parent2;
+
+	if (cg_read(child2, "cgroup.controllers", buf, sizeof(buf)))
+		goto cleanup_all;
+
+	if (!cg_read_strstr(child2, "cgroup.controllers", "memory"))
+		goto cleanup_all;
+
+	ret = KSFT_PASS;
+
+cleanup_all:
+	cg_destroy(child2);
+cleanup_parent2:
+	cg_destroy(parent2);
+cleanup_free2:
+	free(parent2);
+	free(child2);
+cleanup_child:
+	cg_destroy(child);
+cleanup_parent:
+	cg_destroy(parent);
+cleanup_free:
+	free(parent);
+	free(child);
+
+	return ret;
+}
+
+static int alloc_anon_50M_check(const char *cgroup, void *arg)
+{
+	size_t size = MB(50);
+	char *buf, *ptr;
+	long anon, current;
+	int ret = -1;
+
+	buf = malloc(size);
+	for (ptr = buf; ptr < buf + size; ptr += PAGE_SIZE)
+		*ptr = 0;
+
+	current = cg_read_long(cgroup, "memory.current");
+	if (current < size)
+		goto cleanup;
+
+	if (!values_close(size, current, 3))
+		goto cleanup;
+
+	anon = cg_read_key_long(cgroup, "memory.stat", "anon ");
+	if (anon < 0)
+		goto cleanup;
+
+	if (!values_close(anon, current, 3))
+		goto cleanup;
+
+	ret = 0;
+cleanup:
+	free(buf);
+	return ret;
+}
+
+static int alloc_pagecache_50M_check(const char *cgroup, void *arg)
+{
+	size_t size = MB(50);
+	int ret = -1;
+	long current, file;
+	int fd;
+
+	fd = get_temp_fd();
+	if (fd < 0)
+		return -1;
+
+	if (alloc_pagecache(fd, size))
+		goto cleanup;
+
+	current = cg_read_long(cgroup, "memory.current");
+	if (current < size)
+		goto cleanup;
+
+	file = cg_read_key_long(cgroup, "memory.stat", "file ");
+	if (file < 0)
+		goto cleanup;
+
+	if (!values_close(file, current, 10))
+		goto cleanup;
+
+	ret = 0;
+
+cleanup:
+	close(fd);
+	return ret;
+}
+
+/*
+ * This test create a memory cgroup, allocates
+ * some anonymous memory and some pagecache
+ * and check memory.current and some memory.stat values.
+ */
+static int test_memcg_current(const char *root)
+{
+	int ret = KSFT_FAIL;
+	long current;
+	char *memcg;
+
+	memcg = cg_name(root, "memcg_test");
+	if (!memcg)
+		goto cleanup;
+
+	if (cg_create(memcg))
+		goto cleanup;
+
+	current = cg_read_long(memcg, "memory.current");
+	if (current != 0)
+		goto cleanup;
+
+	if (cg_run(memcg, alloc_anon_50M_check, NULL))
+		goto cleanup;
+
+	if (cg_run(memcg, alloc_pagecache_50M_check, NULL))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	cg_destroy(memcg);
+	free(memcg);
+
+	return ret;
+}
+
+static int alloc_pagecache_50M(const char *cgroup, void *arg)
+{
+	int fd = (long)arg;
+
+	return alloc_pagecache(fd, MB(50));
+}
+
+static int alloc_pagecache_50M_noexit(const char *cgroup, void *arg)
+{
+	int fd = (long)arg;
+	int ppid = getppid();
+
+	if (alloc_pagecache(fd, MB(50)))
+		return -1;
+
+	while (getppid() == ppid)
+		sleep(1);
+
+	return 0;
+}
+
+static int alloc_anon_noexit(const char *cgroup, void *arg)
+{
+	int ppid = getppid();
+
+	if (alloc_anon(cgroup, arg))
+		return -1;
+
+	while (getppid() == ppid)
+		sleep(1);
+
+	return 0;
+}
+
+/*
+ * Wait until processes are killed asynchronously by the OOM killer
+ * If we exceed a timeout, fail.
+ */
+static int cg_test_proc_killed(const char *cgroup)
+{
+	int limit;
+
+	for (limit = 10; limit > 0; limit--) {
+		if (cg_read_strcmp(cgroup, "cgroup.procs", "") == 0)
+			return 0;
+
+		usleep(100000);
+	}
+	return -1;
+}
+
+/*
+ * First, this test creates the following hierarchy:
+ * A       memory.min = 50M,  memory.max = 200M
+ * A/B     memory.min = 50M,  memory.current = 50M
+ * A/B/C   memory.min = 75M,  memory.current = 50M
+ * A/B/D   memory.min = 25M,  memory.current = 50M
+ * A/B/E   memory.min = 500M, memory.current = 0
+ * A/B/F   memory.min = 0,    memory.current = 50M
+ *
+ * Usages are pagecache, but the test keeps a running
+ * process in every leaf cgroup.
