Received-SPF: softfail (google.com: domain of transitioning linux-kernel-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) client-ip=23.128.96.19;
Subject: Re: [PATCH 4/4] selftests/memfd: add tests for MFD_NOEXEC
To:     Daniel Verkamp <dverkamp@chromium.org>, linux-mm@kvack.org
Cc:     linux-kernel@vger.kernel.org, linux-kselftest@vger.kernel.org,
        Andrew Morton <akpm@linux-foundation.org>,
        Hugh Dickins <hughd@google.com>,
        Mattias Nissler <mnissler@chromium.org>,
        Dmitry Torokhov <dmitry.torokhov@gmail.com>,
        Kees Cook <keescook@chromium.org>,
        Shuah Khan <skhan@linuxfoundation.org>
References: <20220401220834.307660-1-dverkamp@chromium.org>
 <20220401220834.307660-5-dverkamp@chromium.org>
From:   Shuah Khan <skhan@linuxfoundation.org>
Message-ID: <c8114977-a757-60d0-d6b7-9d3f91568019@linuxfoundation.org>
Date:   Thu, 7 Apr 2022 14:03:18 -0600
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.8.1
MIME-Version: 1.0
In-Reply-To: <20220401220834.307660-5-dverkamp@chromium.org>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 4/1/22 4:08 PM, Daniel Verkamp wrote:
> Tests that ensure MFD_NOEXEC memfds have the appropriate mode bits and
> cannot be chmod-ed into being executable.
> 
> Signed-off-by: Daniel Verkamp <dverkamp@chromium.org>
> ---
>   tools/testing/selftests/memfd/memfd_test.c | 34 ++++++++++++++++++++++
>   1 file changed, 34 insertions(+)
> 
> diff --git a/tools/testing/selftests/memfd/memfd_test.c b/tools/testing/selftests/memfd/memfd_test.c
> index fdb0e46e9df9..a79567161cdf 100644
> --- a/tools/testing/selftests/memfd/memfd_test.c
> +++ b/tools/testing/selftests/memfd/memfd_test.c
> @@ -32,6 +32,10 @@
>   #define F_SEAL_EXEC	0x0020
>   #endif
>   
> +#ifndef MFD_NOEXEC
> +#define MFD_NOEXEC	0x0008U
> +#endif
> +
>   /*
>    * Default is not to test hugetlbfs
>    */
> @@ -959,6 +963,35 @@ static void test_seal_exec(void)
>   	close(fd);
>   }
>   
> +/*
> + * Test memfd_create with MFD_NOEXEC flag
> + * Test that MFD_NOEXEC applies F_SEAL_EXEC and prevents change of exec bits
> + */
> +static void test_noexec(void)
> +{
> +	int fd;
> +
> +	printf("%s NOEXEC\n", memfd_str);
> +
> +	/* Create with NOEXEC and ALLOW_SEALING */
> +	fd = mfd_assert_new("kern_memfd_noexec",
> +			    mfd_def_size,
> +			    MFD_CLOEXEC | MFD_ALLOW_SEALING | MFD_NOEXEC);

Don't we need to check fd here?

> +	mfd_assert_mode(fd, 0666);
> +	mfd_assert_has_seals(fd, F_SEAL_EXEC);
> +	mfd_fail_chmod(fd, 0777);
> +	close(fd);
> +
> +	/* Create with NOEXEC but without ALLOW_SEALING */
> +	fd = mfd_assert_new("kern_memfd_noexec",
> +			    mfd_def_size,
> +			    MFD_CLOEXEC | MFD_NOEXEC);

What happens when mfd_assert_new() fails - don't we need to check fd?

> +	mfd_assert_mode(fd, 0666);
> +	mfd_assert_has_seals(fd, F_SEAL_EXEC | F_SEAL_SEAL);
> +	mfd_fail_chmod(fd, 0777);
> +	close(fd);
> +}
> +
>   /*
>    * Test sharing via dup()
>    * Test that seals are shared between dupped FDs and they're all equal.
> @@ -1132,6 +1165,7 @@ int main(int argc, char **argv)
>   
>   	test_create();
>   	test_basic();
> +	test_noexec();
>   
>   	test_seal_write();
>   	test_seal_future_write();
> 

fd isn't checked in the other test F_SEAL_EXEC in the 3/4 patch.

thanks,
-- Shuah