From:	ROBERT_FINLEY AT ntsc DOT navy DOT mil
Date:	Tue, 05 May 98 10:19:38 EST
Message-Id:	<9804058943.AA894394660@CCMAIL.NTSC.NAVY.MIL>
To:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
Cc:	nate AT cartsys DOT com, djgpp AT delorie DOT com
Subject:	RE: Make "Clock Skew" problem.

My point was that it was the __DJGPP__ (Change 1) which was left out of the
original code that made the difference.  The original "Change 2" line ("...mtime
-1 > now...") worked sometimes to "handle" (whether rightly or wrongly) the
warning when Change 1 was added.  Both of the other "Change 2-s" ("...mtime-3 >
now..." and "mtime > now + 3") handled the warning always (whether rightly or
wrongly).

Rob Finley

-----Original Message-----
From:   Eli Zaretskii <eliz AT is DOT elta DOT co DOT il> 
Sent:   Tuesday, May 05, 1998 6:48 AM
To:     ROBERT FINLEY
Cc:     nate AT cartsys DOT com; djgpp AT delorie DOT com
Subject:        RE: Make "Clock Skew" problem.


On Mon, 4 May 1998 ROBERT_FINLEY AT ntsc DOT navy DOT mil wrote:

> It turns out the Change 1 shuts-up make and Change 2 has no impact.

This is a wrong conclusion.  Change 1 *includes* Change 2, so it is
better to use Change 2 alone (it's simpler).

There was no need to test the original ifdef, since DJGPP doesn't
define WINDOWS32, so unless you add __DJGPP__ or __MSDOS__, that code
won't be compiled into the DJGPP program.

> Now, weather or not the messenger has been shot or the problem is
> fixed, I don't know.

The problem is NOT fixed, you just told Make to ignore it.  IMHO,
leaving the warning is a better way, since then at least you know
something might be wrong and can look closer.  If make is shut up, you
are just oblivious to potential problems.

- Raw text -