+ * Then it creates A/G and creates a significant
+ * memory pressure in it.
+ *
+ * A/B    memory.current ~= 50M
+ * A/B/C  memory.current ~= 33M
+ * A/B/D  memory.current ~= 17M
+ * A/B/E  memory.current ~= 0
+ *
+ * After that it tries to allocate more than there is
+ * unprotected memory in A available, and checks
+ * checks that memory.min protects pagecache even
+ * in this case.
+ */
+static int test_memcg_min(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent[3] = {NULL};
+	char *children[4] = {NULL};
+	long c[4];
+	int i, attempts;
+	int fd;
+
+	fd = get_temp_fd();
+	if (fd < 0)
+		goto cleanup;
+
+	parent[0] = cg_name(root, "memcg_test_0");
+	if (!parent[0])
+		goto cleanup;
+
+	parent[1] = cg_name(parent[0], "memcg_test_1");
+	if (!parent[1])
+		goto cleanup;
+
+	parent[2] = cg_name(parent[0], "memcg_test_2");
+	if (!parent[2])
+		goto cleanup;
+
+	if (cg_create(parent[0]))
+		goto cleanup;
+
+	if (cg_read_long(parent[0], "memory.min")) {
+		ret = KSFT_SKIP;
+		goto cleanup;
+	}
+
+	if (cg_write(parent[0], "cgroup.subtree_control", "+memory"))
+		goto cleanup;
+
+	if (cg_write(parent[0], "memory.max", "200M"))
+		goto cleanup;
+
+	if (cg_write(parent[0], "memory.swap.max", "0"))
+		goto cleanup;
+
+	if (cg_create(parent[1]))
+		goto cleanup;
+
+	if (cg_write(parent[1], "cgroup.subtree_control", "+memory"))
+		goto cleanup;
+
+	if (cg_create(parent[2]))
+		goto cleanup;
+
+	for (i = 0; i < ARRAY_SIZE(children); i++) {
+		children[i] = cg_name_indexed(parent[1], "child_memcg", i);
+		if (!children[i])
+			goto cleanup;
+
+		if (cg_create(children[i]))
+			goto cleanup;
+
+		if (i == 2)
+			continue;
+
+		cg_run_nowait(children[i], alloc_pagecache_50M_noexit,
+			      (void *)(long)fd);
+	}
+
+	if (cg_write(parent[0], "memory.min", "50M"))
+		goto cleanup;
+	if (cg_write(parent[1], "memory.min", "50M"))
+		goto cleanup;
+	if (cg_write(children[0], "memory.min", "75M"))
+		goto cleanup;
+	if (cg_write(children[1], "memory.min", "25M"))
+		goto cleanup;
+	if (cg_write(children[2], "memory.min", "500M"))
+		goto cleanup;
+	if (cg_write(children[3], "memory.min", "0"))
+		goto cleanup;
+
+	attempts = 0;
+	while (!values_close(cg_read_long(parent[1], "memory.current"),
+			     MB(150), 3)) {
+		if (attempts++ > 5)
+			break;
+		sleep(1);
+	}
+
+	if (cg_run(parent[2], alloc_anon, (void *)MB(148)))
+		goto cleanup;
+
+	if (!values_close(cg_read_long(parent[1], "memory.current"), MB(50), 3))
+		goto cleanup;
+
+	for (i = 0; i < ARRAY_SIZE(children); i++)
+		c[i] = cg_read_long(children[i], "memory.current");
+
+	if (!values_close(c[0], MB(33), 10))
+		goto cleanup;
+
+	if (!values_close(c[1], MB(17), 10))
+		goto cleanup;
+
+	if (!values_close(c[2], 0, 1))
+		goto cleanup;
+
+	if (!cg_run(parent[2], alloc_anon, (void *)MB(170)))
+		goto cleanup;
+
+	if (!values_close(cg_read_long(parent[1], "memory.current"), MB(50), 3))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	for (i = ARRAY_SIZE(children) - 1; i >= 0; i--) {
+		if (!children[i])
+			continue;
+
+		cg_destroy(children[i]);
+		free(children[i]);
+	}
+
+	for (i = ARRAY_SIZE(parent) - 1; i >= 0; i--) {
+		if (!parent[i])
+			continue;
+
+		cg_destroy(parent[i]);
+		free(parent[i]);
+	}
+	close(fd);
+	return ret;
+}
+
+/*
+ * First, this test creates the following hierarchy:
+ * A       memory.low = 50M,  memory.max = 200M
+ * A/B     memory.low = 50M,  memory.current = 50M
+ * A/B/C   memory.low = 75M,  memory.current = 50M
+ * A/B/D   memory.low = 25M,  memory.current = 50M
+ * A/B/E   memory.low = 500M, memory.current = 0
+ * A/B/F   memory.low = 0,    memory.current = 50M
+ *
+ * Usages are pagecache.
