diff mbox series

[bpf-next,v2,7/7] selftests/bpf: Add more tests for kptrs in maps

Message ID 20230221200646.2500777-8-memxor@gmail.com (mailing list archive)
State Superseded
Delegated to: BPF
Headers show
Series Add support for kptrs in more BPF maps | expand

Checks

Context Check Description
netdev/tree_selection success Clearly marked for bpf-next
netdev/fixes_present success Fixes tag not required for -next series
netdev/subject_prefix success Link
netdev/cover_letter success Series has a cover letter
netdev/patch_count success Link
netdev/header_inline success No static functions without inline keyword in header files
netdev/build_32bit success Errors and warnings before: 0 this patch: 0
netdev/cc_maintainers warning 10 maintainers not CCed: linux-kselftest@vger.kernel.org john.fastabend@gmail.com sdf@google.com shuah@kernel.org jolsa@kernel.org song@kernel.org martin.lau@linux.dev mykolal@fb.com haoluo@google.com yhs@fb.com
netdev/build_clang success Errors and warnings before: 0 this patch: 0
netdev/module_param success Was 0 now: 0
netdev/verify_signedoff success Signed-off-by tag matches author and committer
netdev/check_selftest success No net selftest shell script
netdev/verify_fixes success No Fixes tag
netdev/build_allmodconfig_warn success Errors and warnings before: 0 this patch: 0
netdev/checkpatch fail CHECK: Macro argument 'map' may be better as '(map)' to avoid precedence issues ERROR: Macros with multiple statements should be enclosed in a do - while loop WARNING: 'accomodate' may be misspelled - perhaps 'accommodate'? WARNING: Macros with flow control statements should be avoided WARNING: line length of 100 exceeds 80 columns WARNING: line length of 82 exceeds 80 columns WARNING: line length of 83 exceeds 80 columns WARNING: line length of 84 exceeds 80 columns WARNING: line length of 85 exceeds 80 columns WARNING: line length of 87 exceeds 80 columns WARNING: line length of 90 exceeds 80 columns WARNING: line length of 91 exceeds 80 columns WARNING: line length of 93 exceeds 80 columns WARNING: line length of 94 exceeds 80 columns WARNING: line length of 97 exceeds 80 columns WARNING: macros should not use a trailing semicolon
netdev/kdoc success Errors and warnings before: 0 this patch: 0
netdev/source_inline success Was 0 now: 0
bpf/vmtest-bpf-next-VM_Test-11 success Logs for test_maps on s390x with gcc
bpf/vmtest-bpf-next-VM_Test-12 success Logs for test_maps on x86_64 with gcc
bpf/vmtest-bpf-next-VM_Test-13 success Logs for test_maps on x86_64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-17 success Logs for test_progs on x86_64 with gcc
bpf/vmtest-bpf-next-VM_Test-18 fail Logs for test_progs on x86_64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-22 success Logs for test_progs_no_alu32 on x86_64 with gcc
bpf/vmtest-bpf-next-VM_Test-23 success Logs for test_progs_no_alu32 on x86_64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-27 success Logs for test_progs_no_alu32_parallel on x86_64 with gcc
bpf/vmtest-bpf-next-VM_Test-28 success Logs for test_progs_no_alu32_parallel on x86_64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-32 success Logs for test_progs_parallel on x86_64 with gcc
bpf/vmtest-bpf-next-VM_Test-33 success Logs for test_progs_parallel on x86_64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-37 success Logs for test_verifier on x86_64 with gcc
bpf/vmtest-bpf-next-VM_Test-38 success Logs for test_verifier on x86_64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-10 success Logs for test_maps on aarch64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-14 fail Logs for test_progs on aarch64 with gcc
bpf/vmtest-bpf-next-VM_Test-15 fail Logs for test_progs on aarch64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-19 fail Logs for test_progs_no_alu32 on aarch64 with gcc
bpf/vmtest-bpf-next-VM_Test-20 fail Logs for test_progs_no_alu32 on aarch64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-24 success Logs for test_progs_no_alu32_parallel on aarch64 with gcc
bpf/vmtest-bpf-next-VM_Test-25 success Logs for test_progs_no_alu32_parallel on aarch64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-29 success Logs for test_progs_parallel on aarch64 with gcc
bpf/vmtest-bpf-next-VM_Test-30 success Logs for test_progs_parallel on aarch64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-34 success Logs for test_verifier on aarch64 with gcc
bpf/vmtest-bpf-next-VM_Test-35 success Logs for test_verifier on aarch64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-21 success Logs for test_progs_no_alu32 on s390x with gcc
bpf/vmtest-bpf-next-VM_Test-26 success Logs for test_progs_no_alu32_parallel on s390x with gcc
bpf/vmtest-bpf-next-VM_Test-36 success Logs for test_verifier on s390x with gcc
bpf/vmtest-bpf-next-VM_Test-16 success Logs for test_progs on s390x with gcc
bpf/vmtest-bpf-next-VM_Test-31 success Logs for test_progs_parallel on s390x with gcc
bpf/vmtest-bpf-next-PR success PR summary
bpf/vmtest-bpf-next-VM_Test-1 success Logs for ${{ matrix.test }} on ${{ matrix.arch }} with ${{ matrix.toolchain }}
bpf/vmtest-bpf-next-VM_Test-2 success Logs for ShellCheck
bpf/vmtest-bpf-next-VM_Test-3 fail Logs for build for aarch64 with gcc
bpf/vmtest-bpf-next-VM_Test-4 fail Logs for build for aarch64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-5 fail Logs for build for s390x with gcc
bpf/vmtest-bpf-next-VM_Test-6 success Logs for build for x86_64 with gcc
bpf/vmtest-bpf-next-VM_Test-7 success Logs for build for x86_64 with llvm-16
bpf/vmtest-bpf-next-VM_Test-8 success Logs for llvm-toolchain
bpf/vmtest-bpf-next-VM_Test-9 success Logs for set-matrix

