Date: Fri, 24 Feb 2006 16:38:08 -0500
From: Christopher Faylor <cgf-no-personal-reply-please AT cygwin DOT com>
To: cygwin AT cygwin DOT com
Subject: Re: Shells hang during script execution
Message-ID: <20060224213808.GA7638@trixie.casa.cgf.cx>
Reply-To: cygwin AT cygwin DOT com
References: <43FF6AED DOT 1060102 AT steeleye DOT com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <43FF6AED.1060102@steeleye.com>
User-Agent: Mutt/1.5.11
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
Precedence: bulk
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com

On Fri, Feb 24, 2006 at 03:22:05PM -0500, Paul Clements wrote:
>On Thu, Feb 23, 2006 at 04:40:33PM -0500, Christopher Faylor wrote:
>>Actually, I spoke too soon.  I don't understand your analysis.  You
>>make this statement:
>
>>>However, if the queueing of the blocked signal happens right after the
>>>signal mask change, then we miss the signal.
>
>>but I don't understand how you came to this conclusion.  If the signal
>>mask changes, then the queue should be flushed -- that's what the
>>__SIGFLUSH is for.  So, even if something is added to the queue while
>>the mask is changing, the function which changes the mask should
>>eventually call send_sig to cause the wait_sig thread to run the queue.
>
>>So, this patch would seem to be a band-aid over some other problem.
>
>OK, I think I see what is happening now:
>
>set_signal_mask() does indeed call sig_dispatch_pending() to do a flush
>when the signal mask changes.  However, when the sigq is empty,
>sig_dispatch_pending() just returns! See:
>
>
>sig_dispatch_pending (bool fast) { -if (exit_state || &_my_tls ==
>_sig_tls || !sigq.start.next) +if (exit_state || &_my_tls == _sig_tls)

I always wondered if that "optimization" was right and now I see that it is
obviously racy.  I've removed it from the source.

>Unfortunately, when I make this change, everything hangs. Not sure what 
>the issue is there. Any ideas?

Not until I ran the test suite, waited five minutes and saw an error.
Attaching gdb to the running process shows another problem.  I'll check
in a fix for this, too.

>(please CC, I'm not on the list)

I've cc'ed you this once however please join the mailing list if you
want to talk about cygwin.

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/