+ * Then it creates A/G an creates a significant
+ * memory pressure in it.
+ *
+ * Then it checks actual memory usages and expects that:
+ * A/B    memory.current ~= 50M
+ * A/B/   memory.current ~= 33M
+ * A/B/D  memory.current ~= 17M
+ * A/B/E  memory.current ~= 0
+ *
+ * After that it tries to allocate more than there is
+ * unprotected memory in A available,
+ * and checks low and oom events in memory.events.
+ */
+static int test_memcg_low(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent[3] = {NULL};
+	char *children[4] = {NULL};
+	long low, oom;
+	long c[4];
+	int i;
+	int fd;
+
+	fd = get_temp_fd();
+	if (fd < 0)
+		goto cleanup;
+
+	parent[0] = cg_name(root, "memcg_test_0");
+	if (!parent[0])
+		goto cleanup;
+
+	parent[1] = cg_name(parent[0], "memcg_test_1");
+	if (!parent[1])
+		goto cleanup;
+
+	parent[2] = cg_name(parent[0], "memcg_test_2");
+	if (!parent[2])
+		goto cleanup;
+
+	if (cg_create(parent[0]))
+		goto cleanup;
+
+	if (cg_read_long(parent[0], "memory.low"))
+		goto cleanup;
+
+	if (cg_write(parent[0], "cgroup.subtree_control", "+memory"))
+		goto cleanup;
+
+	if (cg_write(parent[0], "memory.max", "200M"))
+		goto cleanup;
+
+	if (cg_write(parent[0], "memory.swap.max", "0"))
+		goto cleanup;
+
+	if (cg_create(parent[1]))
+		goto cleanup;
+
+	if (cg_write(parent[1], "cgroup.subtree_control", "+memory"))
+		goto cleanup;
+
+	if (cg_create(parent[2]))
+		goto cleanup;
+
+	for (i = 0; i < ARRAY_SIZE(children); i++) {
+		children[i] = cg_name_indexed(parent[1], "child_memcg", i);
+		if (!children[i])
+			goto cleanup;
+
+		if (cg_create(children[i]))
+			goto cleanup;
+
+		if (i == 2)
+			continue;
+
+		if (cg_run(children[i], alloc_pagecache_50M, (void *)(long)fd))
+			goto cleanup;
+	}
+
+	if (cg_write(parent[0], "memory.low", "50M"))
+		goto cleanup;
+	if (cg_write(parent[1], "memory.low", "50M"))
+		goto cleanup;
+	if (cg_write(children[0], "memory.low", "75M"))
+		goto cleanup;
+	if (cg_write(children[1], "memory.low", "25M"))
+		goto cleanup;
+	if (cg_write(children[2], "memory.low", "500M"))
+		goto cleanup;
+	if (cg_write(children[3], "memory.low", "0"))
+		goto cleanup;
+
+	if (cg_run(parent[2], alloc_anon, (void *)MB(148)))
+		goto cleanup;
+
+	if (!values_close(cg_read_long(parent[1], "memory.current"), MB(50), 3))
+		goto cleanup;
+
+	for (i = 0; i < ARRAY_SIZE(children); i++)
+		c[i] = cg_read_long(children[i], "memory.current");
+
+	if (!values_close(c[0], MB(33), 10))
+		goto cleanup;
+
+	if (!values_close(c[1], MB(17), 10))
+		goto cleanup;
+
+	if (!values_close(c[2], 0, 1))
+		goto cleanup;
+
+	if (cg_run(parent[2], alloc_anon, (void *)MB(166))) {
+		fprintf(stderr,
+			"memory.low prevents from allocating anon memory\n");
+		goto cleanup;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(children); i++) {
+		oom = cg_read_key_long(children[i], "memory.events", "oom ");
+		low = cg_read_key_long(children[i], "memory.events", "low ");
+
+		if (oom)
+			goto cleanup;
+		if (i < 2 && low <= 0)
+			goto cleanup;
+		if (i >= 2 && low)
+			goto cleanup;
+	}
+
+	ret = KSFT_PASS;
+
+cleanup:
+	for (i = ARRAY_SIZE(children) - 1; i >= 0; i--) {
+		if (!children[i])
+			continue;
+
+		cg_destroy(children[i]);
+		free(children[i]);
+	}
+
+	for (i = ARRAY_SIZE(parent) - 1; i >= 0; i--) {
+		if (!parent[i])
+			continue;
+
+		cg_destroy(parent[i]);
+		free(parent[i]);
+	}
+	close(fd);
+	return ret;
+}
+
+static int alloc_pagecache_max_30M(const char *cgroup, void *arg)
+{
+	size_t size = MB(50);
+	int ret = -1;
+	long current;
+	int fd;
+
+	fd = get_temp_fd();
+	if (fd < 0)
+		return -1;
+
+	if (alloc_pagecache(fd, size))
+		goto cleanup;
+
+	current = cg_read_long(cgroup, "memory.current");
+	if (current <= MB(29) || current > MB(30))
+		goto cleanup;
+
+	ret = 0;
+
+cleanup:
+	close(fd);
+	return ret;
+
+}
+
+/*
+ * This test checks that memory.high limits the amount of
+ * memory which can be consumed by either anonymous memory
+ * or pagecache.