Commit Message

Kumar Kartikeya Dwivedi Feb. 21, 2023, 8:06 p.m. UTC
Firstly, ensure programs successfully load when using all of the
supported maps. Then, extend existing tests to test more cases at
runtime. We are currently testing both the synchronous freeing of items
and asynchronous destruction when map is freed, but the code needs to be
adjusted a bit to be able to also accomodate support for percpu maps.

We now do a delete on the item (and update for array maps which has a
similar effect for kptrs) to perform a synchronous free of the kptr, and
test destruction both for the synchronous and asynchronous deletion.
Next time the program runs, it should observe the refcount as 1 since
all existing references should have been released by then. By running
the program after both possible paths freeing kptrs, we establish that
they correctly release resources. Next, we augment the existing test to
also test the same code path shared by all local storage maps using a
task local storage map.

Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
---
 .../selftests/bpf/prog_tests/map_kptr.c       | 122 ++++--
 tools/testing/selftests/bpf/progs/map_kptr.c  | 353 +++++++++++++++---
 2 files changed, 410 insertions(+), 65 deletions(-)
diff mbox series

Patch

diff --git a/tools/testing/selftests/bpf/prog_tests/map_kptr.c b/tools/testing/selftests/bpf/prog_tests/map_kptr.c
index 3533a4ecad01..550497ee04f0 100644
--- a/tools/testing/selftests/bpf/prog_tests/map_kptr.c
+++ b/tools/testing/selftests/bpf/prog_tests/map_kptr.c
@@ -7,67 +7,143 @@ 
 
 static void test_map_kptr_success(bool test_run)
 {
+	LIBBPF_OPTS(bpf_test_run_opts, lopts);
 	LIBBPF_OPTS(bpf_test_run_opts, opts,
 		.data_in = &pkt_v4,
 		.data_size_in = sizeof(pkt_v4),
 		.repeat = 1,
 	);
+	int key = 0, ret, cpu;
 	struct map_kptr *skel;
-	int key = 0, ret;
-	char buf[16];
+	struct bpf_link *link;
+	char buf[16], *pbuf;
 
 	skel = map_kptr__open_and_load();
 	if (!ASSERT_OK_PTR(skel, "map_kptr__open_and_load"))
 		return;
 
