Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com
Date: Wed, 4 Sep 2002 23:21:03 -0400
From: Christopher Faylor <cygwin AT cygwin DOT com>
To: cygwin AT cygwin DOT com
Subject: Re: NTVDM error
Message-ID: <20020905032103.GA10551@redhat.com>
Reply-To: cygwin AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
References: <3D767411 DOT CE1BDB47 AT t-online DOT de> <20020904211855 DOT GA7950 AT redhat DOT com> <3D7687DD DOT A51BFF0C AT t-online DOT de>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <3D7687DD.A51BFF0C@t-online.de>
User-Agent: Mutt/1.3.23.1i

On Thu, Sep 05, 2002 at 12:23:25AM +0200, Andreas Hadler wrote:
>There was no offense intended.

I was not offended.  I am habitually exasperated by people who offer bug
reports with no details but with wild theories about the causes of their
problems gleaned from misreading of email archives.

>The first one occurs when "checking for ANSI C header files..." during
>"Configuring in powerpc-eabi/libiberty" for the full blown gcc. It seems
>to be independent of the source version and independent of my installed
>compiler version (if the build process comes as far).

Checking for ANSI C header files does not scream out "Symlink problem!"
to me.

>I've seen the problem of NTVDM errors reported more than one time in
>this list. The most reasonable explanation seems to be the (undesired)
>execution of some non-executable code, like as trying to execute
>symlinks, as some posters suggested.

But, if you think about it, if Cygwin was actually trying to execute a
symlink directly rather than the program it referred to then it would be
broken.

If you are referring to trying to run a symlink from, say, the command
line, or the Start->run menu then, that could be a problem.  Symlinks
are a cygwin construction and Windows doesn't know anything about them.

When you are running a configure script you are in a cygwin environmennt.
Cygwin would be a very poor UNIX emulation environment if you were unable
to run a symlink linked to a program.

>I am in no way under the apprehension that this is a (well) known
>problem, neither with symlinks, nor with cygwin - au contraire, it seems
>to happen in just some not well analysed cases (mostly because of the
>OPs not reporting). If I'm able to help with my observation I'm glad to
>do so. But please bear with me - concerning cygwin, i'm just a dumb
>user, loving to have the unix tools handy on my windoze box.

I think someone else hit the nail on the head.  The cross configure
script may be building non-Windows applications and trying to run them.
On UNIX that would cause a nice clean failure.  On Windows, it will
cause other strange and random results.

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/