From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 15778 invoked by alias); 27 Dec 2001 14:30:20 -0000 Mailing-List: contact gdb-patches-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sources.redhat.com Received: (qmail 15735 invoked from network); 27 Dec 2001 14:30:13 -0000 Received: from unknown (HELO mail.science.uva.nl) (146.50.4.51) by sources.redhat.com with SMTP; 27 Dec 2001 14:30:13 -0000 Received: from soliton.wins.uva.nl [146.50.20.20] by mail.science.uva.nl with ESMTP (sendmail 8.11.6/config 11.24). id fBRESbL19127; Thu, 27 Dec 2001 15:28:37 +0100 (MET) Received: from localhost by soliton.wins.uva.nl (sendmail 8.11.6/config 11.15). id fBRESo320021; Thu, 27 Dec 2001 15:28:50 +0100 (MET) X-Organisation: Faculty of Science, University of Amsterdam, The Netherlands X-URL: http://www.science.uva.nl/ To: gdb-patches Subject: Re: [RFC]: Use `gdb_compile' instead of `target_link' for linking asm-source testcase References: <20011218145813.Y21898@cygbert.vinschen.de> <3C1F8D1C.77D4D94E@redhat.com> <20011220123119.P21898@cygbert.vinschen.de> <3C21ED48.938C6556@redhat.com> <20011220170841.T21898@cygbert.vinschen.de> From: Mark Kettenis Date: Thu, 27 Dec 2001 06:30:00 -0000 In-Reply-To: Corinna Vinschen's message of Thu, 20 Dec 2001 17:08:41 +0100 Message-ID: X-Mailer: Gnus v5.5/Emacs 20.2 X-SW-Source: 2001-12/txt/msg00585.txt.bz2 Corinna Vinschen writes: > On Thu, Dec 20, 2001 at 08:53:12AM -0500, Fernando Nasser wrote: > > Corinna Vinschen wrote: > > > > > > On Tue, Dec 18, 2001 at 01:38:20PM -0500, Fernando Nasser wrote: > > > > This looks reasonable to me. > > > > > > > > Fernando > > > > > > Since there's no further discussion or any complaints, does that mean > > > the patch is ok and approved? > > > > > > > Yes. Thanks. > > Thanks, applied. Unfortunately this breaks the test on FreeBSD 4.4. gdb_compile automatically links against libc, and on FreeBSD that makes the link fail since libc depends on certain symbols in crt0.o, which we don't link in because of -nostartfiles. Adding -nostdlib isn't an option since gdb_compile also adds -lm which depends on libc. Removing -nostartfiles isn't an option either since it would result in confilicting symbols. Instead of reverting Corrina's patch, we could add "-static" to ldflags (see the attached patch). Opinions? Index: testsuite/ChangeLog from Mark Kettenis * gdb.asm/asm-source.exp: Add ldflags=-static for gdb_compile. Index: testsuite/gdb.asm/asm-source.exp =================================================================== RCS file: /cvs/src/src/gdb/testsuite/gdb.asm/asm-source.exp,v retrieving revision 1.16 diff -u -p -r1.16 asm-source.exp --- testsuite/gdb.asm/asm-source.exp 2001/12/20 22:12:20 1.16 +++ testsuite/gdb.asm/asm-source.exp 2001/12/27 14:27:51 @@ -81,7 +81,7 @@ if {[target_assemble ${src2} asmsrc2.o " gdb_suppress_entire_file "Testcase compile failed, so all tests in this file will automatically fail." } -set opts "debug ldflags=-nostartfiles" +set opts "debug ldflags=-nostartfiles ldflags=-static" foreach i ${link-flags} { append opts " ldflags=$i" }