Received-SPF: pass (google.com: domain of linux-kernel+bounces-182958-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Date: Sat, 18 May 2024 13:36:41 -0700
From: Kees Cook <keescook@chromium.org>
To: Alexandre Belloni <alexandre.belloni@bootlin.com>
Cc: Shuah Khan <shuah@kernel.org>, Masahiro Yamada <masahiroy@kernel.org>,
	linux-rtc@vger.kernel.org, linux-kselftest@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-hardening@vger.kernel.org
Subject: Re: [PATCH] selftests: rtc: rtctest: Do not open-code
 TEST_HARNESS_MAIN
Message-ID: <202405181335.00B6D2F@keescook>
References: <20240518001655.work.053-kees@kernel.org>
 <20240518202354d5422c77@mail.local>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20240518202354d5422c77@mail.local>

On Sat, May 18, 2024 at 10:23:54PM +0200, Alexandre Belloni wrote:
> On 17/05/2024 17:16:58-0700, Kees Cook wrote:
> > Argument processing is specific to the test harness code. Any optional
> > information needs to be passed via environment variables. Move alternate
> > path to the RTC_DEV environment variable. Also do not open-code
> > TEST_HARNESS_MAIN because its definition may change.
> 
> Th main issue doing that is that this breaks the main use case of
> rtctest as /dev/rtc1 is usually the main target for those tests. Having
> the RTC_DEV environment variable only documented n this commit message
> is definitively not enough, I'm going to have to handle zillion of
> complaints that this is not working anymore.

Hm, maybe switch the default to /dev/rtc1? Maybe there's a better way to
integrate arguments into a test runner. Right now the core harness code
is doing the argument parsing...

-- 
Kees Cook