Mailing-List: contact cygwin-help@sourceware.cygnus.com; run by ezmlm
Sender: cygwin-owner@sourceware.cygnus.com
Message-ID: <380F5E6D.6FF50587@cs.unc.edu>
Date: Thu, 21 Oct 1999 14:41:49 -0400
From: Jeffrey Juliano <juliano@cs.unc.edu>
MIME-Version: 1.0
To: cygwin@sourceware.cygnus.com
Subject: Re: B20.1 gettimeofday returns incorrect tv_secs
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Paul Berrevoets wrote:

>Fixed in recent snapshots:
>
>     1999-06-04  DJ Delorie  <dj@cygnus.com>
>
>          * times.cc (totimeval): scale sub properly.
>          (gettimeofday): don't bias by timezone.

Thanks for posting the changelog entry.  However, I still have a
question.

In this changelog entry, and in the mail archive, I've seen many
references to a /timezone bias/ bug(fix).  But, I haven't found anything
specifically addressing whether the following problem has been fixed:

 * in cygwin B20.1, the zero point for the result of gettimeofday is
   not 1970.  Instead it is (either when the computer was turned
   on or when you logged in?)

 * this causes problems when trying to synch the clocks between
   a unix box and a cygwin NT box.

I hope this has been changed as well in recent snapshots.  can someone
please verify?  If so, are they in the version on the CD released last
week?

   [FWIW, neither "scale" now "timezone bias"
    sounds to me like what I'm describing]

thanks,

-jeff

--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe@sourceware.cygnus.com