From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 16513 invoked by alias); 21 Oct 2007 13:05:14 -0000 Received: (qmail 16380 invoked by uid 22791); 21 Oct 2007 13:05:12 -0000 X-Spam-Check-By: sourceware.org Received: from NaN.false.org (HELO nan.false.org) (208.75.86.248) by sourceware.org (qpsmtpd/0.31) with ESMTP; Sun, 21 Oct 2007 13:05:07 +0000 Received: from nan.false.org (localhost [127.0.0.1]) by nan.false.org (Postfix) with ESMTP id A098C9833D; Sun, 21 Oct 2007 13:05:05 +0000 (GMT) Received: from caradoc.them.org (22.svnf5.xdsl.nauticom.net [209.195.183.55]) by nan.false.org (Postfix) with ESMTP id 7AF079829E; Sun, 21 Oct 2007 13:05:05 +0000 (GMT) Received: from drow by caradoc.them.org with local (Exim 4.68) (envelope-from ) id 1IjaUK-0002vn-7T; Sun, 21 Oct 2007 09:05:04 -0400 Date: Sun, 21 Oct 2007 18:08:00 -0000 From: Daniel Jacobowitz To: Ulrich Weigand Cc: gdb-patches@sourceware.org Subject: Re: [commit] Use -mabi=altivec for AltiVec tests Message-ID: <20071021130504.GA10828@caradoc.them.org> Mail-Followup-To: Ulrich Weigand , gdb-patches@sourceware.org References: <200710211235.l9LCZ2Mk002698@d12av02.megacenter.de.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200710211235.l9LCZ2Mk002698@d12av02.megacenter.de.ibm.com> User-Agent: Mutt/1.5.15 (2007-04-09) X-IsSubscribed: yes Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2007-10/txt/msg00473.txt.bz2 On Sun, Oct 21, 2007 at 02:35:02PM +0200, Ulrich Weigand wrote: > Hello, > > both altivec-abi.exp and altivec-regs.exp had a number of test cases > failing because the test case source code was compiled without using > the -mabi=altivec flag. As GDB expects the inferior to use the > AltiVec ABI, we need to compile the test case accordingly ... > > Committed to mainline. I don't think this is a good idea. -mabi=altivec is an ABI-changing option; you're supposed to match the ABI of your system libraries. It will work out OK in practice though since the stack alignment is already large enough on powerpc-linux. I committed a GCC HEAD patch several weeks ago to mark the vector ABI of PowerPC object files. I also posted a linker patch to propogate this into linked objects and I have the matching patch to GDB to auto-detect the correct ABI and support both in GDB. That also fixes this test. Unfortunately the linker patch is hung up right now because it notices this same dubious action (linking -mabi=altivec and -mabi=no-altivec code) and warns about it. That causes a bunch of GCC testcase failures. The GCC testcases in question use, but do not actually need, -mabi=altivec; they just need a nasty bug in -maltivec -mabi=no-altivec fixed and my patch for that didn't get any comments :-( Maybe I should commit the linker patch without the warning? -- Daniel Jacobowitz CodeSourcery