delorie.com/archives/browse.cgi   search  
Mail Archives: cygwin/2001/03/30/19:43:26

Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT sources DOT redhat DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin/>
List-Post: <mailto:cygwin AT sources DOT redhat DOT com>
List-Help: <mailto:cygwin-help AT sources DOT redhat DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender: cygwin-owner AT sources DOT redhat DOT com
Delivered-To: mailing list cygwin AT sources DOT redhat DOT com
Date: Fri, 30 Mar 2001 15:39:56 -0500
From: Christopher Faylor <cgf AT redhat DOT com>
To: cygwin AT cygwin DOT com
Subject: Re: -mno-cygwin
Message-ID: <20010330153956.C30299@redhat.com>
Reply-To: cygwin AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
References: <044601c0b929$91291400$0200a8c0 AT lifelesswks> <20010330110640 DOT B12883 AT redhat DOT com> <3AC4C0C3 DOT 6C3DD7F9 AT yahoo DOT com> <20010330133443 DOT A29516 AT redhat DOT com> <3AC4E5F9 DOT E84767B8 AT yahoo DOT com>
Mime-Version: 1.0
User-Agent: Mutt/1.3.11i
In-Reply-To: <3AC4E5F9.E84767B8@yahoo.com>; from earnie_boyd@yahoo.com on Fri, Mar 30, 2001 at 03:00:57PM -0500

On Fri, Mar 30, 2001 at 03:00:57PM -0500, Earnie Boyd wrote:
>> Except what's a "msvcrt" or a "crtdll"?  They're just strings of characters.
>> 
>> Expecting people to know about these DLLs or any package name other than
>> cygwin or (maybe) unix seems like it would still be subject to confusion.
>
>I expect that know one knows anything except perhaps the creator and
>that can be said to be questionable?! ;^)  Why doesn't -mno-cygwin mean
>that no runtime is supplied at all and I have to supply my own?  As you
>can see "no-cygwin" is "just a string of characters".  It has no more
>meaning than msvcrt or crtdll and is in fact less descriptive.

Presumably, if someone downloaded the Cygwin package they would be able
to infer that an option with "no" and "cygwin" in the name might, just
possibly, produce an executable that doesn't rely on cygwin.  So, if
they stumbled across this option in some mailing list discussion or
other they might stand a chance of doing the right thing.

The original proposal was that we have a -mmingw switch.  That would
presume that a person who wanted to produce a binary that didn't rely
on cygwin would find -mmingw more intuitive.  I don't think that a
naive user who stumbles upon a switch named "-mmingw" is going to
be more apt to think "Aha!  That must produce native windows apps!"

This applies to -mmsvcrt and -mcrtdll as well.  What's a msv?  What's
a crt?  Why would I care?

If you are proposing that it would be a nice convenience to have these
options for people who know what they mean, I certainly wouldn't argue.
I just don't see how you can assert that they would make things easier
to understand.

The bottom line is that, IMO, if you have to do research to figure out
the right option to use, then it really doesn't matter all that much
what the option is called.  Given this, IMO, the one thing that does
make sense, is to stick with what has historically been used.  For this
reason, I think that nuking -mno-cygwin is apt to cause more confusion
than it saves.

cgf

P.S. Btw, -mcygwin is a valid switch.

--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

- Raw text -


  webmaster     delorie software   privacy  
  Copyright © 2019   by DJ Delorie     Updated Jul 2019