From mboxrd@z Thu Jan  1 00:00:00 1970
From: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Subject: Re: [PATCH] net/virtio-user: fix O_CLOEXEC undeclared
 error
Date: Wed, 29 Jun 2016 09:56:10 +0800
Message-ID: <20160629015610.GA23111@yliu-dev.sh.intel.com>
References: <1466948986-12255-1-git-send-email-jianfeng.tan@intel.com>
 <57724644.2070107@intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Cc: Jianfeng Tan <jianfeng.tan@intel.com>, dev@dpdk.org,
 huawei.xie@intel.com, Panu Matilainen <pmatilai@redhat.com>,
 Christian Ehrhardt <christian.ehrhardt@canonical.com>,
 Thomas Monjalon <thomas.monjalon@6wind.com>
To: Ferruh Yigit <ferruh.yigit@intel.com>
Return-path: <dev-bounces@dpdk.org>
Received: from mga01.intel.com (mga01.intel.com [192.55.52.88])
 by dpdk.org (Postfix) with ESMTP id 52A6F2904
 for <dev@dpdk.org>; Wed, 29 Jun 2016 03:55:53 +0200 (CEST)
Content-Disposition: inline
In-Reply-To: <57724644.2070107@intel.com>
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

On Tue, Jun 28, 2016 at 10:41:24AM +0100, Ferruh Yigit wrote:
> On 6/26/2016 2:49 PM, Jianfeng Tan wrote:
> > On some older systems, such as SUSE 11, the compiling error shows
> > as:
> >    .../dpdk/drivers/net/virtio/virtio_user/virtio_user_dev.c:67:22:
> >          error: =E2=80=98O_CLOEXEC=E2=80=99 undeclared (first use in =
this function)
> >=20
> > The fix is to declare _GNU_SOURCE macro before include fcntl.h.
> >=20
> > Fixes: 37a7eb2ae816 ("net/virtio-user: add device emulation layer")
> >=20
> > Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
> > ---
> >  drivers/net/virtio/virtio_user/virtio_user_dev.c | 1 +
> >  1 file changed, 1 insertion(+)
> >=20
> > diff --git a/drivers/net/virtio/virtio_user/virtio_user_dev.c b/drive=
rs/net/virtio/virtio_user/virtio_user_dev.c
> > index 3d12a32..180f824 100644
> > --- a/drivers/net/virtio/virtio_user/virtio_user_dev.c
> > +++ b/drivers/net/virtio/virtio_user/virtio_user_dev.c
> > @@ -31,6 +31,7 @@
> >   *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DA=
MAGE.
> >   */
> > =20
> > +#define _GNU_SOURCE
> This flag enables glibc extensions, and kind of problem on code
> portability. How big problem is it?

Honestly, I don't know. But I was also thinking that define _GNU_SOURCE
here is an overkill to me. For this issue, we may simply fix it by
reference the O_CLOEXEC flag only when it's defined.

> Is there any platform that DPDK runs on which uses a libc that isn't
> compatible with what this flag provides?
>=20
> And it seems this flag already used in DPDK, including libraries.

Yes, "git grep _GNU_SOURCE" shows a lot of such usage.

	--yliu

> If we
> are agree on using glibc extensions, does it make sense to move this
> flag into a single common location (like mk/exec-env/) ?
>=20
> >  #include <stdint.h>
> >  #include <stdio.h>
> >  #include <fcntl.h>
> >=20