Linux Documentation
 help / color / mirror / Atom feed
From: Malte Wechter <maltewechter@gmail.com>
To: "Brendan Higgins" <brendan.higgins@linux.dev>,
	"David Gow" <david@davidgow.net>,
	"Rae Moar" <raemoar63@gmail.com>,
	"Miguel Ojeda" <ojeda@kernel.org>,
	"Boqun Feng" <boqun@kernel.org>, "Gary Guo" <gary@garyguo.net>,
	"Björn Roy Baron" <bjorn3_gh@protonmail.com>,
	"Benno Lossin" <lossin@kernel.org>,
	"Andreas Hindborg" <a.hindborg@kernel.org>,
	"Alice Ryhl" <aliceryhl@google.com>,
	"Trevor Gross" <tmgross@umich.edu>,
	"Danilo Krummrich" <dakr@kernel.org>,
	"Daniel Almeida" <daniel.almeida@collabora.com>,
	"Tamir Duberstein" <tamird@kernel.org>,
	"Alexandre Courbot" <acourbot@nvidia.com>,
	"Onur Özkan" <work@onurozkan.dev>,
	"Jonathan Corbet" <corbet@lwn.net>,
	"Shuah Khan" <skhan@linuxfoundation.org>,
	"Randy Dunlap" <rdunlap@infradead.org>
Cc: linux-kselftest@vger.kernel.org, kunit-dev@googlegroups.com,
	 linux-kernel@vger.kernel.org, rust-for-linux@vger.kernel.org,
	 workflows@vger.kernel.org, linux-doc@vger.kernel.org,
	 Malte Wechter <maltewechter@gmail.com>
Subject: [PATCH v3 4/4] Documentation: kunit: add documentation and examples to kunit documentation
Date: Tue, 13 Oct 2026 12:22:24 +0200	[thread overview]
Message-ID: <20261013-lockdep-kunit-v3-4-cdb8ff720c1f@gmail.com> (raw)
In-Reply-To: <20261013-lockdep-kunit-v3-0-cdb8ff720c1f@gmail.com>

add documentation section to Documentation/dev-tools/kunit
named "Failing Tests on Lockdep Reports and Kernel Warnings", which
explains the additional configurations namely:
CONFIG_KUNIT_FAIL_TEST_ON_LOCKDEP and CONFIG_KUNIT_FAIL_TEST_ON_WARN,
when and how they are useful as well as an example of an invocation.

Signed-off-by: Malte Wechter <maltewechter@gmail.com>
---
 .../dev-tools/kunit/failing_on_warnings.rst        | 78 ++++++++++++++++++++++
 Documentation/dev-tools/kunit/index.rst            |  1 +
 2 files changed, 79 insertions(+)

