From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 24 Jan 2003 14:54:57 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 24 Jan 2003 14:54:57 -0500 Received: from pizda.ninka.net ([216.101.162.242]:12012 "EHLO pizda.ninka.net") by vger.kernel.org with ESMTP id ; Fri, 24 Jan 2003 14:54:56 -0500 Date: Fri, 24 Jan 2003 11:52:52 -0800 (PST) Message-Id: <20030124.115252.24531266.davem@redhat.com> To: bgoglin@ens-lyon.fr Cc: linux-kernel@vger.kernel.org Subject: Re: AH transformation broken since 2.5.56 From: "David S. Miller" In-Reply-To: <20030124100530.GA32263@ens-lyon.fr> References: <20030124100530.GA32263@ens-lyon.fr> X-FalunGong: Information control. X-Mailer: Mew version 2.1 on Emacs 21.1 / Mule 5.0 (SAKAKI) Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org From: Brice Goglin Date: Fri, 24 Jan 2003 11:05:30 +0100 My problem was based on the fact that you can disable CONFIG_CRYPTO_HMAC by disabling CONFIG_CRYPTO. But this will not disable CONFIG_INET_AH. Shouldn't there be a fix in dependencies between CONFIG_CRYPTO and CONFIG_CRYPTO_HMAC, or between CONFIG_INET_AH and CONFIG_CRYPTO ? If you override the defaults, the responsibility lands in your hands to do the right thing. The only facility we have right now is to choose the defaults sensibly for you, and if you look at crypto/Kconfig we are doing exactly that. It checks there fore whether AH or ESP have been enabled, and chooses a default based upon that. Also, CRYPTO selection comes after the ipsec choices. So the only thing we can do is make decisions based upon whether you've enabled AH or ESP not the other way around. Whether there should be a way to FORCE config options on or off (instead of controlling the default) to avoid situations like this is a seperate topic.