delorie.com/archives/browse.cgi   search  
Mail Archives: cygwin/1998/01/28/19:42:39

From: jeffdbREMOVETHIS AT netzone DOT com (Mikey)
Subject: Re: Problems with mingw32 GCC 2.8.0 !!!
28 Jan 1998 19:42:39 -0800 :
Message-ID: <34d09b2b.11419585.cygnus.gnu-win32@smtp.netzone.com>
References: <Pine DOT LNX DOT 3 DOT 96 DOT 980128204057 DOT 2538A-100000 AT zoo-station DOT student DOT utwente DOT nl>
Reply-To: jeffdbREMOVETHIS AT netzone DOT com
Mime-Version: 1.0
To: Jan-Jaap van der Heijden <janjaap AT Wit381304 DOT student DOT utwente DOT nl>,
gnu-win32 AT cygnus DOT com

Well let's get cygnus to change to -windows, -dll that way
you could use

gcc (for mingw32 console)
gcc -windows (for mingw32 GUI exe's)
gcc -dll (for mingw32 console dll's)
gcc -windows -dll (for mingw32 GUI dll's)
gcc -posix (for cygwin32 console)
gcc -posix -dll (for cygwin32 dll's)

Didn't Geoff say that B19 or B20 was going
to be native win32 with a posix extension?

Anyone can do this for themselves obviously,
but it's much easier to upgrade if you can get it into the
FSF sources.

CC="gcc -posix" ./configure --options (for unix software)
( or maybe setup autoconf to do it for us where necessary)
make cc=gcc                                            (for vc++ software)

It's pretty ridiculous ask developers to maintain ~40MB of extra
compiler tools, when you can do the same thing with
a few hundred bytes extra in the specs file ;^)

The first goal of any gcc port is to be able to replace the
native compiler. (in this case nmake/cl/link/lib/rc)
Which egcs-mingw32 is already pretty close to.

I would consider mingw32 a port, as opposed to cygwin32
which is actually an emulation library, able to replace
the posix subsystem on NT and add a posix subsystem
on 9x. (with a MUCH better one ;)

Not I hasten to add that I'm taking anything away from
cygnus, without that, we wouldn't have this ;^)
	Thanks Guys!!!

On Wed, 28 Jan 1998 20:57:10 +0000 (WET), you wrote:

>On Wed, 28 Jan 1998, Mumit Khan wrote:
>
>> jeffdbREMOVETHIS AT netzone DOT com (Mikey) writes:
>> > Jan, Colin, Mumit, Geoff, and all other developers/maintainers
>> > 
>> > using -mXXX is a very BAD idea for a gcc option.
>> > 
>> > -m is reserved for processor specifications
>> 
>> That's what the docs say as well, but Kenner has the final say as 
>> FSF's GCC maintainer. 
>> 
>> JJ: did Kenner say why he wanted it this way?
>
>Kenner himself changed "-windows" to "-mwindows". This was probably for
>the sake of compatibility with cygwin32.
>
>When I couldn't talk him out of that, I argued that "-dll" should be
>renamed to "-mdll" to be consistent . BTW: I have seen *.lreg files
>around  when linking a dll, so the compiler was confused with "-dl"
>(preform a debug dump). I never bothered to make it reproducable.
>
>Most -mXXX options deal with target hardware dependant switches, and
>I agree that neither "-mwindows" nor "-mdll" is one of those.
>
> JanJaap
>
>---
>With sufficient thrust, pigs fly just fine. However, this is not
>necessarily a good idea. It is hard to be sure where they are going
>to land, and it could be dangerous sitting under them as they fly
>overhead.  -- RFC1925.
>


=====================================================
Linux a platform built by, and for users, standing on
the firm legs of reliability, and speed.

Microsoft Windows, a platform without a leg to stand on.

(jeffdbREMOVETHIS AT netzone DOT com)
delete REMOVETHIS from the above to reply
         Mikey
-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request AT cygnus DOT com" with one line of text: "help".

- Raw text -


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