linux-kernel - [RFC PATCH 0/2] KUnit device API proposal

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-ID: <20230325043104.3761770-1-davidgow@google.com>
Date:   Sat, 25 Mar 2023 12:31:02 +0800
From:   David Gow <davidgow@...gle.com>
To:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Matti Vaittinen <mazziesaccount@...il.com>,
        Maxime Ripard <maxime@...no.tech>,
        Brendan Higgins <brendan.higgins@...ux.dev>,
        Stephen Boyd <sboyd@...nel.org>,
        Shuah Khan <skhan@...uxfoundation.org>
Cc:     David Gow <davidgow@...gle.com>,
        "Rafael J . Wysocki" <rafael@...nel.org>,
        Andy Shevchenko <andriy.shevchenko@...ux.intel.com>,
        Heikki Krogerus <heikki.krogerus@...ux.intel.com>,
        Jonathan Cameron <jic23@...nel.org>,
        linux-kernel@...r.kernel.org, linux-kselftest@...r.kernel.org,
        kunit-dev@...glegroups.com
Subject: [RFC PATCH 0/2] KUnit device API proposal

Hi all,

This is a follow-up to the conversation[1] about adding helpers to create a
struct device for use in KUnit tests. At the moment, most tests are
using root_device_register(), which doesn't quite fit, and a few are
using platform_devices instead.

This adds a KUnit-specific equivalent: kunit_device_register(), which
creates a device which will be automatically cleaned up on test exit
(such as, for example, if an assertion fails).
It's also possible to unregister it earlier with
kunit_device_unregister().

This can replace the root_device_register() users pretty comfortably,
though doesn't resolve the issue with devm_ resources not being released
properly as laid out in [2]. Updating the implementation here to use a
'kunit' bus should, I think, be reasonably straightforward.

The first patch in the series is an in-progress implementation of a
separate new 'kunit_defer()' API, upon which this device implementation
is built.

If the overall idea seems good, I'll make sure to add better
tests/documentation, and patches converting existing tests to this API.

Cheers,
-- David

[1]: https://lore.kernel.org/linux-kselftest/bad670ee135391eb902bd34b8bcbe777afabc7fd.1679474247.git.mazziesaccount@gmail.com/
[2]: https://lore.kernel.org/linux-kselftest/20230324123157.bbwvfq4gsxnlnfwb@houat/

---
David Gow (2):
  kunit: resource: Add kunit_defer() functionality
  kunit: Add APIs for managing devices

 include/kunit/device.h   |  25 +++++++++
 include/kunit/resource.h |  87 +++++++++++++++++++++++++++++++
 lib/kunit/Makefile       |   1 +
 lib/kunit/device.c       |  68 ++++++++++++++++++++++++
 lib/kunit/resource.c     | 110 +++++++++++++++++++++++++++++++++++++++
 5 files changed, 291 insertions(+)
 create mode 100644 include/kunit/device.h
 create mode 100644 lib/kunit/device.c

-- 
2.40.0.348.gf938b09366-goog