Re: [PATCH bpf-next v2 5/5] selftests/bpf: Add test for "bpftool feature" command

From: Quentin Monnet
Date: Fri Feb 21 2020 - 06:28:57 EST


2020-02-21 04:17 UTC+0100 ~ Michal Rostecki <mrostecki@xxxxxxxxxxxx>
Add Python module with tests for "bpftool feature" command, which mainly
wheck whether the "full" option is working properly.

Signed-off-by: Michal Rostecki <mrostecki@xxxxxxxxxxxx>
---
tools/testing/selftests/.gitignore | 5 +-
tools/testing/selftests/bpf/Makefile | 3 +-
tools/testing/selftests/bpf/test_bpftool.py | 228 ++++++++++++++++++++
tools/testing/selftests/bpf/test_bpftool.sh | 5 +
4 files changed, 239 insertions(+), 2 deletions(-)
create mode 100644 tools/testing/selftests/bpf/test_bpftool.py
create mode 100755 tools/testing/selftests/bpf/test_bpftool.sh

diff --git a/tools/testing/selftests/.gitignore b/tools/testing/selftests/.gitignore
index 61df01cdf0b2..304fdf1a21dc 100644
--- a/tools/testing/selftests/.gitignore
+++ b/tools/testing/selftests/.gitignore
@@ -3,4 +3,7 @@ gpiogpio-hammer
gpioinclude/
gpiolsgpio
tpm2/SpaceTest.log
-tpm2/*.pyc
+
+# Python bytecode and cache
+__pycache__/
+*.py[cod]
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 257a1aaaa37d..e7d822259c50 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -62,7 +62,8 @@ TEST_PROGS := test_kmod.sh \
test_tc_tunnel.sh \
test_tc_edt.sh \
test_xdping.sh \
- test_bpftool_build.sh
+ test_bpftool_build.sh \
+ test_bpftool.sh
TEST_PROGS_EXTENDED := with_addr.sh \
with_tunnels.sh \
diff --git a/tools/testing/selftests/bpf/test_bpftool.py b/tools/testing/selftests/bpf/test_bpftool.py
new file mode 100644
index 000000000000..7f545feaec98
--- /dev/null
+++ b/tools/testing/selftests/bpf/test_bpftool.py
@@ -0,0 +1,228 @@
+# Copyright (c) 2020 SUSE LLC.
+#
+# This software is licensed under the GNU General License Version 2,
+# June 1991 as shown in the file COPYING in the top-level directory of this
+# source tree.
+#
+# THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS"
+# WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING,
+# BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
+# FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE
+# OF THE PROGRAM IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME
+# THE COST OF ALL NECESSARY SERVICING, REPAIR OR CORRECTION.

SPDX tag instead of boilerplate?

+
+import collections
+import functools
+import json
+import os
+import socket
+import subprocess
+import unittest
+
+
+# Add the source tree of bpftool and /usr/local/sbin to PATH
+cur_dir = os.path.dirname(os.path.realpath(__file__))
+bpftool_dir = os.path.abspath(os.path.join(cur_dir, "..", "..", "..", "..",
+ "tools", "bpf", "bpftool"))
+os.environ["PATH"] = bpftool_dir + ":/usr/local/sbin:" + os.environ["PATH"]
+
+# Probe sections
+SECTION_SYSTEM_CONFIG_PATTERN = b"Scanning system configuration..."
+SECTION_SYSCALL_CONFIG_PATTERN = b"Scanning system call availability..."
+SECTION_PROGRAM_TYPES_PATTERN = b"Scanning eBPF program types..."
+SECTION_MAP_TYPES_PATTERN = b"Scanning eBPF map types..."
+SECTION_HELPERS_PATTERN = b"Scanning eBPF helper functions..."
+SECTION_MISC_PATTERN = b"Scanning miscellaneous eBPF features..."
+
+
+class IfaceNotFoundError(Exception):
+ pass
+
+
+class UnprivilegedUserError(Exception):
+ pass
+
+
+def _bpftool(args, json=True):
+ _args = ["bpftool"]
+ if json:
+ _args.append("-j")
+ _args.extend(args)
+
+ res = subprocess.run(_args, capture_output=True)
+ return res.stdout
+
+
+def bpftool(args):
+ return _bpftool(args, json=False)
+
+
+def bpftool_json(args):
+ res = _bpftool(args)
+ return json.loads(res)
+
+
+def get_default_iface():
+ for iface in socket.if_nameindex():
+ if iface[1] != "lo":
+ return iface[1]
+ raise IfaceNotFoundError("Could not find any network interface to probe")
+
+
+def default_iface(f):
+ @functools.wraps(f)
+ def wrapper(*args, **kwargs):
+ iface = get_default_iface()
+ return f(*args, iface, **kwargs)
+ return wrapper
+
+
+class TestBpftool(unittest.TestCase):
+ @classmethod
+ def setUpClass(cls):
+ if os.getuid() != 0:
+ raise UnprivilegedUserError("This test suite eeeds root privileges")

Typo: eeeds

+
+ def _assert_pattern_not_in_dict(self, dct, pattern, check_keys=False):
+ """Check if all string values inside dictionary do not containe the

Typo: containe

+ given pattern.
+ """
+ for key, value in dct.items():
+ if check_keys:
+ self.assertNotIn(pattern, key)
+ if isinstance(value, dict):
+ self._assert_pattern_not_in_dict(value, pattern,
+ check_keys=True)
+ elif isinstance(value, str):
+ self.assertNotIn(pattern, value)
+
+ @default_iface
+ def test_feature_dev(self, iface):
+ expected_patterns = [
+ SECTION_SYSCALL_CONFIG_PATTERN,
+ SECTION_PROGRAM_TYPES_PATTERN,
+ SECTION_MAP_TYPES_PATTERN,
+ SECTION_HELPERS_PATTERN,
+ SECTION_MISC_PATTERN,
+ ]

Mixed feeling on the tests with plain output, as we keep telling people that plain output should not be parsed (not reliable, may change). But if you want to run one or two tests with it, why not, I guess.

