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
To:	cygwin AT cygwin DOT com
X-Injected-Via-Gmane:	http://gmane.org/
Path:	not-for-mail
From:	"Dylan Cuthbert" <dylan AT q-games DOT com>
Newsgroups:	gmane.os.cygwin
Subject:	Re: Available for test: gcc-3.1.1-3, gcc2-2.95.3-9, gcc-mingw-3_1-20020516-2
Date:	Wed, 17 Jul 2002 23:03:02 +0900
Lines:	48
Message-ID:	<ah3tb7$1s9$1@main.gmane.org>
References:	<20020715210550 DOT GA17619 AT redhat DOT com> <ah0bkt$s62$1 AT main DOT gmane DOT org> <ah38u7$i1g$1 AT main DOT gmane DOT org> <20020717130113 DOT GD28334 AT redhat DOT com>
NNTP-Posting-Host:	pppa20.kyoto-ip.dti.ne.jp
Mime-Version:	1.0
X-Trace:	main.gmane.org 1026914472 1929 210.159.246.20 (17 Jul 2002 14:01:12 GMT)
X-Complaints-To:	usenet AT main DOT gmane DOT org
NNTP-Posting-Date:	Wed, 17 Jul 2002 14:01:12 +0000 (UTC)
X-Priority:	3
X-MSMail-Priority:	Normal
X-Newsreader:	Microsoft Outlook Express 6.00.2600.0000
X-MimeOLE:	Produced By Microsoft MimeOLE V6.00.2600.0000

If the debugger worked, yes I would *of course* run gdb and see where it
crashes.

At the moment when I run the debugger (the latest exp version of gdb) it
puts its "start of main" breakpoint in a piece of code that is nowhere near
main and when I try to step or run my program, my program seg-faults.  The
debugger at this point freezes completely.  I have compiled the entire
source code with -ggdb and even -g but to no avail.

As I may have mentioned, the program we are developing uses a large amount
of modern C++ (boost/mpl/loki etc) and it seems to kind of push the compiler
and gdb to their limits.  Simpler programs I've compiled and tested work
fine.

I'm clutching at straws, which is why I was asking about ABI changes
between -march=pentiumpro and -march=pentium2/3, would alignment adjustments
in the linker (I think you mentioned you had to revert some changes to get
the old gcc to work properly?)

Now, the reasons I'm mailing to this list are that the program works *fine*
without the -mpentium2/3 option to gcc.  It also works *fine* even with
those options if I run it from a command.com bash shell.  The program itself
is a 3d OpenGL program and a non-cygwin application (although not currently
compiling with -mno-cygwin because mingw wasn't up to gccv3.1 until
recently), the program (when it runs from a command.com bash shell) exhibits
absolutely no problems at all.  When it is able to run from rxvt at all,
simply no output appears whatsoever... when it doesn't run, it crashes
immediately and doesn't even write a stackdump file.

The problem might simply be gcc-related but its odd how it is broken "in
combination" with rxvt.

Just trying to catch some of these problems before they get to other
people... any help appreciated.

Regards

---------------------------------
Q-Games, Dylan Cuthbert.
http://www.q-games.com

PS. rxvt seems to have a no. of problems - sometimes pushing ctrl-o simply
freezes it completely, I haven't tracked down the exact conditions but it is
really annoying sometimes. .Clicking on the top-right close-window button
when it is running a child-process causes the window to disappear but the
task to keep running.- type "sleep 1h" into an rxvt shell, close the window
with the X button and type ps in another shell and you'll see what I mean...




--
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/

- Raw text -