Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sources.redhat.com/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com
X-MimeOLE: Produced By Microsoft Exchange V6.0.4417.0
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: RE: [ANNOUNCEMENT] Updated: cygwin-1.3.11-3
Date: Tue, 25 Jun 2002 14:54:47 -0400
Message-ID: <BADF3C947A1BD54FBA75C70C241B0B9E10AA45@ex02.idirect.net>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
From: "Harig, Mark A." <maharig AT idirect DOT net>
To: "Conrad Scott" <Conrad DOT Scott AT dsl DOT pipex DOT com>,
   "Heribert Dahms" <heribert_dahms AT icon-scm DOT com>, <cygwin AT cygwin DOT com>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by delorie.com id g5PIw9M30924

Thank you for making it (the timing issue) public.
It helped resolve a problem I was having.

I have a process that does the following steps:

   1. generate assembly language from c code
   2. assemble the .asm into .o
   3. archive the .o into .a

With 'make' this worked well on the first pass
through these three steps, but if I invoked 'make'
immediately after the .a was generated, then 'make'
thought that the .a was out-of-date with respect to
some -- but not all -- of the .o files.

The reason?  The recorded dates for the .o's within
the .a was within the same second as the timestamp
for the .o's.  If I ran 'make' a second time, then
the timestamps of "internal" .o's was different from
the .o files.  A third invocation of 'make' found that
the .a was up to date.  But this is an error because the
.a should have been up to date after the first invocation.  

The "solution"?  I added step 1a. "sleep 1":

   1. generate assembly language from c code
   1a. sleep 1
   2. assemble the .asm into .o
   3. archive the .o into .a

I don't know if this will be a common problem or not.

Should it be considered part of the standard Cygwin build
of 'make' that the '--disable-nsec-timestamps' be used
during configuration?


> -----Original Message-----
> From: Conrad Scott [mailto:Conrad DOT Scott AT dsl DOT pipex DOT com]
> Sent: Monday, June 24, 2002 6:29 PM
> To: Heribert Dahms; cygwin AT cygwin DOT com
> Subject: Re: [ANNOUNCEMENT] Updated: cygwin-1.3.11-3
> 
> 
> "Heribert Dahms" <heribert_dahms AT icon-scm DOT com> wrote:
> > Hi Conrad,
> >
> > wow, you seem to have a really fast machine, where
> > you can run multiple nontrivial makes per second 8-)
> 
> Not *nontrivial* perhaps . . .
> 
> It's part of a (still ongoing) attempt to write a makefile that
> handles compiling TeX files, where you've got to re-do the processing
> an arbitrary number of times until the output settles down: I've
> currently got a makefile that touches marker files depending on
> various conditions and then re-invokes itself. But some of the phases
> don't take very long to run, so sometimes things don't get re-compiled
> when they should. Actually, I should probably be more delicate than to
> discuss it in public . . . it's *really* foul. Then again, if it ever
> works, it'll all of a sudden be beautiful :-)
> 
> // Conrad
> 
> 
> 
> 
> --
> Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
> Bug reporting:         http://cygwin.com/bugs.html
> Documentation:         http://cygwin.com/docs.html
> FAQ:                   http://cygwin.com/faq/
> 
> 

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/