Merge tag 'cgroup-for-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup

Pull cgroup updates from Tejun Heo:

 - Allow css_rstat_updated() in NMI context to enable memory accounting
   for allocations in NMI context.

 - /proc/cgroups doesn't contain useful information for cgroup2 and was
   updated to only show v1 controllers. This unfortunately broke
   something in the wild. Add an option to bring back the old behavior
   to ease transition.

 - selftest updates and other cleanups.

* tag 'cgroup-for-6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
  cgroup: Add compatibility option for content of /proc/cgroups
  selftests/cgroup: fix cpu.max tests
  cgroup: llist: avoid memory tears for llist_node
  selftests: cgroup: Fix missing newline in test_zswap_writeback_one
  selftests: cgroup: Allow longer timeout for kmem_dead_cgroups cleanup
  memcg: cgroup: call css_rstat_updated irrespective of in_nmi()
  cgroup: remove per-cpu per-subsystem locks
  cgroup: make css_rstat_updated nmi safe
  cgroup: support to enable nmi-safe css_rstat_updated
  selftests: cgroup: Fix compilation on pre-cgroupns kernels
  selftests: cgroup: Optionally set up v1 environment
  selftests: cgroup: Add support for named v1 hierarchies in test_core
  selftests: cgroup_util: Add helpers for testing named v1 hierarchies
  Documentation: cgroup: add section explaining controller availability
  cgroup: Drop sock_cgroup_classid() dummy implementation

6 files changed, 130 insertions, 33 deletions

diff --git a/tools/testing/selftests/cgroup/lib/cgroup_util.c b/tools/testing/selftests/cgroup/lib/cgroup_util.c
index 8832f3d1cb61..0e89fcff4d05 100644
--- a/tools/testing/selftests/cgroup/lib/cgroup_util.c
+++ b/tools/testing/selftests/cgroup/lib/cgroup_util.c
@@ -19,6 +19,8 @@
 #include "cgroup_util.h"
 #include "../../clone3/clone3_selftests.h"
 
+bool cg_test_v1_named;
+
 /* Returns read len on success, or -errno on failure. */
 ssize_t read_text(const char *path, char *buf, size_t max_len)
 {
@@ -361,7 +363,7 @@ int cg_enter_current(const char *cgroup)
 
 int cg_enter_current_thread(const char *cgroup)
 {
-	return cg_write(cgroup, "cgroup.threads", "0");
+	return cg_write(cgroup, CG_THREADS_FILE, "0");
 }
 
 int cg_run(const char *cgroup,
diff --git a/tools/testing/selftests/cgroup/lib/include/cgroup_util.h b/tools/testing/selftests/cgroup/lib/include/cgroup_util.h
index adb2bc193183..c69cab66254b 100644
--- a/tools/testing/selftests/cgroup/lib/include/cgroup_util.h
+++ b/tools/testing/selftests/cgroup/lib/include/cgroup_util.h
@@ -13,6 +13,10 @@
 
 #define TEST_UID	65534 /* usually nobody, any !root is fine */
 
+#define CG_THREADS_FILE (!cg_test_v1_named ? "cgroup.threads" : "tasks")
+#define CG_NAMED_NAME "selftest"
+#define CG_PATH_FORMAT (!cg_test_v1_named ? "0::%s" : (":name=" CG_NAMED_NAME ":%s"))
+
 /*
  * Checks if two given values differ by less than err% of their sum.
  */
@@ -65,3 +69,4 @@ extern int dirfd_open_opath(const char *dir);
 extern int cg_prepare_for_wait(const char *cgroup);
 extern int memcg_prepare_for_wait(const char *cgroup);
 extern int cg_wait_for(int fd);
+extern bool cg_test_v1_named;
diff --git a/tools/testing/selftests/cgroup/test_core.c b/tools/testing/selftests/cgroup/test_core.c
index a5672a91d273..a360e2eb2eef 100644
--- a/tools/testing/selftests/cgroup/test_core.c
+++ b/tools/testing/selftests/cgroup/test_core.c
@@ -5,6 +5,8 @@
 #include <linux/sched.h>
 #include <sys/types.h>
 #include <sys/mman.h>
+#include <sys/mount.h>
+#include <sys/stat.h>
 #include <sys/wait.h>
 #include <unistd.h>
 #include <fcntl.h>
@@ -19,6 +21,9 @@
 #include "cgroup_util.h"
 
 static bool nsdelegate;
+#ifndef CLONE_NEWCGROUP
+#define CLONE_NEWCGROUP 0
+#endif
 
 static int touch_anon(char *buf, size_t size)
 {
@@ -148,6 +153,9 @@ static int test_cgcore_populated(const char *root)
 	int cgroup_fd = -EBADF;
 	pid_t pid;
 
