selftests/bpf: Add selftests for cpumask iter

Add selftests for the newly added cpumask iter.
- cpumask_iter_success
  - The number of CPUs should be expected when iterating over the cpumask
  - percpu data extracted from the percpu struct should be expected
  - It can work in both non-sleepable and sleepable prog
  - RCU lock is only required by bpf_iter_cpumask_new()
  - It is fine without calling bpf_iter_cpumask_next()

- cpumask_iter_failure
  - RCU lock is required in sleepable prog
  - The cpumask to be iterated over can't be NULL
  - bpf_iter_cpumask_destroy() is required after calling
    bpf_iter_cpumask_new()
  - bpf_iter_cpumask_destroy() can only destroy an initilialized iter
  - bpf_iter_cpumask_next() must use an initilialized iter

The result as follows,

  torvalds#64/37   cpumask/test_cpumask_iter:OK
  torvalds#64/38   cpumask/test_cpumask_iter_sleepable:OK
  torvalds#64/39   cpumask/test_cpumask_iter_sleepable:OK
  torvalds#64/40   cpumask/test_cpumask_iter_next_no_rcu:OK
  torvalds#64/41   cpumask/test_cpumask_iter_no_next:OK
  torvalds#64/42   cpumask/test_cpumask_iter:OK
  torvalds#64/43   cpumask/test_cpumask_iter_no_rcu:OK
  torvalds#64/44   cpumask/test_cpumask_iter_no_destroy:OK
  torvalds#64/45   cpumask/test_cpumask_iter_null_pointer:OK
  torvalds#64/46   cpumask/test_cpumask_iter_next_uninit:OK
  torvalds#64/47   cpumask/test_cpumask_iter_destroy_uninit:OK
  torvalds#64      cpumask:OK

Signed-off-by: Yafang Shao <laoar.shao@gmail.com>

tools/testing/selftests/bpf/config

@@ -78,6 +78,7 @@ CONFIG_NF_CONNTRACK_MARK=y
 CONFIG_NF_DEFRAG_IPV4=y
 CONFIG_NF_DEFRAG_IPV6=y
 CONFIG_NF_NAT=y
+CONFIG_PSI=y
 CONFIG_RC_CORE=y
 CONFIG_SECURITY=y
 CONFIG_SECURITYFS=y

tools/testing/selftests/bpf/prog_tests/cpumask.c

@@ -1,9 +1,15 @@
 // SPDX-License-Identifier: GPL-2.0
 /* Copyright (c) 2023 Meta Platforms, Inc. and affiliates. */
 
+#define _GNU_SOURCE
+#include <sched.h>
+
 #include <test_progs.h>
 #include "cpumask_failure.skel.h"
 #include "cpumask_success.skel.h"
+#include "cpumask_iter_success.skel.h"
+#include "cpumask_iter_failure.skel.h"
+#include "cgroup_helpers.h"
 
 static const char * const cpumask_success_testcases[] = {
 	"test_alloc_free_cpumask",
@@ -61,6 +67,142 @@ static void verify_success(const char *prog_name)
 	cpumask_success__destroy(skel);
 }
 
