From: fabio AT joplin DOT colorado DOT edu (Fabio Somenzi)
Subject: Re: Default linker output filename
9 Nov 1996 23:20:06 -0800
Sender: daemon AT cygnus DOT com
Approved: cygnus DOT gnu-win32 AT cygnus DOT com
Distribution: cygnus
Message-ID: <199611100554.WAA16073.cygnus.gnu-win32@joplin.colorado.edu>
Original-To: gnu-win32 AT cygnus DOT com
Original-Sender: owner-gnu-win32 AT cygnus DOT com

Here's is my 2 cents:

Not many makefiles generate a.out. What I see most of the times is
extensionless names. For instance, the GNU confugure scripts normally
generate conftest. I typically patch the configure script. It's just one
line that needs to be changed. Still, it would be nice if it were
automated.

Looking into the configure scripts, I've noticed that there is an option
whose name is somethng like "--program-suffix." I think it wouldn't be
difficult to set this automatically with an autoconf-generated test.
This would allow makefiles to be produced with the proper extensions.
And a.out would become a.out.exe if necessary.

This puts the burden on the shoulders of the maintainers of each
package. However, it is a rather light burden. I wouldn't mind making
the necessary changes to my autoconf files, or to my more traditional
makefiles. (Actually, I've already started doing it.)

If the .exe suffix were appended by the linker, the problem would only
be half-solved. For instance, that would turn real makefile targets into
phony ones. This would cause relinking even if the target is up-to-date.
Changing gmake to look for either foo or foo.exe seems a course fraught
with peril.

In summary, I'd vote for leaving things as they are. That is, a.out
remains a.out.

Fabio
--
Fabio Somenzi          | Phone: 303-492-3466
University of Colorado | Fax:   303-492-2758
ECE Dept.              | Email: Fabio AT Colorado DOT EDU
Boulder CO 80309-0425  | WWW:   http://vlsi.colorado.edu/~fabio
-
For help on using this list, send a message to
"gnu-win32-request AT cygnus DOT com" with one line of text: "help".