+ */
+static int test_memcg_high(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *memcg;
+	long high;
+
+	memcg = cg_name(root, "memcg_test");
+	if (!memcg)
+		goto cleanup;
+
+	if (cg_create(memcg))
+		goto cleanup;
+
+	if (cg_read_strcmp(memcg, "memory.high", "max\n"))
+		goto cleanup;
+
+	if (cg_write(memcg, "memory.swap.max", "0"))
+		goto cleanup;
+
+	if (cg_write(memcg, "memory.high", "30M"))
+		goto cleanup;
+
+	if (cg_run(memcg, alloc_anon, (void *)MB(100)))
+		goto cleanup;
+
+	if (!cg_run(memcg, alloc_pagecache_50M_check, NULL))
+		goto cleanup;
+
+	if (cg_run(memcg, alloc_pagecache_max_30M, NULL))
+		goto cleanup;
+
+	high = cg_read_key_long(memcg, "memory.events", "high ");
+	if (high <= 0)
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	cg_destroy(memcg);
+	free(memcg);
+
+	return ret;
+}
+
+/*
+ * This test checks that memory.max limits the amount of
+ * memory which can be consumed by either anonymous memory
+ * or pagecache.
+ */
+static int test_memcg_max(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *memcg;
+	long current, max;
+
+	memcg = cg_name(root, "memcg_test");
+	if (!memcg)
+		goto cleanup;
+
+	if (cg_create(memcg))
+		goto cleanup;
+
+	if (cg_read_strcmp(memcg, "memory.max", "max\n"))
+		goto cleanup;
+
+	if (cg_write(memcg, "memory.swap.max", "0"))
+		goto cleanup;
+
+	if (cg_write(memcg, "memory.max", "30M"))
+		goto cleanup;
+
+	/* Should be killed by OOM killer */
+	if (!cg_run(memcg, alloc_anon, (void *)MB(100)))
+		goto cleanup;
+
+	if (cg_run(memcg, alloc_pagecache_max_30M, NULL))
+		goto cleanup;
+
+	current = cg_read_long(memcg, "memory.current");
+	if (current > MB(30) || !current)
+		goto cleanup;
+
+	max = cg_read_key_long(memcg, "memory.events", "max ");
+	if (max <= 0)
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	cg_destroy(memcg);
+	free(memcg);
+
+	return ret;
+}
+
+static int alloc_anon_50M_check_swap(const char *cgroup, void *arg)
+{
+	long mem_max = (long)arg;
+	size_t size = MB(50);
+	char *buf, *ptr;
+	long mem_current, swap_current;
+	int ret = -1;
+
+	buf = malloc(size);
+	for (ptr = buf; ptr < buf + size; ptr += PAGE_SIZE)
+		*ptr = 0;
+
+	mem_current = cg_read_long(cgroup, "memory.current");
+	if (!mem_current || !values_close(mem_current, mem_max, 3))
+		goto cleanup;
+
+	swap_current = cg_read_long(cgroup, "memory.swap.current");
+	if (!swap_current ||
+	    !values_close(mem_current + swap_current, size, 3))
+		goto cleanup;
+
+	ret = 0;
+cleanup:
+	free(buf);
+	return ret;
+}
+
+/*
+ * This test checks that memory.swap.max limits the amount of
+ * anonymous memory which can be swapped out.