+static const char * const cpumask_iter_success_testcases[] = {
+	"test_cpumask_iter",
+	"test_cpumask_iter_sleepable",
+};
+
+static int read_percpu_data(struct bpf_link *link, int nr_cpu_exp, int nr_running_exp)
+{
+	int iter_fd, len, item, nr_running, psi_running, nr_cpus, err = -1;
+	char buf[128];
+	size_t left;
+	char *p;
+
+	iter_fd = bpf_iter_create(bpf_link__fd(link));
+	if (!ASSERT_GE(iter_fd, 0, "iter_fd"))
+		return -1;
+
+	memset(buf, 0, sizeof(buf));
+	left = ARRAY_SIZE(buf);
+	p = buf;
+	while ((len = read(iter_fd, p, left)) > 0) {
+		p += len;
+		left -= len;
+	}
+
+	item = sscanf(buf, "nr_running %u nr_cpus %u psi_running %u\n",
+		      &nr_running, &nr_cpus, &psi_running);
+	if (!ASSERT_EQ(item, 3, "seq_format"))
+		goto out;
+	if (!ASSERT_EQ(nr_cpus, nr_cpu_exp, "nr_cpus"))
+		goto out;
+	if (!ASSERT_GE(nr_running, nr_running_exp, "nr_running"))
+		goto out;
+	if (!ASSERT_GE(psi_running, nr_running_exp, "psi_running"))
+		goto out;
+
+	err = 0;
+out:
+	close(iter_fd);
+	return err;
+}
+
+static void verify_iter_success(const char *prog_name)
+{
+	DECLARE_LIBBPF_OPTS(bpf_iter_attach_opts, opts);
+	int cgrp_fd, nr_cpus, err, i, chosen = 0;
+	struct cpumask_iter_success *skel;
+	union bpf_iter_link_info linfo;
+	struct bpf_program *prog;
+	struct bpf_link *link;
+	cpu_set_t set;
+
+	if (setup_cgroup_environment())
+		return;
+
+	/* Utilize the cgroup iter */
+	cgrp_fd = get_root_cgroup();
+	if (!ASSERT_GE(cgrp_fd, 0, "create_cgrp"))
+		goto cleanup;
+
+	skel = cpumask_iter_success__open();
+	if (!ASSERT_OK_PTR(skel, "cpumask_iter_success__open"))
+		goto close_fd;
+
+	skel->bss->pid = getpid();
+	nr_cpus = libbpf_num_possible_cpus();
+
+	err = cpumask_iter_success__load(skel);
+	if (!ASSERT_OK(err, "cpumask_iter_success__load"))
+		goto destroy;
+
+	prog = bpf_object__find_program_by_name(skel->obj, prog_name);
+	if (!ASSERT_OK_PTR(prog, "bpf_object__find_program_by_name"))
+		goto destroy;
+
+	memset(&linfo, 0, sizeof(linfo));
+	linfo.cgroup.cgroup_fd = cgrp_fd;
+	linfo.cgroup.order = BPF_CGROUP_ITER_SELF_ONLY;
+	opts.link_info = &linfo;
+	opts.link_info_len = sizeof(linfo);
+	link = bpf_program__attach_iter(prog, &opts);
+	if (!ASSERT_OK_PTR(link, "bpf_program__attach"))
+		goto destroy;
+
+	/* Case 1): Enable all possible CPUs */
+	CPU_ZERO(&set);
+	for (i = 0; i < nr_cpus; i++)
+		CPU_SET(i, &set);
+	err = sched_setaffinity(skel->bss->pid, sizeof(set), &set);
+	if (!ASSERT_OK(err, "setaffinity_all_cpus"))
+		goto free_link;
+	err = read_percpu_data(link, nr_cpus, 1);
+	if (!ASSERT_OK(err, "read_percpu_data"))
+		goto free_link;
+	if (!ASSERT_OK(skel->bss->err, "null_rq"))
+		goto free_link;
+
+	/* Case 2): CPU0 only */
+	CPU_ZERO(&set);
+	CPU_SET(0, &set);
+	err = sched_setaffinity(skel->bss->pid, sizeof(set), &set);
+	if (!ASSERT_OK(err, "setaffinity_cpu0"))
+		goto free_link;
+	err = read_percpu_data(link, 1, 1);
+	if (!ASSERT_OK(err, "read_percpu_data"))
+		goto free_link;
+	if (!ASSERT_OK(skel->bss->err, "null_rq_psi"))
+		goto free_link;
+
+	/* Case 3): Partial CPUs */
+	CPU_ZERO(&set);
+	for (i = 0; i < nr_cpus; i++) {
+		if (i < 4 && i & 0x1)
+			continue;
+		if (i > 8 && i & 0x2)
+			continue;
+		CPU_SET(i, &set);
+		chosen++;
+	}
+	err = sched_setaffinity(skel->bss->pid, sizeof(set), &set);
+	if (!ASSERT_OK(err, "setaffinity_partial_cpus"))
+		goto free_link;
+	err = read_percpu_data(link, chosen, 1);
+	if (!ASSERT_OK(err, "read_percpu_data"))
+		goto free_link;
+	ASSERT_OK(skel->bss->err, "null_rq_psi");
+
+free_link:
+	bpf_link__destroy(link);
+destroy:
+	cpumask_iter_success__destroy(skel);
+close_fd:
+	close(cgrp_fd);
+cleanup:
+	cleanup_cgroup_environment();
+}
+
 void test_cpumask(void)
 {
 	int i;
@@ -74,4 +216,14 @@ void test_cpumask(void)
 
