selftests/bpf: Fix some bugs in map_lookup_percpu_elem testcase

comments from Andrii Nakryiko, details in here:
https://lore.kernel.org/lkml/20220511093854.411-1-zhoufeng.zf@bytedance.com/T/

use /* */ instead of //
use libbpf_num_possible_cpus() instead of sysconf(_SC_NPROCESSORS_ONLN)
use 8 bytes for value size
fix memory leak
use ASSERT_EQ instead of ASSERT_OK
add bpf_loop to fetch values on each possible CPU

Fixes: ed7c1377 ("selftests/bpf: add test case for bpf_map_lookup_percpu_elem")
Signed-off-by: Feng Zhou <zhoufeng.zf@bytedance.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Yonghong Song <yhs@fb.com>
Link: https://lore.kernel.org/bpf/20220518025053.20492-1-zhoufeng.zf@bytedance.com

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

 // SPDX-License-Identifier: GPL-2.0
-// Copyright (c) 2022 Bytedance
+/* Copyright (c) 2022 Bytedance */
 
 #include <test_progs.h>
-
 #include "test_map_lookup_percpu_elem.skel.h"
 
-#define TEST_VALUE  1
-
 void test_map_lookup_percpu_elem(void)
 {
 	struct test_map_lookup_percpu_elem *skel;
-	int key = 0, ret;
-	int nr_cpus = sysconf(_SC_NPROCESSORS_ONLN);
-	int *buf;
+	__u64 key = 0, sum;
+	int ret, i, nr_cpus = libbpf_num_possible_cpus();
+	__u64 *buf;
 
-	buf = (int *)malloc(nr_cpus*sizeof(int));
+	buf = malloc(nr_cpus*sizeof(__u64));
 	if (!ASSERT_OK_PTR(buf, "malloc"))
 		return;
-	memset(buf, 0, nr_cpus*sizeof(int));
-	buf[0] = TEST_VALUE;
 
-	skel = test_map_lookup_percpu_elem__open_and_load();
-	if (!ASSERT_OK_PTR(skel, "test_map_lookup_percpu_elem__open_and_load"))
-		return;
+	for (i = 0; i < nr_cpus; i++)
+		buf[i] = i;
+	sum = (nr_cpus - 1) * nr_cpus / 2;
+
+	skel = test_map_lookup_percpu_elem__open();
+	if (!ASSERT_OK_PTR(skel, "test_map_lookup_percpu_elem__open"))
+		goto exit;
+
+	skel->rodata->my_pid = getpid();
+	skel->rodata->nr_cpus = nr_cpus;
+
+	ret = test_map_lookup_percpu_elem__load(skel);
+	if (!ASSERT_OK(ret, "test_map_lookup_percpu_elem__load"))
+		goto cleanup;
+
 	ret = test_map_lookup_percpu_elem__attach(skel);
-	ASSERT_OK(ret, "test_map_lookup_percpu_elem__attach");
+	if (!ASSERT_OK(ret, "test_map_lookup_percpu_elem__attach"))
+		goto cleanup;
 
 	ret = bpf_map_update_elem(bpf_map__fd(skel->maps.percpu_array_map), &key, buf, 0);
 	ASSERT_OK(ret, "percpu_array_map update");
@@ -37,10 +45,14 @@ void test_map_lookup_percpu_elem(void)
 
 	syscall(__NR_getuid);
 
-	ret = skel->bss->percpu_array_elem_val == TEST_VALUE &&
-	      skel->bss->percpu_hash_elem_val == TEST_VALUE &&
-	      skel->bss->percpu_lru_hash_elem_val == TEST_VALUE;
-	ASSERT_OK(!ret, "bpf_map_lookup_percpu_elem success");
+	test_map_lookup_percpu_elem__detach(skel);
+
+	ASSERT_EQ(skel->bss->percpu_array_elem_sum, sum, "percpu_array lookup percpu elem");
+	ASSERT_EQ(skel->bss->percpu_hash_elem_sum, sum, "percpu_hash lookup percpu elem");
+	ASSERT_EQ(skel->bss->percpu_lru_hash_elem_sum, sum, "percpu_lru_hash lookup percpu elem");
 