+ */
+static int test_memcg_swap_max(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *memcg;
+	long max;
+
+	if (!is_swap_enabled())
+		return KSFT_SKIP;
+
+	memcg = cg_name(root, "memcg_test");
+	if (!memcg)
+		goto cleanup;
+
+	if (cg_create(memcg))
+		goto cleanup;
+
+	if (cg_read_long(memcg, "memory.swap.current")) {
+		ret = KSFT_SKIP;
+		goto cleanup;
+	}
+
+	if (cg_read_strcmp(memcg, "memory.max", "max\n"))
+		goto cleanup;
+
+	if (cg_read_strcmp(memcg, "memory.swap.max", "max\n"))
+		goto cleanup;
+
+	if (cg_write(memcg, "memory.swap.max", "30M"))
+		goto cleanup;
+
+	if (cg_write(memcg, "memory.max", "30M"))
+		goto cleanup;
+
+	/* Should be killed by OOM killer */
+	if (!cg_run(memcg, alloc_anon, (void *)MB(100)))
+		goto cleanup;
+
+	if (cg_read_key_long(memcg, "memory.events", "oom ") != 1)
+		goto cleanup;
+
+	if (cg_read_key_long(memcg, "memory.events", "oom_kill ") != 1)
+		goto cleanup;
+
+	if (cg_run(memcg, alloc_anon_50M_check_swap, (void *)MB(30)))
+		goto cleanup;
+
+	max = cg_read_key_long(memcg, "memory.events", "max ");
+	if (max <= 0)
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	cg_destroy(memcg);
+	free(memcg);
+
+	return ret;
+}
+
+/*
+ * This test disables swapping and tries to allocate anonymous memory
+ * up to OOM. Then it checks for oom and oom_kill events in
+ * memory.events.
+ */
+static int test_memcg_oom_events(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *memcg;
+
+	memcg = cg_name(root, "memcg_test");
+	if (!memcg)
+		goto cleanup;
+
+	if (cg_create(memcg))
+		goto cleanup;
+
+	if (cg_write(memcg, "memory.max", "30M"))
+		goto cleanup;
+
+	if (cg_write(memcg, "memory.swap.max", "0"))
+		goto cleanup;
+
+	if (!cg_run(memcg, alloc_anon, (void *)MB(100)))
+		goto cleanup;
+
+	if (cg_read_strcmp(memcg, "cgroup.procs", ""))
+		goto cleanup;
+
+	if (cg_read_key_long(memcg, "memory.events", "oom ") != 1)
+		goto cleanup;
+
+	if (cg_read_key_long(memcg, "memory.events", "oom_kill ") != 1)
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	cg_destroy(memcg);
+	free(memcg);
+
+	return ret;
+}
+
+struct tcp_server_args {
+	unsigned short port;
+	int ctl[2];
+};
+
+static int tcp_server(const char *cgroup, void *arg)
+{
+	struct tcp_server_args *srv_args = arg;
+	struct sockaddr_in6 saddr = { 0 };
+	socklen_t slen = sizeof(saddr);
+	int sk, client_sk, ctl_fd, yes = 1, ret = -1;
+
+	close(srv_args->ctl[0]);
+	ctl_fd = srv_args->ctl[1];
+
+	saddr.sin6_family = AF_INET6;
+	saddr.sin6_addr = in6addr_any;
+	saddr.sin6_port = htons(srv_args->port);
+
+	sk = socket(AF_INET6, SOCK_STREAM, 0);
+	if (sk < 0)
+		return ret;
+
+	if (setsockopt(sk, SOL_SOCKET, SO_REUSEADDR, &yes, sizeof(yes)) < 0)
+		goto cleanup;
+
+	if (bind(sk, (struct sockaddr *)&saddr, slen)) {
+		write(ctl_fd, &errno, sizeof(errno));
+		goto cleanup;
+	}
+
+	if (listen(sk, 1))
+		goto cleanup;
+
+	ret = 0;
+	if (write(ctl_fd, &ret, sizeof(ret)) != sizeof(ret)) {
+		ret = -1;
+		goto cleanup;
+	}
+
+	client_sk = accept(sk, NULL, NULL);
+	if (client_sk < 0)
+		goto cleanup;
+
+	ret = -1;
+	for (;;) {
+		uint8_t buf[0x100000];
+
+		if (write(client_sk, buf, sizeof(buf)) <= 0) {
+			if (errno == ECONNRESET)
+				ret = 0;
+			break;
+		}
+	}
+
+	close(client_sk);
+
+cleanup:
+	close(sk);
+	return ret;
+}
+
+static int tcp_client(const char *cgroup, unsigned short port)
+{
+	const char server[] = "localhost";
+	struct addrinfo *ai;
+	char servport[6];
+	int retries = 0x10; /* nice round number */
+	int sk, ret;
+
+	snprintf(servport, sizeof(servport), "%hd", port);
+	ret = getaddrinfo(server, servport, NULL, &ai);
+	if (ret)
+		return ret;
+
+	sk = socket(ai->ai_family, ai->ai_socktype, ai->ai_protocol);
+	if (sk < 0)
+		goto free_ainfo;
+