-	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref), &opts);
-	ASSERT_OK(ret, "test_map_kptr_ref refcount");
-	ASSERT_OK(opts.retval, "test_map_kptr_ref retval");
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref1), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref1 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref1 retval");
 	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref2), &opts);
 	ASSERT_OK(ret, "test_map_kptr_ref2 refcount");
 	ASSERT_OK(opts.retval, "test_map_kptr_ref2 retval");
 
+	link = bpf_program__attach(skel->progs.test_ls_map_kptr_ref1);
+	if (!ASSERT_OK_PTR(link, "bpf_program__attach ref1"))
+		goto exit;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_ls_map_kptr_ref1), &lopts);
+	ASSERT_OK(ret, "test_ls_map_kptr_ref1 refcount");
+	ASSERT_EQ((lopts.retval << 16) >> 16, 9000, "test_ls_map_kptr_ref1 retval");
+	if (!ASSERT_OK(bpf_link__destroy(link), "bpf_link__destroy"))
+		goto exit;
+
+	link = bpf_program__attach(skel->progs.test_ls_map_kptr_ref2);
+	if (!ASSERT_OK_PTR(link, "bpf_program__attach ref2"))
+		goto exit;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_ls_map_kptr_ref2), &lopts);
+	ASSERT_OK(ret, "test_ls_map_kptr_ref2 refcount");
+	ASSERT_EQ((lopts.retval << 16) >> 16, 9000, "test_ls_map_kptr_ref2 retval");
+	if (!ASSERT_OK(bpf_link__destroy(link), "bpf_link__destroy"))
+		goto exit;
+
 	if (test_run)
 		goto exit;
 
+	cpu = libbpf_num_possible_cpus();
+	if (!ASSERT_GT(cpu, 0, "libbpf_num_possible_cpus"))
+		goto exit;
+
+	pbuf = calloc(cpu, sizeof(buf));
+	if (!ASSERT_OK_PTR(pbuf, "calloc(pbuf)"))
+		goto exit;
+
 	ret = bpf_map__update_elem(skel->maps.array_map,
 				   &key, sizeof(key), buf, sizeof(buf), 0);
 	ASSERT_OK(ret, "array_map update");
-	ret = bpf_map__update_elem(skel->maps.array_map,
-				   &key, sizeof(key), buf, sizeof(buf), 0);
-	ASSERT_OK(ret, "array_map update2");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
+
+	ret = bpf_map__update_elem(skel->maps.pcpu_array_map,
+				   &key, sizeof(key), pbuf, cpu * sizeof(buf), 0);
+	ASSERT_OK(ret, "pcpu_array_map update");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
 
-	ret = bpf_map__update_elem(skel->maps.hash_map,
-				   &key, sizeof(key), buf, sizeof(buf), 0);
-	ASSERT_OK(ret, "hash_map update");
 	ret = bpf_map__delete_elem(skel->maps.hash_map, &key, sizeof(key), 0);
 	ASSERT_OK(ret, "hash_map delete");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
+
+	ret = bpf_map__delete_elem(skel->maps.pcpu_hash_map, &key, sizeof(key), 0);
+	ASSERT_OK(ret, "pcpu_hash_map delete");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
 
-	ret = bpf_map__update_elem(skel->maps.hash_malloc_map,
-				   &key, sizeof(key), buf, sizeof(buf), 0);
-	ASSERT_OK(ret, "hash_malloc_map update");
 	ret = bpf_map__delete_elem(skel->maps.hash_malloc_map, &key, sizeof(key), 0);
 	ASSERT_OK(ret, "hash_malloc_map delete");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
+
+	ret = bpf_map__delete_elem(skel->maps.pcpu_hash_malloc_map, &key, sizeof(key), 0);
+	ASSERT_OK(ret, "pcpu_hash_malloc_map delete");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
 
