Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261787AbTI3Xeb (ORCPT ); Tue, 30 Sep 2003 19:34:31 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261798AbTI3Xea (ORCPT ); Tue, 30 Sep 2003 19:34:30 -0400 Received: from opus.cs.columbia.edu ([128.59.20.100]:20958 "EHLO opus.cs.columbia.edu") by vger.kernel.org with ESMTP id S261787AbTI3Xe2 (ORCPT ); Tue, 30 Sep 2003 19:34:28 -0400 Subject: compartmentalizing INADDR_ANY From: Shaya Potter To: linux-kernel@vger.kernel.org Content-Type: text/plain Message-Id: <1064964305.14962.13.camel@zaphod> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.4.4 Date: Tue, 30 Sep 2003 19:25:06 -0400 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org I'm trying to work on a project to compartmentalize INADDR_ANY so that an application doesn't have to be modified, but can run in an environment where it has less access to machine resources. It's mostly proof of concept/researchy so solution can be ugly for now, but what would I have to do to effectively bind a single socket to multiple adapters, so that one can mimic INADDR_ANY behavior? I'm not anywhere comfortable with the networking code yet, but in looking through the code briefly, it seems like there are these "hash buckets" that get searched through to see if/what socket is bound to a particular port. Is it as simple as mucking with those internals and having multiple of those buckets point to the same socket? has anyone done anything like this b4, or even thought about it? oh, and I'd appreciate being CC'd as my l-k backlog is back to over 10k message :( thanks, shaya - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/