From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754718AbYDUGyK (ORCPT ); Mon, 21 Apr 2008 02:54:10 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751703AbYDUGxx (ORCPT ); Mon, 21 Apr 2008 02:53:53 -0400 Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:45330 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1751438AbYDUGxw (ORCPT ); Mon, 21 Apr 2008 02:53:52 -0400 Date: Sun, 20 Apr 2008 23:53:54 -0700 (PDT) Message-Id: <20080420.235354.104901823.davem@davemloft.net> To: mingo@elte.hu Cc: linville@tuxdriver.com, tomas.winkler@intel.com, linux-kernel@vger.kernel.org, kaber@trash.net, torvalds@linux-foundation.org, akpm@linux-foundation.org, netdev@vger.kernel.org, netfilter-devel@vger.kernel.org, mabbas@linux.intel.com, ischram@telenet.be, rjw@sisk.pl Subject: Re: [build bug] nf_nat_proto_sctp.c:(.text+0x6d173): undefined reference to `crc32c_le' From: David Miller In-Reply-To: <20080421062244.GA862@elte.hu> References: <20080421062244.GA862@elte.hu> X-Mailer: Mew version 5.2 on Emacs 22.1 / 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 List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Ingo Molnar Date: Mon, 21 Apr 2008 08:22:44 +0200 > net/built-in.o: In function `sctp_manip_pkt': > nf_nat_proto_sctp.c:(.text+0x6d173): undefined reference to `crc32c_le' > nf_nat_proto_sctp.c:(.text+0x6d180): undefined reference to `crc32c_le' > nf_nat_proto_sctp.c:(.text+0x6d196): undefined reference to `crc32c_le' > nf_nat_proto_sctp.c:(.text+0x6d1c6): undefined reference to `crc32c_le' > > http://redhat.com/~mingo/misc/config-Mon_Apr_21_01_52_14_CEST_2008.bad Thanks for your report Ingo. Patrick, I'm not so sure how to fix this, or even what's wrong. NF_NAT_PROTO_SCTP does a proper select on LIBCRC32C, but somehow NF_NAT_PROTO_SCTP=y and LIBCRC32C=m. I notice that IP_SCTP=m in this config. So perhaps the issue is the case where two features do a select on an infrastructure config option. Maybe whoever does the last 'select' gets to decide if the option gets set to 'y' or 'm'? It's just a guess, as I still for the life of me can't figure out how these dependencies get sorted by the kbuild system in situations like this.