Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Unsubscribe: <mailto:cygwin-unsubscribe-archive-cygwin=delorie DOT com AT sourceware DOT cygnus DOT com>
List-Subscribe: <mailto:cygwin-subscribe AT sourceware DOT cygnus DOT com>
List-Archive: <http://sourceware.cygnus.com/ml/cygwin/>
List-Post: <mailto:cygwin AT sourceware DOT cygnus DOT com>
List-Help: <mailto:cygwin-help AT sourceware DOT cygnus DOT com>, <http://sourceware.cygnus.com/ml/#faqs>
Sender: cygwin-owner AT sourceware DOT cygnus DOT com
Delivered-To: mailing list cygwin AT sourceware DOT cygnus DOT com
From: Chris Faylor <cgf AT cygnus DOT com>
Date: Fri, 24 Sep 1999 07:42:57 -0400
To: "J.A. Bezemer" <costar AT panic DOT et DOT tudelft DOT nl>
Cc: cygwin AT sourceware DOT cygnus DOT com
Subject: Re: B20.1 & Win95 fork/socket bug, or Why Rsync Doesn't Work
Message-ID: <19990924074257.A740@cygnus.com>
Reply-To: cygwin AT sourceware DOT cygnus DOT com
Mail-Followup-To: "J.A. Bezemer" <costar AT panic DOT et DOT tudelft DOT nl>,
	cygwin AT sourceware DOT cygnus DOT com
References: <Pine DOT LNX DOT 3 DOT 96 DOT 990924130420 DOT 5468H-100000 AT panic DOT et DOT tudelft DOT nl>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 0.95.6i
In-Reply-To: <Pine.LNX.3.96.990924130420.5468H-100000@panic.et.tudelft.nl>; from J.A. Bezemer on Fri, Sep 24, 1999 at 01:26:04PM +0200

On Fri, Sep 24, 1999 at 01:26:04PM +0200, J.A. Bezemer wrote:
>I tried to compile/run rsync (sources from
>http://www.student.uni-koeln.de/cygwin/Distribution/Source/rsync/) under
>Win95/Cygwin B20.1. Compiles allright, but doesn't run. Under 98 it runs
>perfectly well, so I suspect this problem is somewhere in Win95. It would be
>very nice if cygwin could do a workaround.

Patches gratefully accepted.

-Chris Faylor
-Win32 Manager
-Cygnus Solutions

>The problem is this: in do_recv() in rsync's main.c there's this structure:
>
>	f_in is filedescriptor of a socket connected to the rsync server
>
>        fork
>           child does read(f_in)
>
>        parent does write(f_in)
>
>Parent's write() is okay, but childs read() doesn't receive anything, and the
>child process just waits forever.
>
>I don't know if this affects only sockets, or all filedescriptors.
>
>In the mean time, for my specific situation I could remove the fork
>completely, and simply run the child after the parent. This works
>perfectly.
>
>I don't have time to go into more detail, but just ask if you need to know
>more. In case you're interested, My hacked version of rsync is at
>http://panic.et.tudelft.nl/~costar/pseudo-image-kit ("Windows version").
>
>And BTW, set CYGWIN=binmode, otherwise rsync goes all wrong...
>
>Regards,
>  Anne Bezemer

--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com