+ unexpected_patterns = [
+ b"bpf_trace_printk",
+ b"bpf_probe_write_user",
+ ]
+
+ res = bpftool(["feature", "probe", "dev", iface])
+ for pattern in expected_patterns:
+ self.assertIn(pattern, res)
+ for pattern in unexpected_patterns:
+ self.assertNotIn(pattern, res)
+
+ @default_iface
+ def test_feature_dev_json(self, iface):
+ expected_keys = [
+ "syscall_config",
+ "program_types",
+ "map_types",
+ "helpers",
+ "misc",
+ ]
+ unexpected_values = [
+ "bpf_trace_printk",
+ "bpf_probe_write_user",
+ ]
+
+ res = bpftool_json(["feature", "probe", "dev", iface])
+ self.assertCountEqual(res.keys(), expected_keys)
+ for value in unexpected_values:
+ self._assert_pattern_not_in_dict(res, value)
+
+ def test_feature_kernel(self):
+ expected_patterns = [
+ SECTION_SYSTEM_CONFIG_PATTERN,
+ SECTION_SYSCALL_CONFIG_PATTERN,
+ SECTION_PROGRAM_TYPES_PATTERN,
+ SECTION_MAP_TYPES_PATTERN,
+ SECTION_HELPERS_PATTERN,
+ SECTION_MISC_PATTERN,
+ ]
+ unexpected_patterns = [
+ b"bpf_trace_printk",
+ b"bpf_probe_write_user",
+ ]
+
+ res_default1 = bpftool(["feature"])
+ res_default2 = bpftool(["feature", "probe"])
+ res = bpftool(["feature", "probe", "kernel"])
+
+ for pattern in expected_patterns:
+ self.assertIn(pattern, res_default1)
+ self.assertIn(pattern, res_default2)
+ self.assertIn(pattern, res)
+ for pattern in unexpected_patterns:
+ self.assertNotIn(pattern, res_default1)
+ self.assertNotIn(pattern, res_default2)
+ self.assertNotIn(pattern, res)
+
+ def test_feature_kernel_full(self):
+ expected_patterns = [
+ SECTION_SYSTEM_CONFIG_PATTERN,
+ SECTION_SYSCALL_CONFIG_PATTERN,
+ SECTION_PROGRAM_TYPES_PATTERN,
+ SECTION_MAP_TYPES_PATTERN,
+ SECTION_HELPERS_PATTERN,
+ SECTION_MISC_PATTERN,
+ b"bpf_trace_printk",
+ b"bpf_probe_write_user",
+ ]

However, if you do just one test for "kernel full", please favour JSON over plain output.

+
+ res_default = bpftool(["feature", "probe", "full"])
+ res = bpftool(["feature", "probe", "kernel", "full"])
+
+ for pattern in expected_patterns:
+ self.assertIn(pattern, res_default)
+ self.assertIn(pattern, res)
+
+ def test_feature_kernel_json(self):
+ expected_keys = [
+ "system_config",
+ "syscall_config",
+ "program_types",
+ "map_types",
+ "helpers",
+ "misc",
+ ]
+ unexpected_values = [
+ "bpf_trace_printk",
+ "bpf_probe_write_user",
+ ]
+
+ res_default1 = bpftool_json(["feature"])
+ self.assertCountEqual(res_default1.keys(), expected_keys)
+ for value in unexpected_values:
+ self._assert_pattern_not_in_dict(res_default1, value)
+
+ res_default2 = bpftool_json(["feature", "probe"])
+ self.assertCountEqual(res_default2.keys(), expected_keys)
+ for value in unexpected_values:
+ self._assert_pattern_not_in_dict(res_default2, value)
+
+ res = bpftool_json(["feature", "probe", "kernel"])
+ self.assertCountEqual(res.keys(), expected_keys)
+ for value in unexpected_values:
+ self._assert_pattern_not_in_dict(res, value)
+
+ def test_feature_macros(self):
+ expected_patterns = [
+ b"/\*\*\* System call availability \*\*\*/",
+ b"#define HAVE_BPF_SYSCALL",
+ b"/\*\*\* eBPF program types \*\*\*/",
+ b"#define HAVE.*PROG_TYPE",
+ b"/\*\*\* eBPF map types \*\*\*/",
+ b"#define HAVE.*MAP_TYPE",
+ b"/\*\*\* eBPF helper functions \*\*\*/",
+ b"#define HAVE.*HELPER",
+ b"/\*\*\* eBPF misc features \*\*\*/",
+ ]
+
+ res = bpftool(["feature", "probe", "macros"])
+ for pattern in expected_patterns:
+ self.assertRegex(res, pattern)

Could we have (or did I miss it?) a test that compares the output of probes _with_ "full" and _without_ it, to make sure that the only lines that differ are about "bpf_trace_prink" or "bpf_probe_write_user"? Could help determine if we filter out too many elements by mistake.

Thanks,
Quentin

diff --git a/tools/testing/selftests/bpf/test_bpftool.sh b/tools/testing/selftests/bpf/test_bpftool.sh
new file mode 100755
index 000000000000..66690778e36d
--- /dev/null
+++ b/tools/testing/selftests/bpf/test_bpftool.sh
@@ -0,0 +1,5 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2020 SUSE LLC.
+
+python3 -m unittest -v test_bpftool.TestBpftool