X-Recipient:	archive-cygwin AT delorie DOT com
X-Spam-Check-By:	sourceware.org
Date:	Wed, 1 Aug 2012 11:05:00 +0200
From:	Corinna Vinschen <corinna-cygwin AT cygwin DOT com>
To:	cygwin AT cygwin DOT com
Subject:	Re: Using recvmsg() with non-unicast datagram (UDP) socket: inconsistent results
Message-ID:	<20120801090500.GC18636@calimero.vinschen.de>
Reply-To:	cygwin AT cygwin DOT com
Mail-Followup-To:	cygwin AT cygwin DOT com
References:	<A9D8BF3D8A74DF4A925FB541C0F39D2A16C25F128A AT NIHMLBX15 DOT nih DOT gov>
MIME-Version:	1.0
In-Reply-To:	<A9D8BF3D8A74DF4A925FB541C0F39D2A16C25F128A@NIHMLBX15.nih.gov>
User-Agent:	Mutt/1.5.21 (2010-09-15)
Mailing-List:	contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Id:	<cygwin.cygwin.com>
List-Unsubscribe:	<mailto:cygwin-unsubscribe-archive-cygwin=delorie DOT com AT cygwin DOT com>
List-Subscribe:	<mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive:	<http://sourceware.org/ml/cygwin/>
List-Post:	<mailto:cygwin AT cygwin DOT com>
List-Help:	<mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/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

On Jul 31 12:43, Lavrentiev, Anton (NIH/NLM/NCBI) [C] wrote:
> Hi,
> 
> Observed is an inconsistent result when recvmsg() is used on a
> UDP socket with or without ancillary control message parameters.
> 
> Here's is an example:
> 
>     static char buf[1 << 16];
>     struct sockaddr_in sin;
>     struct msghdr msg;
>     struct iovec iov;
> #ifndef CYGWIN_BUG
>     char cmsgbuf[32];
> #endif /*CYGWIN_BUG*/
>     int n;
> 
>     iov.iov_base       = buf;
>     iov.iov_len        = sizeof(buf);
>     memset(&sin, 0, sizeof(sin));
>     memset(&msg, 0, sizeof(msg));
>     msg.msg_name       = (caddr_t) &sin;
>     msg.msg_namelen    = sizeof(sin);
>     msg.msg_iov        = &iov;
>     msg.msg_iovlen     = 1;
> #ifndef CYGWIN_BUG
>     memset(cmsgbuf, 0, sizeof(cmsgbuf));
>     msg.msg_control    = cmsgbuf;
>     msg.msg_controllen = sizeof(cmsgbuf);
> #endif /*CYGWIN_BUG*/
> 
>     if ((n = recvmsg(sock, &msg, 0/*no special flags*/)) <= 0) {
> 	   /* Handle error */
>         return;
>     }
> 
>     /* HERE: msg.msg_flags is inconsistent here for non-unicast UDP sockets */
> 
> For a non-unicast socket "sock", the returned msg.msg_flags will have proper WSA
> bits (MSG_BCAST or MSG_MCAST) set if CYGWIN_BUG is not defined (i.e. the recvmag()
> call is passed non-zero msg.msg_controllen).  Otherwise, msg.msg_flags will always
> get returned 0.
> 
> The relevant code from CYGWIN source suggests that the native WSA flags should
> have got preserved (fhandler_socket.cc, fhandler_socket::recvmsg):
> 
> 	msg->msg_flags = wsamsg.dwFlags;
> 
> yet apparently this had slipped away in fhandler_socket::recv_internal, which
> branches on the presence of msg_controllen (and opts out for use other WSA API
> call if that is not set, even for non-stream sockets):
> 
>      if (wsamsg->Control.len > 0)
>        {
>          ...
>          use_recvmsg = true;
>        }
> 
> Can this please be rectified?

I applied a patch which should fix this problem.  It always uses
WSARecvMsg now, if called from recvmsg, still provided it is possible,
of course (No SOCK_STREAM, no AF_LOCAL).

I'm just generating a new developer snapshot.  Please give the today's
snapshot from http://cygwin.com/snapshots/ a try.  Should be up in about
half an hour.


Thanks,
Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader          cygwin AT cygwin DOT com
Red Hat

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple

- Raw text -