Received-SPF: pass (google.com: domain of linux-kernel+bounces-150561-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Precedence: bulk
MIME-Version: 1.0
References: <20240418131754.58217-3-wander@redhat.com> <9ff84256-c7d2-48e5-b06b-09a993db2c39@web.de>
In-Reply-To: <9ff84256-c7d2-48e5-b06b-09a993db2c39@web.de>
From: Wander Lairson Costa <wander@redhat.com>
Date: Thu, 18 Apr 2024 14:18:59 -0300
Message-ID: <CAAq0SUk7nJ80rrMG+zeVi-XReVJ7sdWTAT9fmi4E+LpYAYa=OA@mail.gmail.com>
Subject: Re: [PATCH 2/2] kunit: avoid memory leak on device register error
To: Markus Elfring <Markus.Elfring@web.de>
Cc: kunit-dev@googlegroups.com, linux-kselftest@vger.kernel.org, 
	kernel-janitors@vger.kernel.org, Brendan Higgins <brendan.higgins@linux.dev>, 
	David Gow <davidgow@google.com>, Rae Moar <rmoar@google.com>, 
	LKML <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Thu, Apr 18, 2024 at 12:24=E2=80=AFPM Markus Elfring <Markus.Elfring@web=
de> wrote:
>
> > If the device register fails, free the allocated memory before
> > returning.
>
> * I suggest to use the word =E2=80=9Cregistration=E2=80=9D (instead of =
=E2=80=9Cregister=E2=80=9D)
>   in the commit message.
>
> * Would you like to add the tag =E2=80=9CFixes=E2=80=9D accordingly?
>
>
> > +++ b/lib/kunit/device.c
> > @@ -131,6 +131,7 @@ static struct kunit_device *kunit_device_register_i=
nternal(struct kunit *test,
> >       err =3D device_register(&kunit_dev->dev);
> >       if (err) {
> >               put_device(&kunit_dev->dev);
> > +             kfree(kunit_dev);
> >               return ERR_PTR(err);
> >       }
>
> Common error handling code can be used instead
> if an additional label would be applied for a corresponding jump target.
>
> How do you think about to increase the application of scope-based resourc=
e management here?
>

I thought about that. But I think the code is simple enough (for now)
to not require an exit label.

> Regards,
> Markus
>