From mboxrd@z Thu Jan 1 00:00:00 1970 From: Antonio Quartulli Subject: Re: Re: [B.A.T.M.A.N.] [PATCH 6/7] batman-adv: Allow to use rntl_link for device creation/deletion Date: Sat, 1 Dec 2012 14:28:02 +0100 Message-ID: <20121201132802.GM24115@ritirata.org> References: <1353715332-4284-1-git-send-email-sven@narfation.org> <1353715332-4284-6-git-send-email-sven@narfation.org> <20121201131108.GL24115@ritirata.org> <15221362.NFC3M7kpXE@sven-laptop.home.narfation.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="pFej7zHSL6C5fFIz" Cc: The list for a Better Approach To Mobile Ad-hoc Networking , netdev@vger.kernel.org To: Sven Eckelmann Return-path: Received: from latitanza.investici.org ([82.94.249.234]:44596 "EHLO latitanza.investici.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751796Ab2LAN2x (ORCPT ); Sat, 1 Dec 2012 08:28:53 -0500 Content-Disposition: inline In-Reply-To: <15221362.NFC3M7kpXE@sven-laptop.home.narfation.org> Sender: netdev-owner@vger.kernel.org List-ID: --pFej7zHSL6C5fFIz Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sat, Dec 01, 2012 at 02:16:38PM +0100, Sven Eckelmann wrote: > Please don't remove netdev from this discussion without a reason. >=20 > On Saturday 01 December 2012 14:11:08 Antonio Quartulli wrote: > > On Sat, Nov 24, 2012 at 01:02:11AM +0100, Sven Eckelmann wrote: > > > The sysfs configuration interface of batman-adv to add/remove > > > soft-interfaces is not deadlock free and doesn't follow the currently > > > common way to create new virtual interfaces. > > >=20 > > > An additional interface though rtnl_link is introduced which provides= easy > > > device creation/deletion with tools like "ip": > > >=20 > > > $ ip link add dev bat0 type batadv > > > $ ip link del dev bat0 > > >=20 > > > Signed-off-by: Sven Eckelmann > >=20 > > Hello Sven, > >=20 > > why are we adding yet another API? What's the purpose? Is this intended= to > > fix the lock bug we get while using sysfs? >=20 > Because this is the normal way to create virtual network devices (please = feel=20 > free to correct me). Well, I've seen different iface types using many tools, e.g. vconfig, tunct= l, brctl.. Not that this justifies the fact that we should do the same (imho having a standard and unified way for creating interfaces would be the best option). But, to be honest, I think it should better discuss how to entirely moving/changing the existent API to a "better one" or to a "new one", inste= ad of starting to maintain two of them from now on with no plan, don't you think = so? > And no, it doesn't fix the sysfs problem because the=20 > sysfs stuff isn't removed. Yeah, I had the same in mind, that's why I was asking for the purpose of th= is. Cheers, --=20 Antonio Quartulli =2E.each of us alone is worth nothing.. Ernesto "Che" Guevara --pFej7zHSL6C5fFIz Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.19 (GNU/Linux) iQIcBAEBAgAGBQJQugXiAAoJEADl0hg6qKeO6iMQALALBgFLW3f6uF8P4IWmx3dQ dSko9VHkHq4aceK5kj2U73T0WrPUdfWlRibth0xMOY9Z0XhugqdkPBaTG5huacca ZeJFXI7zpnCt39vd/HpjMgwaDxYJvJKl2rldGWgmiwtFLVHglaVVFdALqcwKCYEw Hfm49asMp9mcsUCVNO/rUrmPAWeGkngk5On5WtH/bh4FDNm9+sFcrYAZ08C334wM Rlgkxgx9eb1TtsqNBqyxyotgb5paHfcarb861n3iVH568WGyLOPtIsQHVA3ftXcW XUCuOWIONRT1bF/UfJT6mCvj4Vq6Qwcl94x4+r1/AyYNR7LyVr+I+MzplgY6bM3G sUtd3teMWRugaegX0S3Y1qhiLIMWQFxCJYPaafmJ+eRcy9/yFLlLEyimtOo9HJoU g55xU2c3ghoNL7gHJXWflmt27GW9zNYFWqJ/NwB2NhG+9LIoNuYywL/iJUihNOlb tSAq3Lv06npitKahuKsF0xsjuDBeNuINewNZLIx6ylOotNUqrKuafiKbJMjAbd8e dsIYHWKkYY45gbKJQ+VnojKraiiToMvlp3xwlqpL5fVkTiDfYWDJpw9C1X46mbHy Qp+0f8LsNhSmeEb0a0akW6I0p9W9R5Zt6xOeubBsElfubO5a51lR6+RrP6lRW8B+ iJK5jvW9dEIrQrHwRD26 =7fEk -----END PGP SIGNATURE----- --pFej7zHSL6C5fFIz--