+cleanup:
 	test_map_lookup_percpu_elem__destroy(skel);
+exit:
+	free(buf);
 }

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

 // SPDX-License-Identifier: GPL-2.0
-// Copyright (c) 2022 Bytedance
+/* Copyright (c) 2022 Bytedance */
 
 #include "vmlinux.h"
 #include <bpf/bpf_helpers.h>
 
-int percpu_array_elem_val = 0;
-int percpu_hash_elem_val = 0;
-int percpu_lru_hash_elem_val = 0;
+__u64 percpu_array_elem_sum = 0;
+__u64 percpu_hash_elem_sum = 0;
+__u64 percpu_lru_hash_elem_sum = 0;
+const volatile int nr_cpus;
+const volatile int my_pid;
 
 struct {
 	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
 	__uint(max_entries, 1);
 	__type(key, __u32);
-	__type(value, __u32);
+	__type(value, __u64);
 } percpu_array_map SEC(".maps");
 
 struct {
 	__uint(type, BPF_MAP_TYPE_PERCPU_HASH);
 	__uint(max_entries, 1);
-	__type(key, __u32);
-	__type(value, __u32);
+	__type(key, __u64);
+	__type(value, __u64);
 } percpu_hash_map SEC(".maps");
 
 struct {
 	__uint(type, BPF_MAP_TYPE_LRU_PERCPU_HASH);
 	__uint(max_entries, 1);
-	__type(key, __u32);
-	__type(value, __u32);
+	__type(key, __u64);
+	__type(value, __u64);
 } percpu_lru_hash_map SEC(".maps");
 
+struct read_percpu_elem_ctx {
+	void *map;
+	__u64 sum;
+};
+
+static int read_percpu_elem_callback(__u32 index, struct read_percpu_elem_ctx *ctx)
+{
+	__u64 key = 0;
+	__u64 *value;
+
+	value = bpf_map_lookup_percpu_elem(ctx->map, &key, index);
+	if (value)
+		ctx->sum += *value;
+	return 0;
+}
+
 SEC("tp/syscalls/sys_enter_getuid")
 int sysenter_getuid(const void *ctx)
 {
-	__u32 key = 0;
-	__u32 cpu = 0;
-	__u32 *value;
+	struct read_percpu_elem_ctx map_ctx;
 
-	value = bpf_map_lookup_percpu_elem(&percpu_array_map, &key, cpu);
-	if (value)
-		percpu_array_elem_val = *value;
+	if (my_pid != (bpf_get_current_pid_tgid() >> 32))
+		return 0;
 
-	value = bpf_map_lookup_percpu_elem(&percpu_hash_map, &key, cpu);
-	if (value)
-		percpu_hash_elem_val = *value;
+	map_ctx.map = &percpu_array_map;
+	map_ctx.sum = 0;
+	bpf_loop(nr_cpus, read_percpu_elem_callback, &map_ctx, 0);
+	percpu_array_elem_sum = map_ctx.sum;
 
-	value = bpf_map_lookup_percpu_elem(&percpu_lru_hash_map, &key, cpu);
-	if (value)
-		percpu_lru_hash_elem_val = *value;
+	map_ctx.map = &percpu_hash_map;
+	map_ctx.sum = 0;
+	bpf_loop(nr_cpus, read_percpu_elem_callback, &map_ctx, 0);
+	percpu_hash_elem_sum = map_ctx.sum;
+
+	map_ctx.map = &percpu_lru_hash_map;
+	map_ctx.sum = 0;
+	bpf_loop(nr_cpus, read_percpu_elem_callback, &map_ctx, 0);
+	percpu_lru_hash_elem_sum = map_ctx.sum;
 
 	return 0;
 }