-	ret = bpf_map__update_elem(skel->maps.lru_hash_map,
-				   &key, sizeof(key), buf, sizeof(buf), 0);
-	ASSERT_OK(ret, "lru_hash_map update");
 	ret = bpf_map__delete_elem(skel->maps.lru_hash_map, &key, sizeof(key), 0);
 	ASSERT_OK(ret, "lru_hash_map delete");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
+
+	ret = bpf_map__delete_elem(skel->maps.lru_pcpu_hash_map, &key, sizeof(key), 0);
+	ASSERT_OK(ret, "lru_pcpu_hash_map delete");
+	skel->data->ref--;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_map_kptr_ref3), &opts);
+	ASSERT_OK(ret, "test_map_kptr_ref3 refcount");
+	ASSERT_OK(opts.retval, "test_map_kptr_ref3 retval");
+
+	link = bpf_program__attach(skel->progs.test_ls_map_kptr_ref_del);
+	if (!ASSERT_OK_PTR(link, "bpf_program__attach ref_del"))
+		goto exit;
+	ret = bpf_prog_test_run_opts(bpf_program__fd(skel->progs.test_ls_map_kptr_ref_del), &lopts);
+	ASSERT_OK(ret, "test_ls_map_kptr_ref_del delete");
+	skel->data->ref--;
+	ASSERT_EQ((lopts.retval << 16) >> 16, 9000, "test_ls_map_kptr_ref_del retval");
+	if (!ASSERT_OK(bpf_link__destroy(link), "bpf_link__destroy"))
+		goto exit;
 
+	free(pbuf);
 exit:
 	map_kptr__destroy(skel);
 }
 
 void test_map_kptr(void)
 {
-	if (test__start_subtest("success")) {
+	RUN_TESTS(map_kptr_fail);
+
+	if (test__start_subtest("success-map")) {
+		test_map_kptr_success(true);
+
+		ASSERT_OK(kern_sync_rcu(), "sync rcu");
+		/* Observe refcount dropping to 1 on bpf_map_free_deferred */
 		test_map_kptr_success(false);
-		/* Do test_run twice, so that we see refcount going back to 1
-		 * after we leave it in map from first iteration.
-		 */
+
+		ASSERT_OK(kern_sync_rcu(), "sync rcu");
+		/* Observe refcount dropping to 1 on synchronous delete elem */
 		test_map_kptr_success(true);
 	}
-
-	RUN_TESTS(map_kptr_fail);
 }
diff --git a/tools/testing/selftests/bpf/progs/map_kptr.c b/tools/testing/selftests/bpf/progs/map_kptr.c
index 228ec45365a8..f8d7f2adccc9 100644
--- a/tools/testing/selftests/bpf/progs/map_kptr.c
+++ b/tools/testing/selftests/bpf/progs/map_kptr.c
@@ -15,6 +15,13 @@  struct array_map {
 	__uint(max_entries, 1);
 } array_map SEC(".maps");
 
+struct pcpu_array_map {
+	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+	__type(key, int);
+	__type(value, struct map_value);
+	__uint(max_entries, 1);
+} pcpu_array_map SEC(".maps");
+
 struct hash_map {
 	__uint(type, BPF_MAP_TYPE_HASH);
 	__type(key, int);
@@ -22,6 +29,13 @@  struct hash_map {
 	__uint(max_entries, 1);
 } hash_map SEC(".maps");
 
+struct pcpu_hash_map {
+	__uint(type, BPF_MAP_TYPE_PERCPU_HASH);
+	__type(key, int);
+	__type(value, struct map_value);
+	__uint(max_entries, 1);
+} pcpu_hash_map SEC(".maps");
+
 struct hash_malloc_map {
 	__uint(type, BPF_MAP_TYPE_HASH);
 	__type(key, int);
@@ -30,6 +44,14 @@  struct hash_malloc_map {
 	__uint(map_flags, BPF_F_NO_PREALLOC);
 } hash_malloc_map SEC(".maps");
 
+struct pcpu_hash_malloc_map {
+	__uint(type, BPF_MAP_TYPE_PERCPU_HASH);
+	__type(key, int);
+	__type(value, struct map_value);
+	__uint(max_entries, 1);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+} pcpu_hash_malloc_map SEC(".maps");
+
 struct lru_hash_map {
 	__uint(type, BPF_MAP_TYPE_LRU_HASH);
 	__type(key, int);
@@ -37,6 +59,41 @@  struct lru_hash_map {
 	__uint(max_entries, 1);
 } lru_hash_map SEC(".maps");
 
