From mboxrd@z Thu Jan 1 00:00:00 1970 From: Harald Welte Subject: Re: [PATCH net-next 03/14] gtp: Call common functions to get tunnel routes and add dst_cache Date: Tue, 19 Sep 2017 20:09:42 +0800 Message-ID: <20170919120942.dpy5kmkhzws7pqd5@nataraja> References: <20170919003904.5124-1-tom@quantonium.net> <20170919003904.5124-4-tom@quantonium.net> <20170918.211751.1871429944584121281.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: tom@quantonium.net, netdev@vger.kernel.org, pablo@netfilter.org, rohit@quantonium.net To: David Miller Return-path: Received: from ganesha.gnumonks.org ([213.95.27.120]:45320 "EHLO ganesha.gnumonks.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751578AbdISMNN (ORCPT ); Tue, 19 Sep 2017 08:13:13 -0400 Content-Disposition: inline In-Reply-To: <20170918.211751.1871429944584121281.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: Hi Dave, On Mon, Sep 18, 2017 at 09:17:51PM -0700, David Miller wrote: > This and the new dst caching code ignores any source address selection > done by ip_route_output_key() or the new tunnel route lookup helpers. > > Either source address selection should be respected, or if saddr will > never be modified by a route lookup for some specific reason here, > that should be documented. The IP source address is fixed by signaling on the GTP-C control plane and nothing that the kernel can unilaterally decide to change. Such a change of address would have to be decided by and first be signaled on GTP-C to the peer by the userspace daemon, which would then update the PDP context in the kernel. So I guess you're asking us to document that rationale as form of a source code comment ? -- - Harald Welte http://laforge.gnumonks.org/ ============================================================================ "Privacy in residential applications is a desirable marketing option." (ETSI EN 300 175-7 Ch. A6)