+	if (cg_test_v1_named)
+		return KSFT_SKIP;
+
 	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");
@@ -277,6 +285,9 @@ static int test_cgcore_invalid_domain(const char *root)
 	int ret = KSFT_FAIL;
 	char *grandparent = NULL, *parent = NULL, *child = NULL;
 
+	if (cg_test_v1_named)
+		return KSFT_SKIP;
+
 	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");
@@ -339,6 +350,9 @@ static int test_cgcore_parent_becomes_threaded(const char *root)
 	int ret = KSFT_FAIL;
 	char *parent = NULL, *child = NULL;
 
+	if (cg_test_v1_named)
+		return KSFT_SKIP;
+
 	parent = cg_name(root, "cg_test_parent");
 	child = cg_name(root, "cg_test_parent/cg_test_child");
 	if (!parent || !child)
@@ -378,7 +392,8 @@ static int test_cgcore_no_internal_process_constraint_on_threads(const char *roo
 	int ret = KSFT_FAIL;
 	char *parent = NULL, *child = NULL;
 
-	if (cg_read_strstr(root, "cgroup.controllers", "cpu") ||
+	if (cg_test_v1_named ||
+	    cg_read_strstr(root, "cgroup.controllers", "cpu") ||
 	    cg_write(root, "cgroup.subtree_control", "+cpu")) {
 		ret = KSFT_SKIP;
 		goto cleanup;
@@ -430,6 +445,9 @@ static int test_cgcore_top_down_constraint_enable(const char *root)
 	int ret = KSFT_FAIL;
 	char *parent = NULL, *child = NULL;
 
+	if (cg_test_v1_named)
+		return KSFT_SKIP;
+
 	parent = cg_name(root, "cg_test_parent");
 	child = cg_name(root, "cg_test_parent/cg_test_child");
 	if (!parent || !child)
@@ -465,6 +483,9 @@ static int test_cgcore_top_down_constraint_disable(const char *root)
 	int ret = KSFT_FAIL;
 	char *parent = NULL, *child = NULL;
 
+	if (cg_test_v1_named)
+		return KSFT_SKIP;
+
 	parent = cg_name(root, "cg_test_parent");
 	child = cg_name(root, "cg_test_parent/cg_test_child");
 	if (!parent || !child)
@@ -506,6 +527,9 @@ static int test_cgcore_internal_process_constraint(const char *root)
 	int ret = KSFT_FAIL;
 	char *parent = NULL, *child = NULL;
 
+	if (cg_test_v1_named)
+		return KSFT_SKIP;
+
 	parent = cg_name(root, "cg_test_parent");
 	child = cg_name(root, "cg_test_parent/cg_test_child");
 	if (!parent || !child)
@@ -573,7 +597,7 @@ static int test_cgcore_proc_migration(const char *root)
 	}
 
 	cg_enter_current(dst);
-	if (cg_read_lc(dst, "cgroup.threads") != n_threads + 1)
+	if (cg_read_lc(dst, CG_THREADS_FILE) != n_threads + 1)
 		goto cleanup;
 
 	ret = KSFT_PASS;
@@ -605,7 +629,7 @@ static void *migrating_thread_fn(void *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]));
+		snprintf(lines[g], sizeof(lines[g]), CG_PATH_FORMAT, grps[g] + strlen(grps[0]));
 
 	for (i = 0; i < n_iterations; ++i) {
 		cg_enter_current_thread(grps[(i % 2) + 1]);
@@ -642,10 +666,12 @@ static int test_cgcore_thread_migration(const char *root)
 	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_test_v1_named) {
+		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;
@@ -659,7 +685,7 @@ static int test_cgcore_thread_migration(const char *root)
 	if (retval)
 		goto cleanup;
 
-	snprintf(line, sizeof(line), "0::%s", grps[1] + strlen(grps[0]));
+	snprintf(line, sizeof(line), CG_PATH_FORMAT, grps[1] + strlen(grps[0]));
 	if (proc_read_strstr(0, 1, "cgroup", line))
 		goto cleanup;
 
