Message-ID:	<36548068.F854AC79@montana.com>
Date:	Thu, 19 Nov 1998 13:32:40 -0700
From:	bowman <bowman AT montana DOT com>
X-Mailer:	Mozilla 4.5 [en] (Win95; I)
X-Accept-Language:	en
MIME-Version:	1.0
To:	djgpp AT delorie DOT com
Subject:	Re: compiling with Rsxntdj
References:	<19981119152615 DOT 15192 DOT rocketmail AT send104 DOT yahoomail DOT com>
Reply-To:	djgpp AT delorie DOT com

Jeremy Ford wrote:
> 
> appropriate linker. ( I believe you have to use the linker that comes
> with rsxntdj.) I have not been able to test this extensively but the
> programs I have built worked...


I've been switching between the djgpp 'ld.exe' and the one supplied by
rsxntdj. The standard djgpp version seems to work, and produce an output
acceptable to 'ntbind'. This is preferable to the silent failure of the
rsxntdj version.

I also found that #include <stdio.h> will pull in the MS version from
/rsxntdj/include/mssdk, resulting in some errors. I have not played with
this enough to determine the extent of the problem, or its resolution.
So far, I've just used sscanf, and can live with the default and
warning. 'wsprintf' is picked up from the <windows.h>, but I don't think
there is an analogous scanf.

I did find an overview of the mt and st libraries in emxlib.txt. Thanks
for the input on relative sizes; I'll move to the single thread for the
time being.

the 'stabs string' thing in C++ looks like it will require recompiling
the compiler. I don't know if I have that much ambition, especially if
use of the STL will require more hacking of conflicting headers.

- Raw text -