+	ret = connect(sk, ai->ai_addr, ai->ai_addrlen);
+	if (ret < 0)
+		goto close_sk;
+
+	ret = KSFT_FAIL;
+	while (retries--) {
+		uint8_t buf[0x100000];
+		long current, sock;
+
+		if (read(sk, buf, sizeof(buf)) <= 0)
+			goto close_sk;
+
+		current = cg_read_long(cgroup, "memory.current");
+		sock = cg_read_key_long(cgroup, "memory.stat", "sock ");
+
+		if (current < 0 || sock < 0)
+			goto close_sk;
+
+		if (current < sock)
+			goto close_sk;
+
+		if (values_close(current, sock, 10)) {
+			ret = KSFT_PASS;
+			break;
+		}
+	}
+
+close_sk:
+	close(sk);
+free_ainfo:
+	freeaddrinfo(ai);
+	return ret;
+}
+
+/*
+ * This test checks socket memory accounting.
+ * The test forks a TCP server listens on a random port between 1000
+ * and 61000. Once it gets a client connection, it starts writing to
+ * its socket.
+ * The TCP client interleaves reads from the socket with check whether
+ * memory.current and memory.stat.sock are similar.
+ */
+static int test_memcg_sock(const char *root)
+{
+	int bind_retries = 5, ret = KSFT_FAIL, pid, err;
+	unsigned short port;
+	char *memcg;
+
+	memcg = cg_name(root, "memcg_test");
+	if (!memcg)
+		goto cleanup;
+
+	if (cg_create(memcg))
+		goto cleanup;
+
+	while (bind_retries--) {
+		struct tcp_server_args args;
+
+		if (pipe(args.ctl))
+			goto cleanup;
+
+		port = args.port = 1000 + rand() % 60000;
+
+		pid = cg_run_nowait(memcg, tcp_server, &args);
+		if (pid < 0)
+			goto cleanup;
+
+		close(args.ctl[1]);
+		if (read(args.ctl[0], &err, sizeof(err)) != sizeof(err))
+			goto cleanup;
+		close(args.ctl[0]);
+
+		if (!err)
+			break;
+		if (err != EADDRINUSE)
+			goto cleanup;
+
+		waitpid(pid, NULL, 0);
+	}
+
+	if (err == EADDRINUSE) {
+		ret = KSFT_SKIP;
+		goto cleanup;
+	}
+
+	if (tcp_client(memcg, port) != KSFT_PASS)
+		goto cleanup;
+
+	waitpid(pid, &err, 0);
+	if (WEXITSTATUS(err))
+		goto cleanup;
+
+	if (cg_read_long(memcg, "memory.current") < 0)
+		goto cleanup;
+
+	if (cg_read_key_long(memcg, "memory.stat", "sock "))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	cg_destroy(memcg);
+	free(memcg);
+
+	return ret;
+}
+
+/*
+ * This test disables swapping and tries to allocate anonymous memory
+ * up to OOM with memory.group.oom set. Then it checks that all
+ * processes in the leaf (but not the parent) were killed.
+ */
+static int test_memcg_oom_group_leaf_events(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent, *child;
+
+	parent = cg_name(root, "memcg_test_0");
+	child = cg_name(root, "memcg_test_0/memcg_test_1");
+
+	if (!parent || !child)
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_create(child))
+		goto cleanup;
+
+	if (cg_write(parent, "cgroup.subtree_control", "+memory"))
+		goto cleanup;
+
+	if (cg_write(child, "memory.max", "50M"))
+		goto cleanup;
+
+	if (cg_write(child, "memory.swap.max", "0"))
+		goto cleanup;
+
+	if (cg_write(child, "memory.oom.group", "1"))
+		goto cleanup;
+
+	cg_run_nowait(parent, alloc_anon_noexit, (void *) MB(60));
+	cg_run_nowait(child, alloc_anon_noexit, (void *) MB(1));
+	cg_run_nowait(child, alloc_anon_noexit, (void *) MB(1));
+	if (!cg_run(child, alloc_anon, (void *)MB(100)))
+		goto cleanup;
+
+	if (cg_test_proc_killed(child))
+		goto cleanup;
+
+	if (cg_read_key_long(child, "memory.events", "oom_kill ") <= 0)
+		goto cleanup;
+
+	if (cg_read_key_long(parent, "memory.events", "oom_kill ") != 0)
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (child)
+		cg_destroy(child);
+	if (parent)
+		cg_destroy(parent);
+	free(child);
+	free(parent);
+
+	return ret;
+}
+
+/*
+ * This test disables swapping and tries to allocate anonymous memory
+ * up to OOM with memory.group.oom set. Then it checks that all
+ * processes in the parent and leaf were killed.
