From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 1759 invoked by alias); 31 Jul 2012 06:24:37 -0000 Received: (qmail 1744 invoked by uid 22791); 31 Jul 2012 06:24:35 -0000 X-SWARE-Spam-Status: No, hits=-2.7 required=5.0 tests=AWL,BAYES_00,KHOP_THREADED X-Spam-Check-By: sourceware.org Received: from nat28.tlf.novell.com (HELO nat28.tlf.novell.com) (130.57.49.28) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Tue, 31 Jul 2012 06:24:22 +0000 Received: from EMEA1-MTA by nat28.tlf.novell.com with Novell_GroupWise; Tue, 31 Jul 2012 07:24:20 +0100 Message-Id: <5017965C020000780009170B@nat28.tlf.novell.com> Date: Tue, 31 Jul 2012 06:24:00 -0000 From: "Jan Beulich" To: "H.J. Lu" Cc: Subject: Re: [PATCH] gas/x86-64: properly distinguish low and high register ranges References: <500EC9B00200007800090382@nat28.tlf.novell.com> <500FB4C4020000780009058B@nat28.tlf.novell.com> In-Reply-To: Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Mailing-List: contact binutils-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: binutils-owner@sourceware.org X-SW-Source: 2012-07/txt/msg00302.txt.bz2 >>> On 30.07.12 at 18:04, "H.J. Lu" wrote: > On Tue, Jul 24, 2012 at 11:56 PM, Jan Beulich wrote: >>>>> On 24.07.12 at 16:16, "H.J. Lu" wrote: >>> Can you add some testcases? >> >> I knew you would ask this, but sorry, this makes no sense - if test >> cases would are desirable here, they shouldn't be testing just the >> things that this patch fixes, but also any other invalid operand >> combinations. As an example - why would testing that "xlat [r11]" >> isn't accepted be needed, but not e.g. "xlat [ecx]"? >> >> Furthermore, this fixes actually broken behavior, so accepting >> the change shouldn't be dependent upon test case availability. >=20 > What broken behavior does this change fix? I gave an example above - xlat [r11]. Other similar examples involve other string instructions requiring fixed registers as well as the one or two instructions requiring xmm0/ymm0 as their first/last operand. > Why hasn't it been tested in the testsuite? Don't know. Jan