From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jon Loeliger Subject: Re: [PATCH 8/9 V3] Add documentation for the new DTS language. Date: Thu, 02 Oct 2008 15:46:02 -0500 Message-ID: <1222980362.9242.25.camel@ld0161-tx32> References: <48E396A3.809@freescale.com> <20081002011800.GI25598@yookeroo.seuss> <20081002152242.GB22258@ld0162-tx32.am.freescale.net> <20081002.135004.1723231860.imp@bsdimp.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20081002.135004.1723231860.imp-uzTCJ5RojNnQT0dZR+AlfA@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: devicetree-discuss-bounces+gldd-devicetree-discuss=m.gmane.org-mnsaURCQ41sdnm+yROfE0A@public.gmane.org Errors-To: devicetree-discuss-bounces+gldd-devicetree-discuss=m.gmane.org-mnsaURCQ41sdnm+yROfE0A@public.gmane.org To: "M. Warner Losh" Cc: Scott Wood , devicetree-discuss List-Id: devicetree@vger.kernel.org On Thu, 2008-10-02 at 13:50 -0600, M. Warner Losh wrote: > In message: <20081002152242.GB22258-VKaLA/mbEU932VTgPCOETVjVikpgYyvb5NbjCUgZEJk@public.gmane.org> > Scott Wood writes: > : On Thu, Oct 02, 2008 at 11:18:00AM +1000, David Gibson wrote: > : > > I vote against anything similar to the C preprocessor. > : > > : > Why? > : > : It's not nearly as powerful as the semantic approach, and it is > : syntactically clumsy (I do *not* want to have to put a backslash at the > : end of every line within a macro, for example). > > And there are no looping constructs at all in CPP. You flat out can't > do a generalized for loop with the macro language, for example. Its > symbols also do not map 1-1/onto symbols used by the dts files. It > can only grok macros you define for it, so it has no way of knowing > many useful things that dts just knows. > > Basically, it is using a hammer to pound screws in. Sure, it seems to > work, but the results are shoddy. > > Warner Also, it might be worth noting that if one uses "-I dts -O dts" with my patch set, it is pretty-darn similar behavior to a simple pre-processing pass that simply expands the original constructs into a flat DTS file like we have today. jdl