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, 29 Sep 2002 20:27:16 -0400
From: Christopher Faylor <cgf AT redhat DOT com>
To: cygwin-developers AT cygwin DOT com
Subject: Re: Many pthread failures in the test suite, one setgroup failure
Message-ID: <20020930002716.GD27934@redhat.com>
Reply-To: cygwin-developers AT cygwin DOT com
Mail-Followup-To: cygwin-developers AT cygwin DOT com
References: <20020929022338 DOT GA12659 AT redhat DOT com> <1033267203 DOT 4374 DOT 100 DOT camel AT lifelesswks> <20020929024420 DOT GA13416 AT redhat DOT com> <1033271512 DOT 4372 DOT 102 DOT camel AT lifelesswks> <20020929141659 DOT GA23836 AT redhat DOT com> <1033309085 DOT 11273 DOT 69 DOT camel AT lifelesswks> <20020929185604 DOT GA25789 AT redhat DOT com> <1033343397 DOT 30057 DOT 167 DOT camel AT lifelesswks> <20020930000824 DOT GC27934 AT redhat DOT com> <1033344793 DOT 11275 DOT 169 DOT camel AT lifelesswks>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1033344793.11275.169.camel@lifelesswks>
User-Agent: Mutt/1.4i

On Mon, Sep 30, 2002 at 10:13:12AM +1000, Robert Collins wrote:
>On Mon, 2002-09-30 at 10:08, Christopher Faylor wrote:
>> On Mon, Sep 30, 2002 at 09:49:56AM +1000, Robert Collins wrote:
>> >I've implemented syncronisation around the pthread_mutex initialisation.
>> >This will not impact performance except for *really badly* written
>> >programs, as mutex creation is rare.
>> >
>> >I still can't reproduce the problem myself, but the changes just checked
>> >in cause no regressions for the test suite here.
>> >
>> >Chris, does this help?
>> 
>> Actually, it seems to have caused the problem to occur much more now.
>> I'm rerunning the whole testsuite to see if there are regressions.
>
>Urggle. I've got to go do real work for a while. Is there any chance you
>can give me ssh access to the failing machine?

Not easily.  I'm behind a firewall.  I'll tinker some and see if it is
possible.

I was wrong in my original assessment of what is going on.  It's dying
in a different place now.  It is failing on line 94 of condvar6.c.
And, it doesn't always fail.  It seemed to work the first time I ran
it.

cgf