From mboxrd@z Thu Jan 1 00:00:00 1970 From: Joe Perches Subject: Re: [PATCH 6/8] net: gianfar: fix old-style declaration Date: Thu, 16 Jun 2016 11:11:45 -0700 Message-ID: <1466100705.19647.47.camel@perches.com> References: <20160616135020.3154461-1-arnd@arndb.de> <20160616135245.3198419-1-arnd@arndb.de> <20160616135245.3198419-6-arnd@arndb.de> <0d9320db-ae2c-03a2-eb20-e527a69c356b@cogentembedded.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: "David S. Miller" , Yangbo Lu , linux-kernel@vger.kernel.org To: Sergei Shtylyov , Arnd Bergmann , netdev@vger.kernel.org Return-path: In-Reply-To: <0d9320db-ae2c-03a2-eb20-e527a69c356b@cogentembedded.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Thu, 2016-06-16 at 21:02 +0300, Sergei Shtylyov wrote: > On 06/16/2016 04:52 PM, Arnd Bergmann wrote: > > Modern C standards expect the '__inline__' keyword to come before t= he return > > type in a declaration, and we get a warning for this with "make W=3D= 1": [] > > diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net= /ethernet/freescale/gianfar.c [] > > @@ -2275,7 +2275,7 @@ static inline void gfar_tx_checksum(struct sk= _buff *skb, struct txfcb *fcb, > > =A0 fcb->flags =3D flags; > > =A0} > >=20 > > -void inline gfar_tx_vlan(struct sk_buff *skb, struct txfcb *fcb) > > +static inline void gfar_tx_vlan(struct sk_buff *skb, struct txfcb = *fcb) > =A0=A0=A0=A0You don't mention making it *static*. Though the function= can be *static*=A0 > indeed... the current policy also forbids *inline* in the *.c files, = leaving=A0 > the judgement to gcc. While mostly true, (__always_inline vs inline vs nothing), there are many inline uses in this file that could be removed in a separate patch rather than removing just this one.