Mailing-List: contact cygwin-help@cygwin.com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-owner@cygwin.com Mail-Followup-To: cygwin@cygwin.com Delivered-To: mailing list cygwin@cygwin.com Date: Thu, 17 Jul 2003 18:08:27 -0400 From: Christopher Faylor To: cygwin@cygwin.com Subject: Re: 1.5.0 - gettimeofday() out of sync with Windows time Message-ID: <20030717220827.GB24025@redhat.com> Reply-To: cygwin@cygwin.com Mail-Followup-To: cygwin@cygwin.com References: <3F1701B7.5000201@integrian.com> <3F171ABE.6020205@cygwin.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <3F171ABE.6020205@cygwin.com> User-Agent: Mutt/1.4.1i On Thu, Jul 17, 2003 at 05:53:02PM -0400, Larry Hall wrote: >Chris Church wrote: > >>I'm using version 1.5.0 of the cygwin1.dll on Windows 2000 >>Professional. I've also seen the problem described below on earlier >>versions of cygwin (1.3.22). >> >>I've found that the gettimeofday() and ftime() functions do not always >>return the correct system time as reported by Windows. The time() >>function, however, always remains in sync with Windows time. This >>discrepancy occurs when an application is started, then the Windows >>system time changes, then the application calls gettimeofday() or >>ftime() to retrieve the current system time. Both gettimeofday() and >>ftime() always report the same incorrect time. It also appears that the >>time as reported by gettimeofday() and ftime() is based on a counter >>that is initialized to match the system time, but once started, >>increments at a constant rate regardless of any changes to the Windows >>time. (I have not yet dug into the source for Cygwin to see how these >>functions are implemented.) Finally, the time reported by >>gettimeofday() and ftime(), when used within an application that runs >>for an extended period of time, will drift from the Windows time. >> >>Is this a known issue with Cygwin, and are there any patches available? > >Not yet. But this was mentioned on the developer's list this morning >and someone is looking at the problem. Sorry, no, no one is looking at the problem. It's been a known problem since the implementation. -- 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/