@@ -842,6 +868,38 @@ cleanup:
 	return ret;
 }
 
+static int setup_named_v1_root(char *root, size_t len, const char *name)
+{
+	char options[PATH_MAX];
+	int r;
+
+	r = snprintf(root, len, "/mnt/cg_selftest");
+	if (r < 0)
+		return r;
+
+	r = snprintf(options, sizeof(options), "none,name=%s", name);
+	if (r < 0)
+		return r;
+
+	r = mkdir(root, 0755);
+	if (r < 0 && errno != EEXIST)
+		return r;
+
+	r = mount("none", root, "cgroup", 0, options);
+	if (r < 0)
+		return r;
+
+	return 0;
+}
+
+static void cleanup_named_v1_root(char *root)
+{
+	if (!cg_test_v1_named)
+		return;
+	umount(root);
+	rmdir(root);
+}
+
 #define T(x) { x, #x }
 struct corecg_test {
 	int (*fn)(const char *root);
@@ -867,13 +925,18 @@ int main(int argc, char *argv[])
 	char root[PATH_MAX];
 	int i, ret = EXIT_SUCCESS;
 
-	if (cg_find_unified_root(root, sizeof(root), &nsdelegate))
-		ksft_exit_skip("cgroup v2 isn't mounted\n");
+	if (cg_find_unified_root(root, sizeof(root), &nsdelegate)) {
+		if (setup_named_v1_root(root, sizeof(root), CG_NAMED_NAME))
+			ksft_exit_skip("cgroup v2 isn't mounted and could not setup named v1 hierarchy\n");
+		cg_test_v1_named = true;
+		goto post_v2_setup;
+	}
 
 	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");
 
+post_v2_setup:
 	for (i = 0; i < ARRAY_SIZE(tests); i++) {
 		switch (tests[i].fn(root)) {
 		case KSFT_PASS:
@@ -889,5 +952,6 @@ int main(int argc, char *argv[])
 		}
 	}
 
+	cleanup_named_v1_root(root);
 	return ret;
 }
diff --git a/tools/testing/selftests/cgroup/test_cpu.c b/tools/testing/selftests/cgroup/test_cpu.c
index a2b50af8e9ee..2a60e6c41940 100644
--- a/tools/testing/selftests/cgroup/test_cpu.c
+++ b/tools/testing/selftests/cgroup/test_cpu.c
@@ -2,6 +2,7 @@
 
 #define _GNU_SOURCE
 #include <linux/limits.h>
+#include <sys/param.h>
 #include <sys/sysinfo.h>
 #include <sys/wait.h>
 #include <errno.h>
@@ -645,10 +646,16 @@ test_cpucg_nested_weight_underprovisioned(const char *root)
 static int test_cpucg_max(const char *root)
 {
 	int ret = KSFT_FAIL;
-	long usage_usec, user_usec;
-	long usage_seconds = 1;
-	long expected_usage_usec = usage_seconds * USEC_PER_SEC;
+	long quota_usec = 1000;
+	long default_period_usec = 100000; /* cpu.max's default period */
+	long duration_seconds = 1;
+
+	long duration_usec = duration_seconds * USEC_PER_SEC;
+	long usage_usec, n_periods, remainder_usec, expected_usage_usec;
 	char *cpucg;
+	char quota_buf[32];
+
+	snprintf(quota_buf, sizeof(quota_buf), "%ld", quota_usec);
 
 	cpucg = cg_name(root, "cpucg_test");
 	if (!cpucg)
@@ -657,13 +664,13 @@ static int test_cpucg_max(const char *root)
 	if (cg_create(cpucg))
 		goto cleanup;
 
-	if (cg_write(cpucg, "cpu.max", "1000"))
+	if (cg_write(cpucg, "cpu.max", quota_buf))
 		goto cleanup;
 
 	struct cpu_hog_func_param param = {
 		.nprocs = 1,
 		.ts = {
-			.tv_sec = usage_seconds,
+			.tv_sec = duration_seconds,
 			.tv_nsec = 0,
 		},
 		.clock_type = CPU_HOG_CLOCK_WALL,
@@ -672,14 +679,19 @@ static int test_cpucg_max(const char *root)
 		goto cleanup;
 
 	usage_usec = cg_read_key_long(cpucg, "cpu.stat", "usage_usec");
-	user_usec = cg_read_key_long(cpucg, "cpu.stat", "user_usec");
-	if (user_usec <= 0)
+	if (usage_usec <= 0)
 		goto cleanup;
 
-	if (user_usec >= expected_usage_usec)
-		goto cleanup;
+	/*
+	 * The following calculation applies only since
+	 * the cpu hog is set to run as per wall-clock time
+	 */
+	n_periods = duration_usec / default_period_usec;
+	remainder_usec = duration_usec - n_periods * default_period_usec;
+	expected_usage_usec
+		= n_periods * quota_usec + MIN(remainder_usec, quota_usec);
 
