1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
|
# Copyright (c) 2026 Oracle and/or its affiliates
# SPDX-License-Identifier: LGPL-2.1-or-later
from _drgn_util.platform import NORMALIZED_MACHINE_NAME
import drgn.commands._crash # noqa # needed to register crash commands
from drgn.helpers.linux.cpumask import for_each_online_cpu
from tests.linux_kernel import skip_unless_have_stack_tracing
from tests.linux_kernel.vmcore import LinuxVMCoreCrashCommandTestCase
@skip_unless_have_stack_tracing
class TestBtVMCore(LinuxVMCoreCrashCommandTestCase):
# When CPU 0 panics on s390x, the kernel switches to a different stack that
# we don't know how to unwind through. This is a bug. For now,
# vmtest.enter_kdump tries to avoid this so we can get some test coverage,
# and we skip the tests otherwise.
def _skip_if_cpu0_on_s390x(self):
if NORMALIZED_MACHINE_NAME == "s390x" and self.prog["panic_cpu"].counter == 0:
self.skipTest("drgn can't unwind s390x panic stack on CPU 0")
def test_cpu(self):
self._skip_if_cpu0_on_s390x()
cmd = self.check_crash_command("bt -c 0")
self.assertIn("CPU: 0", cmd.stdout)
self.assertNotIn("(active)", cmd.stdout)
def test_panic(self):
self._skip_if_cpu0_on_s390x()
tid = self.prog.crashed_thread().tid
for cmd in ("bt", "bt -p"):
with self.subTest(cmd=cmd):
res = self.check_crash_command(cmd)
self.assertIn("drgn_test_crash_func", res.stdout)
self.assertIn(f"PID: {tid}", res.stdout)
self.assertNotIn("(active)", res.stdout)
def test_all(self):
self._skip_if_cpu0_on_s390x()
cmd = self.check_crash_command("bt -a")
self.assertNotIn("(active)", cmd.stdout)
for cpu in for_each_online_cpu(self.prog):
self.assertIn(f"CPU: {cpu}", cmd.stdout)
|
