Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20220721180214.3223778-1-dlatypov@google.com> <20220721180214.3223778-4-dlatypov@google.com>
 <CABVgOSkHCDt4DPRTCpSgWUSa5_5adEoqwW1fNnXyev9fK1eiuQ@mail.gmail.com>
In-Reply-To: <CABVgOSkHCDt4DPRTCpSgWUSa5_5adEoqwW1fNnXyev9fK1eiuQ@mail.gmail.com>
From:   Daniel Latypov <dlatypov@google.com>
Date:   Fri, 22 Jul 2022 09:49:02 -0700
Message-ID: <CAGS_qxqV0Con7XL0P5XVvdJY73aSLWDu_WD+3DKn2UfD+=rsxQ@mail.gmail.com>
Subject: Re: [PATCH 4/4] kunit: make knuit_kfree() not segfault on invalid inputs
To:     David Gow <davidgow@google.com>
Cc:     Brendan Higgins <brendanhiggins@google.com>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        KUnit Development <kunit-dev@googlegroups.com>,
        "open list:KERNEL SELFTEST FRAMEWORK" 
        <linux-kselftest@vger.kernel.org>,
        Shuah Khan <skhan@linuxfoundation.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Fri, Jul 22, 2022 at 12:35 AM David Gow <davidgow@google.com> wrote:
> _Maybe_ we should no-op if ptr is NULL. I think it's legal for
> free()/kfree(), and while I don't see much use of it for kunit tests,
> maybe it'll save someone confusion down the road.
>
> But I could live with it either way...

That's a good point.
kfree(NULL) is indeed a no-op.

I can see someone writing a parameterized test w/ some code like
  char *buffer = NULL;
  if (param->use_buffer) buffer = kunit_kzalloc(test, 10, GFP_KERNEL);
  ...
  kunit_kfree(test, buffer);
and they'd have every reason to think this should just work.

I think I'll tack this on as an extra patch and send a v2 w/ the
commit subject for this one fixed.