Xref: news2.mv.net comp.os.msdos.djgpp:1225
From: Charles Sandmann <sandmann AT clio DOT rice DOT edu>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: DJGPP v2 woes :(
Date: Wed, 14 Feb 1996 17:38:48 CST
Organization: Rice University, Houston, Texas
Lines: 24
Message-ID: <31227288.sandmann@clio.rice.edu>
References: <4ft5lv$lke AT serra DOT unipi DOT it> <4ft3jr$sb6 AT odin DOT diku DOT dk> <WALD DOT 96Feb14153852 AT woodpecker DOT lcs DOT mit DOT edu>
Reply-To: sandmann AT clio DOT rice DOT edu
NNTP-Posting-Host: clio.rice.edu
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp

> Sorry, but we're all coming out of the woodwork now.  

DJ and I plan to publish a paper with concrete proof that no matter
how extensive a beta you do, bugs will be discovered within 5 hours of
any software release :-)

> I'm hitting the
> same sort of problem (infernal compiler error with when compiling
> template-laden C++ code).  Unfortunately, heapfix isn't a full
> solution for me: it allows me to compile a few more files, but I've
> got a couple of files which won't compile no matter how high I bump the heap.

Since they work on other platforms, two suggestions:
 1) Stubedit the stack size to 512K or 1Mb in CC1PLUS.EXE, and
 2) Try to compile under something like Windows.  If succesful, CWSDPMI has
    a bug.  Please send output of CPP (the preprocessor output) to me
    with an appropriate cc1plus command line, and I'll fix it.

>  (At least, until heapfix bumps the heap beyond x9400, at
> which point I no longer have a runnable CWSDPMI.EXE.)  

Yes, the data segment exceeds 64K, and you known how DOS hates that.  I
doubt anything over 1 or 2 bumps will make any difference anyway.