Mailing-List: contact cygwin-apps-help AT sourceware DOT cygnus DOT com; run by ezmlm
Sender: cygwin-apps-owner AT sourceware DOT cygnus DOT com
List-Subscribe: <mailto:cygwin-apps-subscribe AT sources DOT redhat DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin-apps/>
List-Post: <mailto:cygwin-apps AT sources DOT redhat DOT com>
List-Help: <mailto:cygwin-apps-help AT sources DOT redhat DOT com>, <http://sources.redhat.com/lists.html#faqs>
Delivered-To: mailing list cygwin-apps AT sources DOT redhat DOT com
Date: Wed, 20 Dec 2000 23:36:35 -0500
From: Christopher Faylor <cgf AT redhat DOT com>
To: cygwin-apps AT cygwin DOT com
Subject: Re: Cygwin 1.1.4:  unexpected "make" behaviour
Message-ID: <20001220233635.A27245@redhat.com>
Reply-To: cygwin-apps AT cygwin DOT com
Mail-Followup-To: cygwin-apps AT cygwin DOT com
References: <20001221034647 DOT 26222 DOT qmail AT web118 DOT yahoomail DOT com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.3.11i
In-Reply-To: <20001221034647.26222.qmail@web118.yahoomail.com>; from earnie_boyd@yahoo.com on Wed, Dec 20, 2000 at 07:46:47PM -0800

On Wed, Dec 20, 2000 at 07:46:47PM -0800, Earnie Boyd wrote:
>--- Christopher Faylor <cgf AT redhat DOT com> wrote:
>> On Wed, Dec 20, 2000 at 04:48:41PM -0800, Sammartino, Ryan wrote:
>> >If you look at a slightly larger slice of the code
>> >than diff -c provides, you'll see that this whole thing
>> >is wrapped in a "#ifdef GCC_IS_NATIVE" directive, which is
>> >set near the top of default.c:
>> >
>> >/* Define GCC_IS_NATIVE if gcc is the native development environment on
>> >   your system (gcc/bison/flex vs cc/yacc/lex).  */
>> >/* CYGNUS LOCAL: or __CYGWIN__ */
>> >#if defined(__MSDOS__) || defined(__CYGWIN__)
>> >#define GCC_IS_NATIVE
>> >#endif
>> >
>> >
>> >so it does seem to be a "cygnus only" thing.
>> 
>> Sorry, I don't understand your logic.  If GCC_IS_NATIVE, surely g++
>> either be the default c++ compiler or someone had a reason for not
>> making that the case.  AFAICT, you are modifying a section of make that
>> is untouched by Cygnus/Red Hat.  It's under the control of a define
>> which can be set in non-Cygwin conditions.  Roland McGrath obviously had
>> a reason for doing this.  If it was wrong, then the make maintainer
>> (psmith AT gnu DOT org) should apprised of that fact.
>
>And possibly the correct solution is to remove the `|| defined(__CYGWIN__)' in
>this instance.  The probable case here is that back in version x a patch was
>submitted that took effect in version y so that there is no Cygwin specific
>change anymore.  It would be a Cygwin package maintainers job to see that such
>code patches are submitted back to the source maintainer.

If you are going to trust the cygwin package maintainer to make the
right decision, then, his decision is that the GNU maintainer of the
file in question should be notified.  As far as the comment in the file
is concerned, Cygwin is doing the right thing.

I am not particularly bothered by the default CXX and I don't see any
reason why I should be burdened with going to the effort of trying to
champion a fix that I have little interest in.  I don't have any special
relationship with the make maintainer which would enable me to receive
special consideration.

cgf