From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:53987) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Yk6QX-0000wI-Fu for qemu-devel@nongnu.org; Mon, 20 Apr 2015 03:47:34 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Yk6QU-0000MQ-Nq for qemu-devel@nongnu.org; Mon, 20 Apr 2015 03:47:33 -0400 Received: from mx1.redhat.com ([209.132.183.28]:53117) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Yk6QU-0000MG-IB for qemu-devel@nongnu.org; Mon, 20 Apr 2015 03:47:30 -0400 Received: from int-mx11.intmail.prod.int.phx2.redhat.com (int-mx11.intmail.prod.int.phx2.redhat.com [10.5.11.24]) by mx1.redhat.com (Postfix) with ESMTPS id A655FA10AC for ; Mon, 20 Apr 2015 07:47:29 +0000 (UTC) Message-ID: <1429514329.12547.16.camel@nilsson.home.kraxel.org> From: Gerd Hoffmann Date: Mon, 20 Apr 2015 09:18:49 +0200 In-Reply-To: <1429280557-8887-24-git-send-email-berrange@redhat.com> References: <1429280557-8887-1-git-send-email-berrange@redhat.com> <1429280557-8887-24-git-send-email-berrange@redhat.com> Content-Type: text/plain; charset="UTF-8" Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH v1 RFC 23/34] io: add QIOChannelSocket class List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Daniel P. Berrange" Cc: Paolo Bonzini , qemu-devel@nongnu.org, Stefan Hajnoczi On Fr, 2015-04-17 at 15:22 +0100, Daniel P. Berrange wrote: > Implement a QIOChannel subclass that supports sockets I/O > > TBD check errno handling of windows port & fix watch impl > +struct QIOChannelSocket { > + QIOChannel parent; > + int fd; > + struct sockaddr_storage localAddr; > + socklen_t localAddrLen; > + struct sockaddr_storage remoteAddr; > + socklen_t remoteAddrLen; > +}; Looks like this supports a single listening socket only, correct? That is a long-standing issue we have in qemu. Listening on both ipv4 and ipv6 doesn't work, except when binding the socket to the ipv6 wildcard address which can accept ipv4 connects too on most systems. Would be nice to tackle that while putting the socket code upside down anyway. cheers, Gerd