Date: Fri, 10 Dec 93 04:30:07 PST
From: Advanced Amateur Radio Networking Group <tcp-group@ucsd.edu>
Errors-To: TCP-Group-Errors@UCSD.Edu
Reply-To: TCP-Group@UCSD.Edu
Precedence: Bulk
Subject: TCP-Group Digest V93 #318
To: tcp-group-digest
TCP-Group Digest Fri, 10 Dec 93 Volume 93 : Issue 318
Today's Topics:
NOS compile (2 msgs)
Send Replies or notes for publication to: <TCP-Group@UCSD.Edu>.
Subscription requests to <TCP-Group-REQUEST@UCSD.Edu>.
Problems you can't solve otherwise to brian@ucsd.edu.
Archives of past issues of the TCP-Group Digest are available
(by FTP only) from UCSD.Edu in directory "mailarchives".
We trust that readers are intelligent enough to realize that all text
herein consists of personal comments and does not represent the official
policies or positions of any party. Your mileage may vary. So there.
----------------------------------------------------------------------
Date: Thu, 9 Dec 93 09:43:12 EST
From: crompton@NADC.NADC.NAVY.MIL (D. Crompton)
Subject: NOS compile
To: aroberts%csunb.mit.csu.edu.au@NADC.NADC.NAVY.MIL, tcp-group@ucsd.edu
Adam,
This is a commonly asked question. Actually there should be a note in
the makefile about this. I have put one in in some versions that I did.
The problem is that you are trying to compile in to many modules and there
is not enough DGROUP room in it's 64K segment.
The answer is to use the -Ff=xxxx compiler option. This would go in to
makefile in the same line as other compiler options.
The xxxx is set to the MAXIMUM value that will allow a compile. If the
-Ff line is left off altogether it defaults to 32767.
Setting it to say 1023 means that any data segment larger than that will
be forced to far data, thus freeing up some DGROUP room.
1023 is a standard number and is actually in many of the makefiles. I have
gone as low as 255 in some compiles.
You MUST do a complete recompile (touch *.c or del *.obj) in between the
changes. Just keep lowering it in say 256 steps until in goes.
You CANNOT compile everything in! You will reach a limit where resources
just do not exist and you will have to make a decision to drop some module.
When using the -Ff directive the data that is forced FAR will give a
warning message - Near data xxx is being made far - or whatever - YOu
can safetly ignore this.
The other thing that could be done in the code is to go thru it and make
large data areas (like command table) dar data to begin with. I am not
suggesting you do that, but I know someone who has.
I suggest to the developers that they include a message to this effect in the makefile.
Doug
------------------------------
Date: 09 Dec 1993 08:30:08 GMT
From: "Robert D Nielsen" <0060880@CCMAIL.EMIS.hac.com>
Subject: NOS compile
To: TCP-Group@UCSD.EDU
aroberts@csunb.mit.csu.edu.au writes:
>Hello all, I am trying to compile the latest version of nos that I FTP'd
>recently and am having problems.
>
>Using Borland C++ 3.1 , when I compile the src of jn110x14 (or jn110x10) I
>receive errors at the "TurboLink" line:
>
>......
>ERROR: Group DGROUP exceeds 64K
>
>* error 1 * deleting nos.exe
>.....
>
>I am not a c programmer (yet) but do know about makefile's and config.h files.
>
>If I use the supplied distconf.h as the config.h file it compiles OK, but
>when I want to define some others into it (even just one of : IPACCESS or
>ENCAP or LZW) it crashes and I do not get an .exe file.
Look in MAKEFILE for a line that begins with "CFLAGS= " and change the -Ff
parameter to a smaller number. Don't go any smaller than you need to in
order to get the compilation to work.
Bob Nielsen
w6swe@tapr.org
0060880@ccmail.emis.hac.com
nielsen@igate1.hac.com
------------------------------
End of TCP-Group Digest V93 #318
******************************
******************************