From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jay Vosburgh Subject: Re: [PATCH v2 net-next 5/9] bonding: create a macro for bond mode and use it Date: Thu, 15 May 2014 11:32:41 -0700 Message-ID: <10534.1400178761@localhost.localdomain> References: <1400156967-8231-1-git-send-email-vfalico@gmail.com> <1400156967-8231-6-git-send-email-vfalico@gmail.com> <10028.1400176311@localhost.localdomain> <20140515181436.GA25181@mikrodark.usersys.redhat.com> Cc: netdev@vger.kernel.org, Andy Gospodarek To: Veaceslav Falico Return-path: Received: from youngberry.canonical.com ([91.189.89.112]:40879 "EHLO youngberry.canonical.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755797AbaEOScp (ORCPT ); Thu, 15 May 2014 14:32:45 -0400 In-reply-to: <20140515181436.GA25181@mikrodark.usersys.redhat.com> Sender: netdev-owner@vger.kernel.org List-ID: Veaceslav Falico wrote: >On Thu, May 15, 2014 at 10:51:51AM -0700, Jay Vosburgh wrote: >>Veaceslav Falico wrote: >> >>>CC: Jay Vosburgh >>>CC: Andy Gospodarek >>>Signed-off-by: Veaceslav Falico >>>--- >...snip... >> Would it be better to use "bond_uses_primary(struct bonding *)" >>instead of the above? That would simplify the above calling pattern, >>and shorten the calls elsewhere. Maybe I missed one, but it looks like >>all of the calls to _uses_primary have BOND_MODE(bond) as the argument. > >There's at least one call, when checking the params, where it checks the >int: > >bond_main.c: >4271 if (primary && !bond_mode_uses_primary(bond_mode)) { > >so, either we'll use something else here, or leave it with BOND_MODE()... The something else isn't so bad, and would look better for most callers: static inline bool bond_uses_primary(struct bonding *bond) { return bond_mode_uses_primary(bond->params.mode); } This ends up replacing "USES_PRIMARY(bond->params.mode)" with "bond_uses_primary(bond)" for most call sites, which actually looks like an improvement (it's even shorter). -J --- -Jay Vosburgh, jay.vosburgh@canonical.com