From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1MTbNl-00071P-Du for qemu-devel@nongnu.org; Wed, 22 Jul 2009 08:57:17 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1MTbNg-00071D-OP for qemu-devel@nongnu.org; Wed, 22 Jul 2009 08:57:16 -0400 Received: from [199.232.76.173] (port=52383 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1MTbNg-00071A-I8 for qemu-devel@nongnu.org; Wed, 22 Jul 2009 08:57:12 -0400 Received: from mx20.gnu.org ([199.232.41.8]:31056) by monty-python.gnu.org with esmtps (TLS-1.0:RSA_AES_256_CBC_SHA1:32) (Exim 4.60) (envelope-from ) id 1MTbNg-0006NT-55 for qemu-devel@nongnu.org; Wed, 22 Jul 2009 08:57:12 -0400 Received: from mail-yx0-f188.google.com ([209.85.210.188]) by mx20.gnu.org with esmtp (Exim 4.60) (envelope-from ) id 1MTbNf-0007et-CP for qemu-devel@nongnu.org; Wed, 22 Jul 2009 08:57:11 -0400 Received: by yxe26 with SMTP id 26so249529yxe.4 for ; Wed, 22 Jul 2009 05:57:10 -0700 (PDT) Message-ID: <4A670CA3.6030800@codemonkey.ws> Date: Wed, 22 Jul 2009 07:57:07 -0500 From: Anthony Liguori MIME-Version: 1.0 Subject: Re: [Qemu-devel] [PATCH 00/37] Generate a proper LIBS variable References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: quintela@redhat.com Cc: qemu-devel@nongnu.org quintela@redhat.com wrote: > From: Juan Quintela > > A consequence of this patch series is that LIBS becomes unified for -softmmu and for -linux-user. While this isn't necessarily a bad thing, linux-user uses a special ld script that has the following (at least for x86_64.ld): SEARCH_DIR("/lib64"); SEARCH_DIR("/usr/lib64"); SEARCH_DIR("/usr/local/lib64"); I use Fedora which doesn't package vde, so as a consequence, I have libvdeplug.so installed in /usr/local/lib. When not explicitly specifying search paths, things link quite happily which leads me to believe that this is a valid thing for me to do. I think you need to make one of the following changes: 1) Build separate LIBS variables for -softmmu and -linux-user to restore the behavior from default 2) Figure out how to set the search dirs in the ld script to what ld normally defaults to 3) Figure out a way to use an ld script while still using the default search path In the interim, if you rebase your other series I can push those while you work out the LIBS issues. Regards, Anthony Liguori