From: Jiri Olsa <jolsa@redhat.com>
To: Alexei Starovoitov <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Andrii Nakryiko <andriin@fb.com>,
"Steven Rostedt (VMware)" <rostedt@goodmis.org>
Cc: netdev@vger.kernel.org, bpf@vger.kernel.org,
Martin KaFai Lau <kafai@fb.com>, Song Liu <songliubraving@fb.com>,
Yonghong Song <yhs@fb.com>,
John Fastabend <john.fastabend@gmail.com>,
KP Singh <kpsingh@chromium.org>, Daniel Xu <dxu@dxuuu.xyz>,
Viktor Malik <vmalik@redhat.com>
Subject: [PATCH bpf-next v4 26/27] selftests/bpf: Add attach multi func test
Date: Thu, 26 Aug 2021 21:39:21 +0200 [thread overview]
Message-ID: <20210826193922.66204-27-jolsa@kernel.org> (raw)
In-Reply-To: <20210826193922.66204-1-jolsa@kernel.org>
Adding selftest to check attaching rules for multi func programs.
- attach 2 programs:
fentry/bpf_fentry_test1
fexit/bpf_fentry_test2
- check that we can attach multi func program on top of them:
fentry.multi/bpf_fentry_test*
- check that we cannot attach another multi funct program
that does not cover the same BTF ids (one less):
fentry.multi/bpf_fentry_test[1-7]
fexit.multi/bpf_fentry_test[1-7]
- check that we can no longer attach standard trampoline
programs (below) on top of attached multi func program:
fentry/bpf_fentry_test1
fexit/bpf_fentry_test3
Because the supported wildcards do not allow us to
match just limited set of bpf_fentry_test*, adding
extra code to look it up in kernel's BTF.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
---
.../bpf/prog_tests/multi_attach_check_test.c | 115 ++++++++++++++++++
.../selftests/bpf/progs/multi_attach_check.c | 36 ++++++
| 12 ++
| 12 ++
4 files changed, 175 insertions(+)
create mode 100644 tools/testing/selftests/bpf/prog_tests/multi_attach_check_test.c
create mode 100644 tools/testing/selftests/bpf/progs/multi_attach_check.c
create mode 100644 tools/testing/selftests/bpf/progs/multi_attach_check_extra1.c
create mode 100644 tools/testing/selftests/bpf/progs/multi_attach_check_extra2.c
diff --git a/tools/testing/selftests/bpf/prog_tests/multi_attach_check_test.c b/tools/testing/selftests/bpf/prog_tests/multi_attach_check_test.c
new file mode 100644
index 000000000000..32b23718437d
--- /dev/null
+++ b/tools/testing/selftests/bpf/prog_tests/multi_attach_check_test.c
@@ -0,0 +1,115 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <test_progs.h>
+#include <linux/btf_ids.h>
+#include "multi_attach_check.skel.h"
+#include "multi_attach_check_extra1.skel.h"
+#include "multi_attach_check_extra2.skel.h"
+#include <bpf/btf.h>
+
+static __u32 btf_ids[7];
+
+static int load_btf_ids(void)
+{
+ __u32 i, nr_types, cnt;
+ struct btf *btf;
+
+ btf = btf__load_vmlinux_btf();
+ if (!ASSERT_OK_PTR(btf, "btf__load_vmlinux_btf"))
+ return -1;
+
+ nr_types = btf__get_nr_types(btf);
+
+ for (i = 1, cnt = 0; i <= nr_types && cnt < 7; i++) {
+ const struct btf_type *t = btf__type_by_id(btf, i);
+ const char *name;
+
+ if (!btf_is_func(t))
+ continue;
+
+ name = btf__name_by_offset(btf, t->name_off);
+ if (!name)
+ continue;
+ if (strncmp(name, "bpf_fentry_test", sizeof("bpf_fentry_test") - 1))
+ continue;
+
+ btf_ids[cnt] = i;
+ cnt++;
+ }
+
+ btf__free(btf);
+ return ASSERT_EQ(cnt, 7, "bpf_fentry_test_cnt") ? 0 : -1;
+}
+
+void test_multi_attach_check_test(void)
+{
+ struct bpf_link *link1 = NULL, *link2 = NULL, *link3 = NULL;
+ DECLARE_LIBBPF_OPTS(bpf_link_create_opts, opts);
+ struct multi_attach_check_extra1 *skel_extra1 = NULL;
+ struct multi_attach_check_extra2 *skel_extra2 = NULL;
+ struct multi_attach_check *skel;
+ int link_fd, prog_fd;
+
+ /* Load/attach standard trampolines and on top of it multi
+ * func program. It should succeed.
