From owner-svn-src-all@FreeBSD.ORG Fri Jan 27 12:56:37 2012 Return-Path: Delivered-To: svn-src-all@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 3656C1065670; Fri, 27 Jan 2012 12:56:37 +0000 (UTC) (envelope-from luigi@onelab2.iet.unipi.it) Received: from onelab2.iet.unipi.it (onelab2.iet.unipi.it [131.114.59.238]) by mx1.freebsd.org (Postfix) with ESMTP id E5FC28FC17; Fri, 27 Jan 2012 12:56:36 +0000 (UTC) Received: by onelab2.iet.unipi.it (Postfix, from userid 275) id 0D8C57300A; Fri, 27 Jan 2012 13:58:44 +0100 (CET) Date: Fri, 27 Jan 2012 13:58:44 +0100 From: Luigi Rizzo To: Bruce Evans Message-ID: <20120127125844.GA11690@onelab2.iet.unipi.it> References: <201201260955.q0Q9tG1m075353@svn.freebsd.org> <4F21820B.8040000@rice.edu> <20120127203631.Q1604@besplex.bde.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20120127203631.Q1604@besplex.bde.org> User-Agent: Mutt/1.4.2.3i Cc: svn-src-head@FreeBSD.org, Luigi Rizzo , src-committers@FreeBSD.org, svn-src-all@FreeBSD.org, Alan Cox Subject: Re: svn commit: r230572 - in head/sys/dev: ixgbe netmap X-BeenThere: svn-src-all@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "SVN commit messages for the entire src tree \(except for " user" and " projects" \)" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 27 Jan 2012 12:56:37 -0000 On Fri, Jan 27, 2012 at 08:45:40PM +1100, Bruce Evans wrote: > On Thu, 26 Jan 2012, Alan Cox wrote: > > >On 01/26/2012 03:55, Luigi Rizzo wrote: > > >>Log: > >>... > >> Netmap-related changes for ixgbe: > >>... > >>+#define NKR_PENDINTR 0x1 // Pending interrupt. > >> u_int nkr_num_slots; > >> > >> int nkr_hwofs; /* offset between NIC and netmap ring */ > >> struct netmap_adapter *na; // debugging > >> struct selinfo si; /* poll/select wait queue */ > >>-}; > >>+} __attribute__((__aligned__(64))); > > > >The machine-dependent param.h defines CACHE_LINE_SIZE for use in > >situations like this. > > Also, the machine-independent cdefs.h defines __aligned() for use > in situations like this. All cases that use CACHE_LINE_SIZE in an > alignment statement spell the alignment statement correctly. The > only hard-coded __attribute__() in a line matching CACHE_LINE_SIZE > is for CVMX_CACHE_LINE_SIZE in contrib/octeon-sdk/cvmx-utils.h. > Perhaps contrib'ed code needs to use hard-coded gccisms instead of > hard-coded FreeBSDisms. thanks for the comments, I'll adjust this in some subsequent commit. cheers luigi