From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 28714 invoked by alias); 31 Dec 2006 19:56:54 -0000 Received: (qmail 28704 invoked by uid 22791); 31 Dec 2006 19:56:54 -0000 X-Spam-Check-By: sourceware.org Received: from nevyn.them.org (HELO nevyn.them.org) (66.93.172.17) by sourceware.org (qpsmtpd/0.31.1) with ESMTP; Sun, 31 Dec 2006 19:56:49 +0000 Received: from drow by nevyn.them.org with local (Exim 4.63) (envelope-from ) id 1H16nT-0006KU-8U; Sun, 31 Dec 2006 14:56:43 -0500 Date: Sun, 31 Dec 2006 19:56:00 -0000 From: Daniel Jacobowitz To: Paolo Bonzini Cc: gcc-patches@gcc.gnu.org, Mark Mitchell Subject: Re: libgcc move to the top level Message-ID: <20061231195643.GA24283@nevyn.them.org> Mail-Followup-To: Paolo Bonzini , gcc-patches@gcc.gnu.org, Mark Mitchell References: <45880EB3.7070101@lu.unisi.ch> <20061219191730.GA32355@nevyn.them.org> <4594FBAE.908@lu.unisi.ch> <20061229145039.GA10379@nevyn.them.org> <4595345E.308@lu.unisi.ch> <20061229161307.GA13106@nevyn.them.org> <20061229180139.GA16625@nevyn.them.org> <45962160.40707@lu.unisi.ch> <20061230201926.GA25399@nevyn.them.org> <4597DEB7.90900@lu.unisi.ch> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4597DEB7.90900@lu.unisi.ch> User-Agent: Mutt/1.5.13 (2006-08-11) 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: 2006-12/txt/msg01906.txt.bz2 On Sun, Dec 31, 2006 at 05:00:55PM +0100, Paolo Bonzini wrote: > > >It's not strictly necessary, only for the correctness tests I added to > >libgcc for the transition period. t-frv builds frvbegin.o and frvend.o > >for its own purposes; frv.h puts them in STARTFILE_SPEC and > >ENDFILE_SPEC; linux.h removes them again; but t-linux didn't remove > >them. So they are built but not used for frv-linux on trunk. > >This triggers an error because libgcc notes they are not in the libgcc > >configuration. > > Understood. Did you catch this by inspection of the t-* files, or by > building a cross? In either case, for which files/triplets did you do so? Hybrid: I configured GCC for every potentially affected target, and added a rule to gcc/Makefile.in to display the startfiles. I have bootstrapped powerpc64-linux and x86_64-linux, and run regression tests (no surprises in either). That is the current limit of my testing; I can test any additional configurations requested. -- Daniel Jacobowitz CodeSourcery