+ */
+static int test_memcg_oom_group_parent_events(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *parent, *child;
+
+	parent = cg_name(root, "memcg_test_0");
+	child = cg_name(root, "memcg_test_0/memcg_test_1");
+
+	if (!parent || !child)
+		goto cleanup;
+
+	if (cg_create(parent))
+		goto cleanup;
+
+	if (cg_create(child))
+		goto cleanup;
+
+	if (cg_write(parent, "memory.max", "80M"))
+		goto cleanup;
+
+	if (cg_write(parent, "memory.swap.max", "0"))
+		goto cleanup;
+
+	if (cg_write(parent, "memory.oom.group", "1"))
+		goto cleanup;
+
+	cg_run_nowait(parent, alloc_anon_noexit, (void *) MB(60));
+	cg_run_nowait(child, alloc_anon_noexit, (void *) MB(1));
+	cg_run_nowait(child, alloc_anon_noexit, (void *) MB(1));
+
+	if (!cg_run(child, alloc_anon, (void *)MB(100)))
+		goto cleanup;
+
+	if (cg_test_proc_killed(child))
+		goto cleanup;
+	if (cg_test_proc_killed(parent))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (child)
+		cg_destroy(child);
+	if (parent)
+		cg_destroy(parent);
+	free(child);
+	free(parent);
+
+	return ret;
+}
+
+/*
+ * This test disables swapping and tries to allocate anonymous memory
+ * up to OOM with memory.group.oom set. Then it checks that all
+ * processes were killed except those set with OOM_SCORE_ADJ_MIN
+ */
+static int test_memcg_oom_group_score_events(const char *root)
+{
+	int ret = KSFT_FAIL;
+	char *memcg;
+	int safe_pid;
+
+	memcg = cg_name(root, "memcg_test_0");
+
+	if (!memcg)
+		goto cleanup;
+
+	if (cg_create(memcg))
+		goto cleanup;
+
+	if (cg_write(memcg, "memory.max", "50M"))
+		goto cleanup;
+
+	if (cg_write(memcg, "memory.swap.max", "0"))
+		goto cleanup;
+
+	if (cg_write(memcg, "memory.oom.group", "1"))
+		goto cleanup;
+
+	safe_pid = cg_run_nowait(memcg, alloc_anon_noexit, (void *) MB(1));
+	if (set_oom_adj_score(safe_pid, OOM_SCORE_ADJ_MIN))
+		goto cleanup;
+
+	cg_run_nowait(memcg, alloc_anon_noexit, (void *) MB(1));
+	if (!cg_run(memcg, alloc_anon, (void *)MB(100)))
+		goto cleanup;
+
+	if (cg_read_key_long(memcg, "memory.events", "oom_kill ") != 3)
+		goto cleanup;
+
+	if (kill(safe_pid, SIGKILL))
+		goto cleanup;
+
+	ret = KSFT_PASS;
+
+cleanup:
+	if (memcg)
+		cg_destroy(memcg);
+	free(memcg);
+
+	return ret;
+}
+
+
+#define T(x) { x, #x }
+struct memcg_test {
+	int (*fn)(const char *root);
+	const char *name;
+} tests[] = {
+	T(test_memcg_subtree_control),
+	T(test_memcg_current),
+	T(test_memcg_min),
+	T(test_memcg_low),
+	T(test_memcg_high),
+	T(test_memcg_max),
+	T(test_memcg_oom_events),
+	T(test_memcg_swap_max),
+	T(test_memcg_sock),
+	T(test_memcg_oom_group_leaf_events),
+	T(test_memcg_oom_group_parent_events),
+	T(test_memcg_oom_group_score_events),
+};
+#undef T
+
+int main(int argc, char **argv)
+{
+	char root[PATH_MAX];
+	int i, ret = EXIT_SUCCESS;
+
+	if (cg_find_unified_root(root, sizeof(root)))
+		ksft_exit_skip("cgroup v2 isn't mounted\n");
+
+	/*
+	 * Check that memory controller is available:
+	 * memory is listed in cgroup.controllers
+	 */
+	if (cg_read_strstr(root, "cgroup.controllers", "memory"))