+struct lru_pcpu_hash_map {
+	__uint(type, BPF_MAP_TYPE_LRU_PERCPU_HASH);
+	__type(key, int);
+	__type(value, struct map_value);
+	__uint(max_entries, 1);
+} lru_pcpu_hash_map SEC(".maps");
+
+struct cgrp_ls_map {
+	__uint(type, BPF_MAP_TYPE_CGRP_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct map_value);
+} cgrp_ls_map SEC(".maps");
+
+struct task_ls_map {
+	__uint(type, BPF_MAP_TYPE_TASK_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct map_value);
+} task_ls_map SEC(".maps");
+
+struct inode_ls_map {
+	__uint(type, BPF_MAP_TYPE_INODE_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct map_value);
+} inode_ls_map SEC(".maps");
+
+struct sk_ls_map {
+	__uint(type, BPF_MAP_TYPE_SK_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct map_value);
+} sk_ls_map SEC(".maps");
+
 #define DEFINE_MAP_OF_MAP(map_type, inner_map_type, name)       \
 	struct {                                                \
 		__uint(type, map_type);                         \
@@ -160,6 +217,58 @@  int test_map_kptr(struct __sk_buff *ctx)
 	return 0;
 }
 
+SEC("tp_btf/cgroup_mkdir")
+int BPF_PROG(test_cgrp_map_kptr, struct cgroup *cgrp, const char *path)
+{
+	struct map_value *v;
+
+	v = bpf_cgrp_storage_get(&cgrp_ls_map, cgrp, NULL, BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (v)
+		test_kptr(v);
+	return 0;
+}
+
+SEC("lsm/inode_unlink")
+int BPF_PROG(test_task_map_kptr, struct inode *inode, struct dentry *victim)
+{
+	struct task_struct *task;
+	struct map_value *v;
+
+	task = bpf_get_current_task_btf();
+	if (!task)
+		return 0;
+	v = bpf_task_storage_get(&task_ls_map, task, NULL, BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (v)
+		test_kptr(v);
+	return 0;
+}
+
+SEC("lsm/inode_unlink")
+int BPF_PROG(test_inode_map_kptr, struct inode *inode, struct dentry *victim)
+{
+	struct map_value *v;
+
+	v = bpf_inode_storage_get(&inode_ls_map, inode, NULL, BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (v)
+		test_kptr(v);
+	return 0;
+}
+
+SEC("tc")
+int test_sk_map_kptr(struct __sk_buff *ctx)
+{
+	struct map_value *v;
+	struct bpf_sock *sk;
+
+	sk = ctx->sk;
+	if (!sk)
+		return 0;
+	v = bpf_sk_storage_get(&sk_ls_map, sk, NULL, BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (v)
+		test_kptr(v);
+	return 0;
+}
+
 SEC("tc")
 int test_map_in_map_kptr(struct __sk_buff *ctx)
 {
@@ -189,106 +298,266 @@  int test_map_in_map_kptr(struct __sk_buff *ctx)
 	return 0;
 }
 
-SEC("tc")
-int test_map_kptr_ref(struct __sk_buff *ctx)
+int ref = 1;
+
+static __always_inline
+int test_map_kptr_ref_pre(struct map_value *v)
 {
 	struct prog_test_ref_kfunc *p, *p_st;
 	unsigned long arg = 0;
-	struct map_value *v;
-	int key = 0, ret;
+	int ret;
 
 	p = bpf_kfunc_call_test_acquire(&arg);
 	if (!p)
 		return 1;
+	ref++;
 
 	p_st = p->next;
-	if (p_st->cnt.refs.counter != 2) {
+	if (p_st->cnt.refs.counter != ref) {
 		ret = 2;
 		goto end;
 	}
 
