Mail Archives: djgpp-workers/1996/10/14/13:04:25
Robert Hoehne wrote:
>
> I don't know exactly, what the reason for this idea was, but I think
> this technique is a good one. With this it is possible to use most
> of the UNIX makefiles without changing (in most parts), because they
> use the -o switch without suffix but we (on MS-DOS) need executables
> with .exe. In this way the problem is solved by updating a target
> without a suffix and getting an executable in one step.
>
> I'm realy not understand John what he has against this. Do you have
> so small disk space that you cannot get both files at the same time?
>
> Where is the problem if you have both files?
>
Disk space is not a problem. :) The problem is that there seemed to be
no way to tell djgpp to use a _different_ stubify than the standard when
it runs the linker. Because of the way my makefile for djverify was set
up, invoking it with 'djverify' (well, djvrfy2 now) alone as the target
causes it to spit out an image file _and_ an executable stubbed with the
WRONG stub. For this reason, I thought it would be better if it simply
did not produce an executable if not explicitly told to do so.
However, I have slowly and painfully eked out from the collective djgpp
wisdom that it is possible to tell the binutils which stub/stubify to
use to generate any given executable. So my original point, that there
was no way to either not produce an image or use a different stub,
turned out to be wrong. So, I rest my case, and thank everyone for
their help. :)
--
John M. Aldrich <fighteer AT cs DOT com>
* Anything that happens, happens.
* Anything that, in happening, causes something else to happen,
causes something else to happen.
* Anything that, in happening, causes itself to happen again, happens
again.
* It doesn't necessarily do it in chronological order, though.
--- Douglas Adams
- Raw text -