Date: Tue, 22 Oct 96 15:32:55 BST
Message-Id: <9610221432.AA20384@leopard.proteon.com>
From: Neil Jarvis <Neil DOT Jarvis AT proteon DOT com>
To: eliz AT is DOT elta DOT co DOT il
Cc: djgpp-workers AT delorie DOT com
Subject: Bug in Make 3.75
Reply-To: Neil DOT Jarvis AT proteon DOT com
Mime-Version: 1.0 (generated by tm-edit 7.43)
Content-Type: text/plain; charset=US-ASCII

Eli,

There is a bug in the new port of make, version 3.75 (Newly downloaded
from ftp.simtel.net). When a rule's actions are being executed, and a
particular action returns an errorcode which is ignore (by placing '-'
in front of the command), all subsequent actions inherit the bad error
code, and the make fails.

An example should make it all clear. 

Below is a short C program, ret_code.c, that returns the error code
that is passed as its first parameters. The makefile target 'all'
calls this program twice, the first time ignoring an error level 1,
and then again with an error level of 0. The 'all' target should exit
with a global error level of 0, indicating a pass.

****************************************
ret_code.c
****************************************

int main(int argc, char **argv) 
{
   if (argc == 1)
      return 0;
   else
      return (atoi(argv[1]));
}

****************************************
makefile
****************************************

all:
	-ret_code 1
	ret_code 0


****************************************
Under DJGPP V2.01, make (3.75) generates the following incorrect output:

> make
ret_code 1
make.exe: [all] Error 1 (ignored)
ret_code 0
make.exe: *** [all] Error 1                     <<< This is wrong
>

****************************************
Under SunOS 4.1.3, GNU make (3.74) generates the following correct output:

> make
ret_code 1
make.exe: [all] Error 1 (ignored)
ret_code 0
>

****************************************

This bug prevents anyone rebuilding binutils-2.7, using the V2.01
distribution. The libiberty library makefile has a rule that relies on
an action generating an error, which is ignored, and the error output
is processed to complete the target. This doesn't work....

Hope this can be fixed soon, as I am trying to rebuild a debug version
of 'ar.exe' to investigate why it does not work on a networked drive.....

Cheers,

-- Neil Jarvis, Proteon International R&D, York, UK. (Neil DOT Jarvis AT proteon DOT com)