From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:53941) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WkFLT-0004Mc-Pf for qemu-devel@nongnu.org; Tue, 13 May 2014 12:14:29 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WkFLL-0003jx-2Z for qemu-devel@nongnu.org; Tue, 13 May 2014 12:14:23 -0400 Received: from mail-ee0-x231.google.com ([2a00:1450:4013:c00::231]:64336) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WkFLK-0003js-T9 for qemu-devel@nongnu.org; Tue, 13 May 2014 12:14:15 -0400 Received: by mail-ee0-f49.google.com with SMTP id e53so567984eek.22 for ; Tue, 13 May 2014 09:14:13 -0700 (PDT) Sender: Paolo Bonzini Message-ID: <537244D0.7010707@redhat.com> Date: Tue, 13 May 2014 18:14:08 +0200 From: Paolo Bonzini MIME-Version: 1.0 References: <5372296E.60204@gmail.com> <53723B9D.1050303@msgid.tls.msk.ru> In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] Build Broken? List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Maydell , Michael Tokarev Cc: Tom Musta , QEMU Developers Il 13/05/2014 17:55, Peter Maydell ha scritto: > On 13 May 2014 16:34, Michael Tokarev wrote: >> 13.05.2014 18:17, Tom Musta wrote: >>> The current origin/master is not building for me: >>> >>> LINK qemu-img >>> qemu-img.o: In function `add_format_to_seq': >>> /bghome/tmusta/powerisa/qemu/qemu/qemu-img.c:73: undefined reference to `g_sequence_lookup' >>> collect2: ld returned 1 exit status >>> make: *** [qemu-img] Error 1 >> [] >>> It appears that this code is dependent on glib.h function that is newer than one of my build systems >>> (RHEL 6 update 5, gcc version 4.4.7 20120313 (Red Hat 4.4.7-4)) >>> >>> Is this a known limitation? i.e. is QEMU 2.1 going to prereq a newer version of glib2? >> >> If we're going to require more recent glib (g_sequence_lookup() first appeared in 2.26), >> let's require at least 2.31 with the new threading primitives as well. > > This is the wrong end to approach this question from, I think. > We should start with "which RedHat/Fedora/Debian/SuSE/etc > distro versions do we need to support?" and then can derive > the minimum glib version from that. > > In particular, if RHEL6 doesn't have a glib with > g_sequence_lookup I think it's a fairly straightforward > decision to say we can't use it. I think we should stick for RHEL5's 2.12 for a few months still, at least until RHEL7.0 or 7.1 comes out. Looks like SLES's 2.22 will be the next lowest target. Paolo