+		ksft_exit_skip("memory controller isn't available\n");
+
+	if (cg_read_strstr(root, "cgroup.subtree_control", "memory"))
+		if (cg_write(root, "cgroup.subtree_control", "+memory"))
+			ksft_exit_skip("Failed to set memory controller\n");
+
+	for (i = 0; i < ARRAY_SIZE(tests); i++) {
+		switch (tests[i].fn(root)) {
+		case KSFT_PASS:
+			ksft_test_result_pass("%s\n", tests[i].name);
+			break;
+		case KSFT_SKIP:
+			ksft_test_result_skip("%s\n", tests[i].name);
+			break;
+		default:
+			ret = EXIT_FAILURE;
+			ksft_test_result_fail("%s\n", tests[i].name);
+			break;
+		}
+	}
+
+	return ret;
+}
diff --git a/tools/testing/selftests/cgroup/test_stress.sh b/tools/testing/selftests/cgroup/test_stress.sh
new file mode 100755
index 000000000..3c9c4554d
--- /dev/null
+++ b/tools/testing/selftests/cgroup/test_stress.sh
@@ -0,0 +1,4 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+./with_stress.sh -s subsys -s fork ${OUTPUT:-.}/test_core
diff --git a/tools/testing/selftests/cgroup/with_stress.sh b/tools/testing/selftests/cgroup/with_stress.sh
new file mode 100755
index 000000000..e28c35008
--- /dev/null
+++ b/tools/testing/selftests/cgroup/with_stress.sh
@@ -0,0 +1,101 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+
+# Kselftest framework requirement - SKIP code is 4.
+ksft_skip=4
+
+stress_fork()
+{
+	while true ; do
+		/usr/bin/true
+		sleep 0.01
+	done
+}
+
+stress_subsys()
+{
+	local verb=+
+	while true ; do
+		echo $verb$subsys_ctrl >$sysfs/cgroup.subtree_control
+		[ $verb = "+" ] && verb=- || verb=+
+		# incommensurable period with other stresses
+		sleep 0.011
+	done
+}
+
+init_and_check()
+{
+	sysfs=`mount -t cgroup2 | head -1 | awk '{ print $3 }'`
+	if [ ! -d "$sysfs" ]; then
+		echo "Skipping: cgroup2 is not mounted" >&2
+		exit $ksft_skip
+	fi
+
+	if ! echo +$subsys_ctrl >$sysfs/cgroup.subtree_control ; then
+		echo "Skipping: cannot enable $subsys_ctrl in $sysfs" >&2
+		exit $ksft_skip
+	fi
+
+	if ! echo -$subsys_ctrl >$sysfs/cgroup.subtree_control ; then
+		echo "Skipping: cannot disable $subsys_ctrl in $sysfs" >&2
+		exit $ksft_skip
+	fi
+}
+
+declare -a stresses
+declare -a stress_pids
+duration=5
+rc=0
+subsys_ctrl=cpuset
+sysfs=
+
+while getopts c:d:hs: opt; do
+	case $opt in
+	c)
+		subsys_ctrl=$OPTARG
+		;;
+	d)
+		duration=$OPTARG
+		;;
+	h)
+		echo "Usage $0 [ -s stress ] ... [ -d duration ] [-c controller] cmd args .."
+		echo -e "\t default duration $duration seconds"
+		echo -e "\t default controller $subsys_ctrl"
+		exit
+		;;
+	s)
+		func=stress_$OPTARG
+		if [ "x$(type -t $func)" != "xfunction" ] ; then
+			echo "Unknown stress $OPTARG"
+			exit 1
+		fi
+		stresses+=($func)
+		;;
+	esac
+done
+shift $((OPTIND - 1))
+
+init_and_check
+
+for s in ${stresses[*]} ; do
+	$s &
+	stress_pids+=($!)
+done
+
+
+time=0
+start=$(date +%s)
+
+while [ $time -lt $duration ] ; do
+	$*
+	rc=$?
+	[ $rc -eq 0 ] || break
+	time=$(($(date +%s) - $start))
+done
+
+for pid in ${stress_pids[*]} ; do
+	kill -SIGTERM $pid
+	wait $pid
+done
+
+exit $rc