diff --git a/Documentation/dev-tools/kunit/failing_on_warnings.rst b/Documentation/dev-tools/kunit/failing_on_warnings.rst
new file mode 100644
index 000000000000..cf502995368b
--- /dev/null
+++ b/Documentation/dev-tools/kunit/failing_on_warnings.rst
@@ -0,0 +1,78 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+====================================================
+Failing Tests on Lockdep Reports and Kernel Warnings
+====================================================
+
+By default, a KUnit test only fails if one of its own expectations or
+assertions fails. If the code under test triggers a lockdep report or a
+``WARN()``, the problem is printed to the kernel log, but the test still
+passes and a bug is easily missed.
+
+KUnit provides two optional config options that turn these problems into
+test failures.
+
+Config options
+==============
+
+``CONFIG_KUNIT_FAIL_TEST_ON_LOCKDEP``
+	Fail a test case if lockdep reports a problem (and turns off
+	``debug_locks``) while the test runs, which is useful for any test
+	that takes locks. Locking bugs are reported by the test that
+	triggered them. Requires lockdep, e.g. ``CONFIG_PROVE_LOCKING=y``.
+
+``CONFIG_KUNIT_FAIL_TEST_ON_WARN``
+	Fail a test case if the kernel becomes tainted with ``TAINT_WARN``
+	while the test runs, i.e. if the test triggered a ``WARN()``. Since
+	a warning usually indicates a bug, this is useful for any test that
+	does not intentionally trigger a ``WARN()``. If you trigger a ``WARN()``
+	on purpose use ``kunit_warning_suppress(test)`` otherwise your test
+	will be marked as a failure.
+
+Forcefully skipped tests
+========================
+
+Once lockdep has reported a problem, ``debug_locks`` stays off, and the
+``TAINT_WARN`` taint cannot be cleared. The kernel is therefore considered
+unreliable for the rest of the run, so after the failing test, all
+remaining test cases are *forcefully skipped* instead of being run.
+
+A forcefully skipped test case is reported as a skip with a
+``force skip:`` reason, for example:
+
+.. code-block:: none
+
+	ok 2 foo_test # SKIP force skip: debug_locks are off
+
+Unlike an ordinary skip, a forced skip does not mean the test does not
+apply; it means the test could not be trusted to run. A suite in which
+at least one test case was forcefully skipped is reported as follows:
+
+- **Failed**, if a test case failed, and the remaining test cases were
+  forcefully skipped as a result. This is the common case.
+- **Passed**, if the forcefully skipped test cases only follow passing
+  ones. This is rare, and implies that something other than the test
+  cases themselves triggered the lockdep report or ``TAINT_WARN``.
+- **Forced skipped**, if all test cases in the suite were forcefully
+  skipped, typically because an earlier suite already left the kernel
+  unreliable. In the KTAP output, such a suite is reported as
+  ``not ok``.
+
+The suite statistics count forced skips separately as ``forced_skip``,
+and ``kunit.py`` shows such test cases as ``[FORCE SKIPPED]``.
+
+Running tests with the options enabled
+======================================
+
+The options can be added to a ``kunit.py`` run with ``--kconfig_add``,
+options CONFIG_KUNIT_FAIL_TEST_ON_WARN and CONFIG_KUNIT_FAIL_TEST_ON_LOCKDEP
+are independent of each other. Here is an example of running kunit tests with both configs:
+
+.. code-block:: bash
+
+	./tools/testing/kunit/kunit.py run \
+		--kconfig_add CONFIG_DEBUG_KERNEL=y \
+		--kconfig_add CONFIG_PROVE_LOCKING=y \
+		--kconfig_add CONFIG_KUNIT_FAIL_TEST_ON_WARN=y \
+		--kconfig_add CONFIG_KUNIT_FAIL_TEST_ON_LOCKDEP=y
+
diff --git a/Documentation/dev-tools/kunit/index.rst b/Documentation/dev-tools/kunit/index.rst
index b3593ae29ace..bdef2913da36 100644
--- a/Documentation/dev-tools/kunit/index.rst
+++ b/Documentation/dev-tools/kunit/index.rst
@@ -17,6 +17,7 @@ KUnit - Linux Kernel Unit Testing
 	style
 	faq
 	running_tips
+	failing_on_warnings
 
 This section details the kernel unit testing framework.
 

-- 
2.54.0


      parent reply	other threads:[~2026-10-13 10:22 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-13 10:22 [PATCH v3 0/4] kunit: add optional assertions to catch taint and lockdep warnings Malte Wechter
2026-10-13 10:22 ` [PATCH v3 1/4] kunit: add logic for tracking forcefully skipped test cases/suites Malte Wechter
2026-10-13 10:22 ` [PATCH v3 2/4] kunit: add assertions for lockdep inside of kunit Malte Wechter
2026-10-13 10:22 ` [PATCH v3 3/4] kunit: add assertions for TAINT_WARN " Malte Wechter
2026-10-13 10:22 ` Malte Wechter [this message]

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=20261013-lockdep-kunit-v3-4-cdb8ff720c1f@gmail.com \
    --to=maltewechter@gmail.com \
    --cc=a.hindborg@kernel.org \
    --cc=acourbot@nvidia.com \
    --cc=aliceryhl@google.com \
    --cc=bjorn3_gh@protonmail.com \
    --cc=boqun@kernel.org \
    --cc=brendan.higgins@linux.dev \
    --cc=corbet@lwn.net \
    --cc=dakr@kernel.org \
    --cc=daniel.almeida@collabora.com \
    --cc=david@davidgow.net \
    --cc=gary@garyguo.net \
    --cc=kunit-dev@googlegroups.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=lossin@kernel.org \
    --cc=ojeda@kernel.org \
    --cc=raemoar63@gmail.com \
    --cc=rdunlap@infradead.org \
    --cc=rust-for-linux@vger.kernel.org \
    --cc=skhan@linuxfoundation.org \
    --cc=tamird@kernel.org \
    --cc=tmgross@umich.edu \
    --cc=work@onurozkan.dev \
    --cc=workflows@vger.kernel.org \
    /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