 	RUN_TESTS(cpumask_success);
 	RUN_TESTS(cpumask_failure);
+
+	for (i = 0; i < ARRAY_SIZE(cpumask_iter_success_testcases); i++) {
+		if (!test__start_subtest(cpumask_iter_success_testcases[i]))
+			continue;
+
+		verify_iter_success(cpumask_iter_success_testcases[i]);
+	}
+
+	RUN_TESTS(cpumask_iter_success);
+	RUN_TESTS(cpumask_iter_failure);
 }

tools/testing/selftests/bpf/progs/cpumask_common.h

@@ -55,6 +55,9 @@ void bpf_cpumask_copy(struct bpf_cpumask *dst, const struct cpumask *src) __ksym
 u32 bpf_cpumask_any_distribute(const struct cpumask *src) __ksym;
 u32 bpf_cpumask_any_and_distribute(const struct cpumask *src1, const struct cpumask *src2) __ksym;
 u32 bpf_cpumask_weight(const struct cpumask *cpumask) __ksym;
+int bpf_iter_cpumask_new(struct bpf_iter_cpumask *it, const struct cpumask *mask) __ksym;
+int *bpf_iter_cpumask_next(struct bpf_iter_cpumask *it) __ksym;
+void bpf_iter_cpumask_destroy(struct bpf_iter_cpumask *it) __ksym;
 
 void bpf_rcu_read_lock(void) __ksym;
 void bpf_rcu_read_unlock(void) __ksym;

tools/testing/selftests/bpf/progs/cpumask_iter_failure.c

@@ -0,0 +1,99 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (c) 2024 Yafang Shao <laoar.shao@gmail.com> */
+
+#include "vmlinux.h"
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
+
+#include "bpf_misc.h"
+#include "task_kfunc_common.h"
+#include "cpumask_common.h"
+
+char _license[] SEC("license") = "GPL";
+
+SEC("iter.s/cgroup")
+__failure __msg("R2 must be a rcu pointer")
+int BPF_PROG(test_cpumask_iter_no_rcu, struct bpf_iter_meta *meta, struct cgroup *cgrp)
+{
+	struct task_struct *p;
+	int *cpu;
+
+	p = bpf_task_from_pid(1);
+	if (!p)
+		return 1;
+
+	bpf_for_each(cpumask, cpu, p->cpus_ptr) {
+	}
+	bpf_task_release(p);
+	return 0;
+}
+
+SEC("iter/cgroup")
+__failure __msg("Possibly NULL pointer passed to trusted arg1")
+int BPF_PROG(test_cpumask_iter_null_pointer, struct bpf_iter_meta *meta, struct cgroup *cgrp)
+{
+	struct cpumask *mask = NULL;
+	int *cpu;
+
+	bpf_for_each(cpumask, cpu, mask) {
+	}
+	return 0;
+}
+
+SEC("iter.s/cgroup")
+__failure __msg("Unreleased reference id=3 alloc_insn=10")
+int BPF_PROG(test_cpumask_iter_no_destroy, struct bpf_iter_meta *meta, struct cgroup *cgrp)
+{
+	struct bpf_iter_cpumask it;
+	struct task_struct *p;
+
+	p = bpf_task_from_pid(1);
+	if (!p)
+		return 1;
+
+	bpf_rcu_read_lock();
+	bpf_iter_cpumask_new(&it, p->cpus_ptr);
+	bpf_rcu_read_unlock();
+
+	bpf_iter_cpumask_next(&it);
+	bpf_task_release(p);
+	return 0;
+}
+
+SEC("iter/cgroup")
+__failure __msg("expected an initialized iter_cpumask as arg #1")
+int BPF_PROG(test_cpumask_iter_next_uninit, struct bpf_iter_meta *meta, struct cgroup *cgrp)
+{
+	struct bpf_iter_cpumask *it = NULL;
+
+	bpf_iter_cpumask_next(it);
+	return 0;
+}
+
+SEC("iter/cgroup")
+__failure __msg("expected an initialized iter_cpumask as arg #1")
+int BPF_PROG(test_cpumask_iter_next_uninit2, struct bpf_iter_meta *meta, struct cgroup *cgrp)
+{
+	struct bpf_iter_cpumask it = {};
+
+	bpf_iter_cpumask_next(&it);
+	return 0;
+}
+
+SEC("iter/cgroup")
+__failure __msg("expected an initialized iter_cpumask as arg #1")
+int BPF_PROG(test_cpumask_iter_destroy_uninit, struct bpf_iter_meta *meta, struct cgroup *cgrp)
+{
+	struct bpf_iter_cpumask_kern it = {.cpu = -1};
+	struct bpf_cpumask *mask;
+
+	mask = bpf_cpumask_create();
+	if (!mask)
+		return 1;
+
+	bpf_cpumask_setall(mask);
+	it.mask = &mask->cpumask;
+	bpf_iter_cpumask_destroy((struct bpf_iter_cpumask *)&it);
+	bpf_cpumask_release(mask);
+	return 0;
+}