+ */
+ skel = multi_attach_check__open_and_load();
+ if (!ASSERT_OK_PTR(skel, "multi_attach_check__load"))
+ return;
+
+ link1 = bpf_program__attach(skel->progs.test1);
+ if (!ASSERT_OK_PTR(link1, "multi_attach_check__test1_attach"))
+ goto cleanup;
+
+ link2 = bpf_program__attach(skel->progs.test2);
+ if (!ASSERT_OK_PTR(link2, "multi_attach_check__test2_attach"))
+ goto cleanup;
+
+ link3 = bpf_program__attach(skel->progs.test3);
+ if (!ASSERT_OK_PTR(link3, "multi_attach_check__test3_attach"))
+ goto cleanup;
+
+ if (!ASSERT_OK(load_btf_ids(), "load_btf_ids"))
+ goto cleanup;
+
+ /* There's 8 bpf_fentry_test* functions, get BTF ids for 7 of them
+ * and try to load/link multi func program with them. It should fail
+ * both for fentry.multi ...
+ */
+ opts.multi.btf_ids = btf_ids;
+ opts.multi.btf_ids_cnt = 7;
+
+ prog_fd = bpf_program__fd(skel->progs.test4);
+
+ link_fd = bpf_link_create(prog_fd, 0, BPF_TRACE_FENTRY, &opts);
+ if (!ASSERT_LT(link_fd, 0, "bpf_link_create"))
+ goto cleanup;
+
+ close(link_fd);
+
+ /* ... and fexit.multi */
+ prog_fd = bpf_program__fd(skel->progs.test5);
+
+ link_fd = bpf_link_create(prog_fd, 0, BPF_TRACE_FEXIT, &opts);
+ if (!ASSERT_LT(link_fd, 0, "bpf_link_create"))
+ goto cleanup;
+
+ close(link_fd);
+
+ /* Try to load/attach extra programs on top of multi func programs,
+ * it should fail for both fentry ...
+ */
+ skel_extra1 = multi_attach_check_extra1__open_and_load();
+ if (!ASSERT_ERR_PTR(skel_extra1, "multi_attach_check_extra1__load"))
+ multi_attach_check_extra1__destroy(skel_extra1);
+
+ /* ... and fexit */
+ skel_extra2 = multi_attach_check_extra2__open_and_load();
+ if (!ASSERT_ERR_PTR(skel_extra2, "multi_attach_check_extra2__load"))
+ multi_attach_check_extra2__destroy(skel_extra2);
+
+cleanup:
+ bpf_link__destroy(link1);
+ bpf_link__destroy(link2);
+ bpf_link__destroy(link3);
+ multi_attach_check__destroy(skel);
+}
diff --git a/tools/testing/selftests/bpf/progs/multi_attach_check.c b/tools/testing/selftests/bpf/progs/multi_attach_check.c
new file mode 100644
index 000000000000..cf0f25c69556
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/multi_attach_check.c
@@ -0,0 +1,36 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
+
+char _license[] SEC("license") = "GPL";
+
+SEC("fentry/bpf_fentry_test1")
+int BPF_PROG(test1, int a)
+{
+ return 0;
+}
+
+SEC("fexit/bpf_fentry_test2")
+int BPF_PROG(test2, int a, __u64 b, int ret)
+{
+ return 0;
+}
+
+SEC("fentry.multi/bpf_fentry_test*")
+int BPF_PROG(test3, __u64 a, __u64 b, __u64 c, __u64 d, __u64 e, __u64 f)
+{
+ return 0;
+}
+
+SEC("fentry.multi/bpf_fentry_test1-7")
+int BPF_PROG(test4, __u64 a, __u64 b, __u64 c, __u64 d, __u64 e, __u64 f, int ret)
+{
+ return 0;
+}
+
+SEC("fexit.multi/bpf_fentry_test1-7")
+int BPF_PROG(test5, __u64 a, __u64 b, __u64 c, __u64 d, __u64 e, __u64 f, int ret)
+{
+ return 0;
+}
--git a/tools/testing/selftests/bpf/progs/multi_attach_check_extra1.c b/tools/testing/selftests/bpf/progs/multi_attach_check_extra1.c
new file mode 100644
index 000000000000..c1d816a0206a
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/multi_attach_check_extra1.c
@@ -0,0 +1,12 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
+
+char _license[] SEC("license") = "GPL";
+
+SEC("fentry/bpf_fentry_test1")
+int BPF_PROG(test1, int a)
+{
+ return 0;
+}
--git a/tools/testing/selftests/bpf/progs/multi_attach_check_extra2.c b/tools/testing/selftests/bpf/progs/multi_attach_check_extra2.c
new file mode 100644
index 000000000000..cd66abb4b848
--- /dev/null
+++ b/tools/testing/selftests/bpf/progs/multi_attach_check_extra2.c
@@ -0,0 +1,12 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+#include <bpf/bpf_tracing.h>
+
+char _license[] SEC("license") = "GPL";
+
+SEC("fexit/bpf_fentry_test3")
+int BPF_PROG(test3, int a, __u64 b, int ret)
+{
+ return 0;
+}
--
2.31.1
