Mailing-List: contact cygwin-developers-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-developers-subscribe AT sources DOT redhat DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin-developers/>
List-Post: <mailto:cygwin-developers AT sources DOT redhat DOT com>
List-Help: <mailto:cygwin-developers-help AT sources DOT redhat DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender: cygwin-developers-owner AT sources DOT redhat DOT com
Delivered-To: mailing list cygwin-developers AT sources DOT redhat DOT com
Date: 18 Oct 2001 12:11:35 -0400
Message-ID: <20011018161135.27956.qmail@lizard.curl.com>
From: Jonathan Kamens <jik AT curl DOT com>
To: cygwin-developers AT cygwin DOT com
Subject: Current repository crashes when compiled without debugging,
   doesn't when compiled with debugging

When I compile the current repository without --enable-debugging,
everything crashes on start-up, e.g.:

  m:\>ps
        PID    PPID    PGID     WINPID  TTY  UID    STIME COMMAND
        0 [ttyin] ps 199 handle_exceptions: Exception: STATUS_ACCESS_VIOLATION
        199       1     199        199    0  500 12:07:37 /usr/bin/ps

  m:\>ls
        0 [ttyin] ls 120 handle_exceptions: Exception: STATUS_ACCESS_VIOLATION
    37122 [ttyin] ls 120 open_stackdumpfile: Dumping stack trace to ls.exe.stackdump
  #MAIL~62          core               merging-trunk      setup.ini
  #MAIL~RN          curl               metal              sh.exe.stackdump
  Application Data  curl-web           ns_imap            solutions

  m:\>

When I compile with --enable-debugging, the crash goes away, which
makes this somewhat difficult to debug :-(.  When I try to use
error_start to start gdb when the non-debugging DLL crashes, gdb says
it can't attach to the process.

The 20011016 snapshot doesn't seem to have this problem.  There's no
newer snapshot for me to try, so I can't find out if this is something
unique to my build environment.

If, indeed, something checked into the repository in the last two days
is causing a problem, it's probably better for the people who've
checked in in the past two days to debug it rather than me proceeding
:-).  For me to pursue this any further, I suspect I'd have to set
CYGWIN_SLEEP, step through process start up in a cygwin1.dll compiled
with --enable-debugging, and try to figure out what's being done
differently with --enable-debugging that is preventing the crash.  I
confess I'm loathe to attempt that.

  jik