From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 18694 invoked by alias); 8 Mar 2008 16:36:35 -0000 Received: (qmail 18683 invoked by uid 22791); 8 Mar 2008 16:36:35 -0000 X-Spam-Check-By: sourceware.org Received: from merkur.ins.uni-bonn.de (HELO merkur.ins.uni-bonn.de) (131.220.223.13) by sourceware.org (qpsmtpd/0.31) with ESMTP; Sat, 08 Mar 2008 16:36:14 +0000 Received: from localhost.localdomain (xdsl-87-78-91-55.netcologne.de [87.78.91.55]) by merkur.ins.uni-bonn.de (Postfix) with ESMTP id 40C234000049F; Sat, 8 Mar 2008 17:36:11 +0100 (CET) Received: from ralf by localhost.localdomain with local (Exim 4.63) (envelope-from ) id 1JY21j-0001LB-7G; Sat, 08 Mar 2008 17:36:04 +0100 Date: Sat, 08 Mar 2008 16:36:00 -0000 From: Ralf Wildenhues To: Tom Tromey Cc: Gcc Patch List Subject: Re: Patch: automatic dependencies for gcc Message-ID: <20080308163601.GA31768@ins.uni-bonn.de> Mail-Followup-To: Ralf Wildenhues , Tom Tromey , Gcc Patch List References: <20080308143259.GA28658@ins.uni-bonn.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.17+20080114 (2008-01-14) X-IsSubscribed: yes Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-patches-owner@gcc.gnu.org X-SW-Source: 2008-03/txt/msg00539.txt.bz2 * Tom Tromey wrote on Sat, Mar 08, 2008 at 04:27:59PM CET: > >>>>> "Ralf" == Ralf Wildenhues writes: > Ralf> I ran into a build failure but it may be due to me getting the > Ralf> merge wrong. > > What happened? Hmm, I may have resolved some conflict in Makefile.in wrongly. Seems to be right now, though. > >> +COMPILE = source='$<' object='$@' libtool=no \ > >> + DEPDIR=$(DEPDIR) $(DEPMODE) $(depcomp) $(COMPILE.base) > > Ralf> This looks like $(OUTPUT_OPTION) aka '-o $@' is missing for > Ralf> non-gcc3 which means objects from subdirs end up in gcc/, > Ralf> wrongly. > [...] > > Thanks for noticing this. I will add it. > > Hmm, maybe I can fake a build forcing the use of depcomp even with > gcc. That should be possible with passing am_cv_CC_dependencies_compiler_type=gcc to gcc/configure. > Ralf> Is GCC supposed to be compilable at all by a compiler not understanding > Ralf> "-c -o" (of which there aren't very many out there any more)? > > It depends on what they are. I would prefer not to jump through hoops > here unless such a compiler is the system compiler on an "important" > platform. I've long forgotten what compilers do this, do you know? Need to look it up. I know some SCO C++ compiler did that (but that's irrelevant here). FWIW, it seems the depdirs are not distcleaned (again, still untested). Cheers, Ralf