Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
In-Reply-To: <c3341eb3-623a-b5dd-e0d1-e7b9b6650932@oracle.com>
References: <20180313114205.27714-1-anders.roxell@linaro.org> <c3341eb3-623a-b5dd-e0d1-e7b9b6650932@oracle.com>
From:   Anders Roxell <anders.roxell@linaro.org>
Date:   Wed, 4 Apr 2018 13:36:18 +0200
Message-ID: <CADYN=9K-=7jOHGmUP+GT9uz7Nwcc801tam+6wf6hUjsKhGEndQ@mail.gmail.com>
Subject: Re: [PATCH 1/2] selftests/memfd/memfd_test.c: fix implicit declaration
To:     Mike Kravetz <mike.kravetz@oracle.com>
Cc:     shuah@kernel.org, linux-kselftest@vger.kernel.org,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        David Herrmann <dh.herrmann@gmail.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 14 March 2018 at 02:09, Mike Kravetz <mike.kravetz@oracle.com> wrote:
> On 03/13/2018 04:42 AM, Anders Roxell wrote:
>> gcc warns about implicit declaration.
>>
>> gcc -D_FILE_OFFSET_BITS=3D64 -I../../../../include/uapi/
>>     -I../../../../include/ -I../../../../usr/include/
>>     memfd_test.c common.o  -o memfd_test
>> memfd_test.c: In function =E2=80=98mfd_assert_get_seals=E2=80=99:
>> memfd_test.c:74:6: warning: implicit declaration of function =E2=80=98fc=
ntl=E2=80=99
>>     [-Wimplicit-function-declaration]
>>   r =3D fcntl(fd, F_GET_SEALS);
>>       ^~~~~
>> memfd_test.c: In function =E2=80=98mfd_assert_open=E2=80=99:
>> memfd_test.c:197:6: warning: implicit declaration of function =E2=80=98o=
pen=E2=80=99
>>     [-Wimplicit-function-declaration]
>>   r =3D open(buf, flags, mode);
>>       ^~~~
>> memfd_test.c: In function =E2=80=98mfd_assert_write=E2=80=99:
>> memfd_test.c:328:6: warning: implicit declaration of function =E2=80=98f=
allocate=E2=80=99
>>     [-Wimplicit-function-declaration]
>>   r =3D fallocate(fd,
>>       ^~~~~~~~~
>>
>> In the current code, we include the headers that the functions want
>> according to the man pages, and we add some defines that will be used if
>> they isn't found in glibc.  The defines was added into the kernel source
>> in kernel >=3D 3.16 and glibc requires kernel header files >=3D 3.2.
>>
>> Fixes: 4f5ce5e8d7e2 ("selftests: add memfd_create() + sealing tests")
>> Signed-off-by: Anders Roxell <anders.roxell@linaro.org>
>> ---
>>  tools/testing/selftests/memfd/memfd_test.c | 25 +++++++++++++++++++++++=
+-
>>  1 file changed, 24 insertions(+), 1 deletion(-)
>>
>> diff --git a/tools/testing/selftests/memfd/memfd_test.c b/tools/testing/=
selftests/memfd/memfd_test.c
>> index 10baa1652fc2..0dbeb29c094c 100644
>> --- a/tools/testing/selftests/memfd/memfd_test.c
>> +++ b/tools/testing/selftests/memfd/memfd_test.c
>> @@ -6,7 +6,6 @@
>>  #include <inttypes.h>
>>  #include <limits.h>
>>  #include <linux/falloc.h>
>> -#include <linux/fcntl.h>
>>  #include <linux/memfd.h>
>>  #include <sched.h>
>>  #include <stdio.h>
>> @@ -14,13 +13,37 @@
>>  #include <signal.h>
>>  #include <string.h>
>>  #include <sys/mman.h>
>> +#include <sys/types.h>
>>  #include <sys/stat.h>
>>  #include <sys/syscall.h>
>>  #include <sys/wait.h>
>> +#include <fcntl.h>
>>  #include <unistd.h>
>
> I suspect there is some guiding philosophy for selftests that I am
> unfamiliar with.  However, it seems that tests should use as much
> of the header files in the current kernel source tree as possible.
> This change removes the include of a header in the current source
> tree <linux/fcntl.h>.  It replaces that with the header <fcntl.h>
> from the host system (and some other changes).
>
> To me, this seems like step in the wrong direction.  But, I could
> be totally wrong and perhaps self tests should primarily target the
> host system header files.

So in a way I agree with you. However, what was the design decisions
internal kernel headers vs. headers from Host System ?
For me it isn't clear how/when we use them, its a mix today is it not?

To ignore warnings with -Wno-implicit-function-declaration wont be an
alternative I think.

Cheers,
Anders

> --
> Mike Kravetz