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: Tue, 11 Sep 2001 21:15:23 -0400
From: Jason Tishler <jason AT tishler DOT net>
To: Robert Collins <robert DOT collins AT itdomain DOT com DOT au>,
        cygwin-developers AT cygwin DOT com
Subject: Re: Quick testfeedback...
Message-ID: <20010911211523.B1108@dothill.com>
Mail-Followup-To: Robert Collins <robert DOT collins AT itdomain DOT com DOT au>,
	cygwin-developers AT cygwin DOT com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20010911204019.H1752@dothill.com>
User-Agent: Mutt/1.3.18i

On Tue, Sep 11, 2001 at 08:40:19PM -0400, Jason Tishler wrote:
> On Wed, Sep 12, 2001 at 09:19:19AM +1000, Robert Collins wrote:
> > If it was after the bugfix commit, I'd like to see if we can track this
> > asap...
> 
> I can reproduce it frequently but not every time.  I will debug it first
> thing tomorrow morning.  FYI, it seems to occur right near the end of
> the test and the test still passes.

Doh!  My first shot running broadcast from gdb caused the SIGSEGV:

(gdb) run
Starting program: /home/jt/src/PthreadTest-0.1/broadcast.exe 
before cond wait
..
before cond broadcast
after  cond broadcast
after  cond wait

Program received signal SIGSEGV, Segmentation fault.
[Switching to thread 1804.0x2dc]
0x6100f0cb in call_signal_handler_now ()
    at ../../../../src/winsup/cygwin/exceptions.cc:1132
1132      *sigsave.retaddr_on_stack = sigsave.retaddr;
Current language:  auto; currently c++
(gdb) bt
#0  0x6100f0cb in call_signal_handler_now ()
    at ../../../../src/winsup/cygwin/exceptions.cc:1132
#1  0x6100f12e in sigframe::call_signal_handler (this=0x1e06f360)
    at ../../../../src/winsup/cygwin/exceptions.cc:1146
#2  0x6104ff64 in sleep (seconds=2)
    at ../../../../src/winsup/cygwin/signal.cc:84
#3  0x004011fe in baby () at broadcast.c:56
#4  0x61063b6c in thread_init_wrapper (_arg=0xa011588)
    at ../../../../src/winsup/cygwin/thread.cc:865
#5  0x77e837cd in _cygheap_start ()
(gdb) list
1127    static int __stdcall
1128    call_signal_handler_now ()
1129    {
1130      int sa_flags = sigsave.sa_flags;
1131      sigproc_printf ("sa_flags %p", sa_flags);
1132      *sigsave.retaddr_on_stack = sigsave.retaddr;
1133      sigdelayed0 ();
1134      return sa_flags & SA_RESTART;
1135    }
1136    /* This kludge seems to keep a copy of call_signal_handler_now around
(gdb) p sigsave.retaddr_on_stack
$1 = (DWORD *) 0x0

I really have to go now.  I will pick this up first thing in the morning
assuming that I'm not still in the dog house...  No cable modem there... :,)

Jason