From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gdb-patches-return-51084-listarch-gdb-patches=sources.redhat.com@sourceware.org>
Received: (qmail 31021 invoked by alias); 5 Oct 2007 17:52:45 -0000
Received: (qmail 31012 invoked by uid 22791); 5 Oct 2007 17:52:44 -0000
X-Spam-Check-By: sourceware.org
Received: from mail.codesourcery.com (HELO mail.codesourcery.com) (65.74.133.4)     by sourceware.org (qpsmtpd/0.31) with ESMTP; Fri, 05 Oct 2007 17:52:42 +0000
Received: (qmail 15272 invoked from network); 5 Oct 2007 17:52:40 -0000
Received: from unknown (HELO localhost) (jimb@127.0.0.2)   by mail.codesourcery.com with ESMTPA; 5 Oct 2007 17:52:40 -0000
To: gdb-patches@sourceware.org
Subject: Re: [rfc] Pre-parse XML target descriptions
References: <20071005153350.GA23583@caradoc.them.org> 	<m3ir5llcxu.fsf@codesourcery.com> 	<20071005174009.GA3182@caradoc.them.org>
From: Jim Blandy <jimb@codesourcery.com>
Date: Fri, 05 Oct 2007 17:52:00 -0000
In-Reply-To: <20071005174009.GA3182@caradoc.them.org> (Daniel Jacobowitz's message of "Fri, 5 Oct 2007 13:40:09 -0400")
Message-ID: <m38x6hzbx4.fsf@codesourcery.com>
User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.0.50 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-IsSubscribed: yes
Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm
Precedence: bulk
List-Id: <gdb-patches.sourceware.org>
List-Subscribe: <mailto:gdb-patches-subscribe@sourceware.org>
List-Archive: <http://sourceware.org/ml/gdb-patches/>
List-Post: <mailto:gdb-patches@sourceware.org>
List-Help: <mailto:gdb-patches-help@sourceware.org>, <http://sourceware.org/ml/#faqs>
Sender: gdb-patches-owner@sourceware.org
X-SW-Source: 2007-10/txt/msg00062.txt.bz2


Daniel Jacobowitz <drow at false.org> writes:
> On Fri, Oct 05, 2007 at 09:54:21AM -0700, Jim Blandy wrote:
>> Actually, the mechanism I suggested could be incorporated in any
>> case.  The GDB build process should always check the checksums of the
>> XML input files against those stored in the files built from them,
>> simply to remind developers when they need to regenerate the C files.
>> It's just a good general principle for dealing with pre-generated
>> files.
>
> But honestly, is it worth the trouble?  If we do that we have to
> content with at least these issues:
>
>   - $build != $host; can not run GDB at all
>   - Expat not available; can not parse descriptions
>   - GDB needs the descriptions to build but can't regenerate them
>     until after it is built.
>
> I don't think it's important enough to make the build system
> more complicated.

I think you're imagining something more complicated than I had in
mind.  This is all I meant:

gdb: check-xml-regs

check-xml-regs: foo.check-xml-regs-stamp bar.check-xml-regs-stamp

%.check-xml-regs-stamp: %.xml %.c
        stamp=$$(sed -ne 's|/. stamp: \(.*\) ./|\1|p' < $*.c); \
        if [ "$$(md5sum $*.xml)" = "$$stamp" ]; then \
           touch $@; \
        else \
           echo '.c file must be regenerated from .xml: $*.c' >&2; \
           exit 1; \
        fi