-	v = bpf_map_lookup_elem(&array_map, &key);
-	if (!v) {
-		ret = 3;
-		goto end;
-	}
-
 	p = bpf_kptr_xchg(&v->ref_ptr, p);
 	if (p) {
-		ret = 4;
+		ret = 3;
 		goto end;
 	}
-	if (p_st->cnt.refs.counter != 2)
-		return 5;
+	if (p_st->cnt.refs.counter != ref)
+		return 4;
 
 	p = bpf_kfunc_call_test_kptr_get(&v->ref_ptr, 0, 0);
 	if (!p)
-		return 6;
-	if (p_st->cnt.refs.counter != 3) {
-		ret = 7;
+		return 5;
+	ref++;
+	if (p_st->cnt.refs.counter != ref) {
+		ret = 6;
 		goto end;
 	}
 	bpf_kfunc_call_test_release(p);
-	if (p_st->cnt.refs.counter != 2)
-		return 8;
+	ref--;
+	if (p_st->cnt.refs.counter != ref)
+		return 7;
 
 	p = bpf_kptr_xchg(&v->ref_ptr, NULL);
 	if (!p)
-		return 9;
+		return 8;
 	bpf_kfunc_call_test_release(p);
-	if (p_st->cnt.refs.counter != 1)
-		return 10;
+	ref--;
+	if (p_st->cnt.refs.counter != ref)
+		return 9;
 
 	p = bpf_kfunc_call_test_acquire(&arg);
 	if (!p)
-		return 11;
+		return 10;
+	ref++;
 	p = bpf_kptr_xchg(&v->ref_ptr, p);
 	if (p) {
-		ret = 12;
+		ret = 11;
 		goto end;
 	}
-	if (p_st->cnt.refs.counter != 2)
-		return 13;
+	if (p_st->cnt.refs.counter != ref)
+		return 12;
 	/* Leave in map */
 
 	return 0;
 end:
+	ref--;
 	bpf_kfunc_call_test_release(p);
 	return ret;
 }
 
-SEC("tc")
-int test_map_kptr_ref2(struct __sk_buff *ctx)
+static __always_inline
+int test_map_kptr_ref_post(struct map_value *v)
 {
 	struct prog_test_ref_kfunc *p, *p_st;
-	struct map_value *v;
-	int key = 0;
-
-	v = bpf_map_lookup_elem(&array_map, &key);
-	if (!v)
-		return 1;
 
 	p_st = v->ref_ptr;
-	if (!p_st || p_st->cnt.refs.counter != 2)
-		return 2;
+	if (!p_st || p_st->cnt.refs.counter != ref)
+		return 1;
 
 	p = bpf_kptr_xchg(&v->ref_ptr, NULL);
 	if (!p)
-		return 3;
-	if (p_st->cnt.refs.counter != 2) {
+		return 2;
+	if (p_st->cnt.refs.counter != ref) {
 		bpf_kfunc_call_test_release(p);
-		return 4;
+		return 3;
 	}
 
 	p = bpf_kptr_xchg(&v->ref_ptr, p);
 	if (p) {
 		bpf_kfunc_call_test_release(p);
-		return 5;
+		return 4;
 	}
-	if (p_st->cnt.refs.counter != 2)
-		return 6;
+	if (p_st->cnt.refs.counter != ref)
+		return 5;
+
+	return 0;
+}
+
+#define TEST(map)                            \
+	v = bpf_map_lookup_elem(&map, &key); \
+	if (!v)                              \
+		return -1;                   \
+	ret = test_map_kptr_ref_pre(v);      \
+	if (ret)                             \
+		return ret;
+
+#define TEST_PCPU(map)                                 \
+	v = bpf_map_lookup_percpu_elem(&map, &key, 0); \
+	if (!v)                                        \
+		return -1;                             \
+	ret = test_map_kptr_ref_pre(v);                \
+	if (ret)                                       \
+		return ret;
+
+SEC("tc")
+int test_map_kptr_ref1(struct __sk_buff *ctx)
+{
+	struct map_value *v, val = {};
+	int key = 0, ret;
+
+	bpf_map_update_elem(&hash_map, &key, &val, 0);
+	bpf_map_update_elem(&hash_malloc_map, &key, &val, 0);
+	bpf_map_update_elem(&lru_hash_map, &key, &val, 0);
+
+	bpf_map_update_elem(&pcpu_hash_map, &key, &val, 0);
+	bpf_map_update_elem(&pcpu_hash_malloc_map, &key, &val, 0);
+	bpf_map_update_elem(&lru_pcpu_hash_map, &key, &val, 0);
+
+	TEST(array_map);
+	TEST(hash_map);
+	TEST(hash_malloc_map);
+	TEST(lru_hash_map);
+
+	TEST_PCPU(pcpu_array_map);
+	TEST_PCPU(pcpu_hash_map);
+	TEST_PCPU(pcpu_hash_malloc_map);
+	TEST_PCPU(lru_pcpu_hash_map);
+
+	return 0;
+}
+
+#undef TEST
+#undef TEST_PCPU
+
+#define TEST(map)                            \
+	v = bpf_map_lookup_elem(&map, &key); \
+	if (!v)                              \
+		return -1;                   \
+	ret = test_map_kptr_ref_post(v);     \
+	if (ret)                             \
+		return ret;
+
+#define TEST_PCPU(map)                                 \
+	v = bpf_map_lookup_percpu_elem(&map, &key, 0); \
+	if (!v)                                        \
+		return -1;                             \
+	ret = test_map_kptr_ref_post(v);               \
+	if (ret)                                       \
+		return ret;
+
+SEC("tc")
+int test_map_kptr_ref2(struct __sk_buff *ctx)
+{
+	struct map_value *v;
+	int key = 0, ret;
+
+	TEST(array_map);
+	TEST(hash_map);
+	TEST(hash_malloc_map);
+	TEST(lru_hash_map);
+
+	TEST_PCPU(pcpu_array_map);
+	TEST_PCPU(pcpu_hash_map);
+	TEST_PCPU(pcpu_hash_malloc_map);
+	TEST_PCPU(lru_pcpu_hash_map);
 