-	if (values_close(usage_usec, expected_usage_usec, 95))
+	if (!values_close(usage_usec, expected_usage_usec, 10))
 		goto cleanup;
 
 	ret = KSFT_PASS;
@@ -698,10 +710,16 @@ cleanup:
 static int test_cpucg_max_nested(const char *root)
 {
 	int ret = KSFT_FAIL;
-	long usage_usec, user_usec;
-	long usage_seconds = 1;
-	long expected_usage_usec = usage_seconds * USEC_PER_SEC;
+	long quota_usec = 1000;
+	long default_period_usec = 100000; /* cpu.max's default period */
+	long duration_seconds = 1;
+
+	long duration_usec = duration_seconds * USEC_PER_SEC;
+	long usage_usec, n_periods, remainder_usec, expected_usage_usec;
 	char *parent, *child;
+	char quota_buf[32];
+
+	snprintf(quota_buf, sizeof(quota_buf), "%ld", quota_usec);
 
 	parent = cg_name(root, "cpucg_parent");
 	child = cg_name(parent, "cpucg_child");
@@ -717,13 +735,13 @@ static int test_cpucg_max_nested(const char *root)
 	if (cg_create(child))
 		goto cleanup;
 
-	if (cg_write(parent, "cpu.max", "1000"))
+	if (cg_write(parent, "cpu.max", quota_buf))
 		goto cleanup;
 
 	struct cpu_hog_func_param param = {
 		.nprocs = 1,
 		.ts = {
-			.tv_sec = usage_seconds,
+			.tv_sec = duration_seconds,
 			.tv_nsec = 0,
 		},
 		.clock_type = CPU_HOG_CLOCK_WALL,
@@ -732,14 +750,19 @@ static int test_cpucg_max_nested(const char *root)
 		goto cleanup;
 
 	usage_usec = cg_read_key_long(child, "cpu.stat", "usage_usec");
-	user_usec = cg_read_key_long(child, "cpu.stat", "user_usec");
-	if (user_usec <= 0)
+	if (usage_usec <= 0)
 		goto cleanup;
 
-	if (user_usec >= expected_usage_usec)
-		goto cleanup;
+	/*
+	 * The following calculation applies only since
+	 * the cpu hog is set to run as per wall-clock time
+	 */
+	n_periods = duration_usec / default_period_usec;
+	remainder_usec = duration_usec - n_periods * default_period_usec;
+	expected_usage_usec
+		= n_periods * quota_usec + MIN(remainder_usec, quota_usec);
 
-	if (values_close(usage_usec, expected_usage_usec, 95))
+	if (!values_close(usage_usec, expected_usage_usec, 10))
 		goto cleanup;
 
 	ret = KSFT_PASS;
diff --git a/tools/testing/selftests/cgroup/test_kmem.c b/tools/testing/selftests/cgroup/test_kmem.c
index 96693d8772be..63b3c9aad399 100644
--- a/tools/testing/selftests/cgroup/test_kmem.c
+++ b/tools/testing/selftests/cgroup/test_kmem.c
@@ -308,6 +308,7 @@ static int test_kmem_dead_cgroups(const char *root)
 	char *parent;
 	long dead;
 	int i;
+	int max_time = 20;
 
 	parent = cg_name(root, "kmem_dead_cgroups_test");
 	if (!parent)
@@ -322,7 +323,7 @@ static int test_kmem_dead_cgroups(const char *root)
 	if (cg_run_in_subcgroups(parent, alloc_dcache, (void *)100, 30))
 		goto cleanup;
 
-	for (i = 0; i < 5; i++) {
+	for (i = 0; i < max_time; i++) {
 		dead = cg_read_key_long(parent, "cgroup.stat",
 					"nr_dying_descendants ");
 		if (dead == 0) {
@@ -334,6 +335,8 @@ static int test_kmem_dead_cgroups(const char *root)
 		 * let's wait a bit and repeat.
 		 */
 		sleep(1);
+		if (i > 5)
+			printf("Waiting time longer than 5s; wait: %ds (dead: %ld)\n", i, dead);
 	}
 
 cleanup:
diff --git a/tools/testing/selftests/cgroup/test_zswap.c b/tools/testing/selftests/cgroup/test_zswap.c
index 40de679248b8..e1f578ca2841 100644
--- a/tools/testing/selftests/cgroup/test_zswap.c
+++ b/tools/testing/selftests/cgroup/test_zswap.c
@@ -338,7 +338,7 @@ static int test_zswap_writeback_one(const char *cgroup, bool wb)
 		return -1;
 
 	if (wb != !!zswpwb_after) {
-		ksft_print_msg("zswpwb_after is %ld while wb is %s",
+		ksft_print_msg("zswpwb_after is %ld while wb is %s\n",
 				zswpwb_after, wb ? "enabled" : "disabled");
 		return -1;
 	}