Mailing-List:	contact cygwin-developers-help AT cygwin DOT com; run by ezmlm
List-Subscribe:	<mailto:cygwin-developers-subscribe AT cygwin DOT com>
List-Archive:	<http://sources.redhat.com/ml/cygwin-developers/>
List-Post:	<mailto:cygwin-developers AT cygwin DOT com>
List-Help:	<mailto:cygwin-developers-help AT cygwin DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender:	cygwin-developers-owner AT cygwin DOT com
Delivered-To:	mailing list cygwin-developers AT cygwin DOT com
Date:	Sun, 11 Aug 2002 15:23:01 -0400
From:	Christopher Faylor <cgf AT redhat DOT com>
To:	cygwin-developers AT cygwin DOT com
Subject:	Re: SIGCLD (?) problem
Message-ID:	<20020811192301.GA9017@redhat.com>
Reply-To:	cygwin-developers AT cygwin DOT com
Mail-Followup-To:	cygwin-developers AT cygwin DOT com
References:	<016a01c24148$e61594e0$6132bc3e AT BABEL>
Mime-Version:	1.0
In-Reply-To:	<016a01c24148$e61594e0$6132bc3e@BABEL>
User-Agent:	Mutt/1.3.23.1i

On Sun, Aug 11, 2002 at 04:08:06PM +0100, Conrad Scott wrote:
>So it looks (to the untrained eye) like a race, but I've not got
>anywhere much myself so I thought I'd punt on this one and hope
>some more wizardly being out there might see something in all
>this.

I don't see how there could be a race.  thread 2 should be the
signal thread.  It would be interesting to see what it is doing.
It sounds like it has not been started or something, which would
make the "wait_for_me" function block forever.

I've made some changes and added some assertions throughout the code.
I don't think they will solve the problem but it may make things a
little clearer if there is a problem.

>p.s. I'll keep this process hung in gdb for the moment in case I
>get enthusiastic tomorrow or someone else wants some more
>information.

Could you check on the value of wait_sig_inited?  Is it non-zero?
The problem is that the process processing thread should not have
to be waiting for this.  Although, it looks like it is possible
for this scenario to happen if a process forks, and then quickly
execs.  My just-checked-in-changes change that behavior.  So,
now if it hangs it should hang in fork.

cgf

- Raw text -