 	return 0;
 }
 
+#undef TEST
+#undef TEST_PCPU
+
+SEC("tc")
+int test_map_kptr_ref3(struct __sk_buff *ctx)
+{
+	struct prog_test_ref_kfunc *p;
+	unsigned long sp = 0;
+
+	p = bpf_kfunc_call_test_acquire(&sp);
+	if (!p)
+		return 1;
+	ref++;
+	if (p->cnt.refs.counter != ref) {
+		bpf_kfunc_call_test_release(p);
+		return 2;
+	}
+	bpf_kfunc_call_test_release(p);
+	ref--;
+	return 0;
+}
+
+SEC("fmod_ret/bpf_modify_return_test")
+int BPF_PROG(test_ls_map_kptr_ref1, int a, int *b)
+{
+	struct task_struct *current;
+	struct map_value *v;
+	int ret;
+
+	current = bpf_get_current_task_btf();
+	if (!current)
+		return 100;
+	v = bpf_task_storage_get(&task_ls_map, current, NULL, 0);
+	if (v)
+		return 150;
+	v = bpf_task_storage_get(&task_ls_map, current, NULL, BPF_LOCAL_STORAGE_GET_F_CREATE);
+	if (!v)
+		return 200;
+	ret = test_map_kptr_ref_pre(v);
+	if (ret)
+		return ret;
+	return 9000;
+}
+
+SEC("fmod_ret/bpf_modify_return_test")
+int BPF_PROG(test_ls_map_kptr_ref2, int a, int *b)
+{
+	struct task_struct *current;
+	struct map_value *v;
+	int ret;
+
+	current = bpf_get_current_task_btf();
+	if (!current)
+		return 100;
+	v = bpf_task_storage_get(&task_ls_map, current, NULL, 0);
+	if (!v)
+		return 200;
+	ret = test_map_kptr_ref_post(v);
+	if (ret)
+		return ret;
+	return 9000;
+}
+
+SEC("fmod_ret/bpf_modify_return_test")
+int BPF_PROG(test_ls_map_kptr_ref_del, int a, int *b)
+{
+	struct task_struct *current;
+	struct map_value *v;
+	int ret;
+
+	current = bpf_get_current_task_btf();
+	if (!current)
+		return 100;
+	v = bpf_task_storage_get(&task_ls_map, current, NULL, 0);
+	if (!v)
+		return 200;
+	if (!v->ref_ptr)
+		return 300;
+	ret = bpf_task_storage_delete(&task_ls_map, current);
+	if (ret)
+		return ret;
+	return 9000;
+}
+
 char _license[] SEC("license") = "GPL";