next prev parent reply other threads:[~2021-08-26 19:42 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-26 19:38 [PATCH bpf-next v4 00/27] x86/ftrace/bpf: Add batch support for direct/tracing attach Jiri Olsa
2021-08-26 19:38 ` [PATCH bpf-next v4 01/27] x86/ftrace: Remove extra orig rax move Jiri Olsa
2021-08-26 19:38 ` [PATCH bpf-next v4 02/27] x86/ftrace: Remove fault protection code in prepare_ftrace_return Jiri Olsa
2021-08-26 19:38 ` [PATCH bpf-next v4 03/27] x86/ftrace: Make function graph use ftrace directly Jiri Olsa
2021-08-26 19:38 ` [PATCH bpf-next v4 04/27] tracing: Add trampoline/graph selftest Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 05/27] ftrace: Add ftrace_add_rec_direct function Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 06/27] ftrace: Add multi direct register/unregister interface Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 07/27] ftrace: Add multi direct modify interface Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 08/27] ftrace/samples: Add multi direct interface test module Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 09/27] bpf: Add support to load multi func tracing program Jiri Olsa
2021-08-31 23:17 ` Andrii Nakryiko
2021-09-01 11:32 ` Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 10/27] bpf: Add struct bpf_tramp_node layer Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 11/27] bpf: Factor out bpf_trampoline_init function Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 12/27] bpf: Factor out __bpf_trampoline_lookup function Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 13/27] bpf: Factor out __bpf_trampoline_put function Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 14/27] bpf: Change bpf_trampoline_get to return error pointer Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 15/27] bpf, x64: Allow to use caller address from stack Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 16/27] bpf: Add bpf_trampoline_multi_get/put functions Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 17/27] bpf: Add multi trampoline attach support Jiri Olsa
2021-08-31 23:36 ` Andrii Nakryiko
2021-09-01 0:02 ` Andrii Nakryiko
2021-09-01 11:39 ` Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 18/27] bpf, x64: Store properly return value for trampoline with multi func programs Jiri Olsa
2021-08-31 23:51 ` Andrii Nakryiko
2021-09-01 15:15 ` Jiri Olsa
2021-09-02 3:56 ` Andrii Nakryiko
2021-09-02 12:57 ` Jiri Olsa
2021-09-02 16:54 ` Andrii Nakryiko
2021-09-02 21:55 ` Alexei Starovoitov
2021-09-03 9:50 ` Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 19/27] bpf: Attach multi trampoline with ftrace_ops Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 20/27] libbpf: Add btf__find_by_glob_kind function Jiri Olsa
2021-09-01 0:10 ` Andrii Nakryiko
2021-09-01 11:33 ` Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 21/27] libbpf: Add support to link multi func tracing program Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 22/27] selftests/bpf: Add fentry multi func test Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 23/27] selftests/bpf: Add fexit " Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 24/27] selftests/bpf: Add fentry/fexit " Jiri Olsa
2021-08-26 19:39 ` [PATCH bpf-next v4 25/27] selftests/bpf: Add mixed " Jiri Olsa
2021-08-26 19:39 ` Jiri Olsa [this message]
2021-08-26 19:39 ` [PATCH bpf-next v4 27/27] selftests/bpf: Add ret_mod " Jiri Olsa
2021-08-29 17:04 ` [PATCH bpf-next v4 00/27] x86/ftrace/bpf: Add batch support for direct/tracing attach Alexei Starovoitov
2021-08-30 8:02 ` Jiri Olsa
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210826193922.66204-27-jolsa@kernel.org \
--to=jolsa@redhat.com \
--cc=andriin@fb.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=dxu@dxuuu.xyz \
--cc=john.fastabend@gmail.com \
--cc=kafai@fb.com \
--cc=kpsingh@chromium.org \
--cc=netdev@vger.kernel.org \
--cc=rostedt@goodmis.org \
--cc=songliubraving@fb.com \
--cc=vmalik@redhat.com \
--cc=yhs@fb.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox