From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: network interface state Date: Wed, 14 Nov 2007 16:22:01 -0800 (PST) Message-ID: <20071114.162201.174150647.davem@davemloft.net> References: <473B61C8.3030302@redhat.com> <20071114.153103.230463871.davem@davemloft.net> <473B8EEC.1010906@redhat.com> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org To: drepper@redhat.com Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:53245 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1756233AbXKOAWB (ORCPT ); Wed, 14 Nov 2007 19:22:01 -0500 In-Reply-To: <473B8EEC.1010906@redhat.com> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org From: Ulrich Drepper Date: Wed, 14 Nov 2007 16:12:28 -0800 > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > David Miller wrote: > > Most daemons handle this by listening for events on the netlink > > socket, but I understand how that might not be practical for > > glibc. > > Right, this cannot work. I have no inner loop which I can control. I > cannot install a listener. > > At some point, when we have non-sequential, hidden file descriptors, > I'll be able to leave a socket file descriptor open. But that's about > it. Even then the generation counter interface is likely to be the best > choice. Ok, I'll think about how to implement this. > BTW, is it possible to send both these requests out before starting to > read the results? This would reduce the amount of code quite a bit. Unfortunately, that won't work. Like datagram protocols, netlink assumes one message per sendmsg() call.