From: Hans-Bernhard Broeker <broeker AT physik DOT rwth-aachen DOT de>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Peculiar behavior of program.
Date: 29 Jun 2001 10:11:18 GMT
Organization: Aachen University of Technology (RWTH)
Lines: 47
Message-ID: <9hhk86$pf7$1@nets3.rz.RWTH-Aachen.DE>
References: <Pine DOT SUN DOT 3 DOT 91 DOT 1010625151332 DOT 2920E-100000 AT is> <3b37df15 DOT 286160341 AT news DOT primus DOT ca> <9h9jfv$khe$1 AT nets3 DOT rz DOT RWTH-Aachen DOT DE> <3b3b4284 DOT 210410883 AT news DOT primus DOT ca> <9hfpvq$fqj$1 AT nets3 DOT rz DOT RWTH-Aachen DOT DE> <3b3be356 DOT 251586899 AT news DOT primus DOT ca>
NNTP-Posting-Host: acp3bf.physik.rwth-aachen.de
X-Trace: nets3.rz.RWTH-Aachen.DE 993809478 26087 137.226.32.75 (29 Jun 2001 10:11:18 GMT)
X-Complaints-To: abuse AT rwth-aachen DOT de
NNTP-Posting-Date: 29 Jun 2001 10:11:18 GMT
Originator: broeker@
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Graaagh the Mighty <invalid AT erehwon DOT invalid> wrote:

> Yeah, so obviously the debuggers I used *both* have bugs
> *simultaneously*, or the compiler output a bad object file *and*
> both the debuggers have bugs that cause ungraceful responses to
> pathological or inconsistent input.  

And how are we supposed to be able to tell, with the extremely small
amount of hard fact besides all the rants you've supplied so far?

You keep adding facts to your story piecemeal, where what you really
should have done is go back to your machine, and *try* some of the
suggestions you've been giving.  IIRC, you haven't reported any
results from even one of the first suggestions, yet: adding print
statements to see how deep the recursion really is, at the point of
crash.

Or you could produce something others can work on.  Like: upload the
sources as tarball, somewhere, and/or even the readily compiled
binary. 

>>Let's face it: advanced debugging *is* rocket science, by the popular
>>meaning of the latter word.

> Yeah. In other words, debugging is bad enough, without adding a shoddy
> user interface that has more bucky bits involved than menu items or
> documentation. :-)

No. Debugging is a hard task, which means there's not much point in
polishing the tools for doing it beyond a certain degree.  Real
Tools(tm) may look dirty, but they work.  If you really can't wrap
your mind around using command-line GDB, try RHGDB (a version of GDB
with a TVision user interface, like the one built into RHIDE), or
RHIDE itself.

>>You must be kidding. How could *I* possibly have made implications
>>based upon *your* best of knowledge?

> Well, based on the two debuggers that shipped with 2.02, rather than
> the new (old and reissued?) one that is mysteriously absent from 2.02,

It isn't absent. You just didn't install it, because it's a separate
package.

-- 
Hans-Bernhard Broeker (broeker AT physik DOT rwth-aachen DOT de)
Even if all the snow were burnt, ashes would remain.