From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 30338 invoked by alias); 5 Dec 2001 18:23:57 -0000 Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org Received: (qmail 28827 invoked from network); 5 Dec 2001 18:22:33 -0000 Received: from unknown (HELO ftpbox.mot.com) (129.188.136.101) by sources.redhat.com with SMTP; 5 Dec 2001 18:22:33 -0000 Received: [from mothost.mot.com (mothost.mot.com [129.188.137.101]) by ftpbox.mot.com (ftpbox 2.1) with ESMTP id LAA24573 for ; Wed, 5 Dec 2001 11:22:33 -0700 (MST)] Received: [from mail.wm.sps.mot.com ([199.10.246.2]) by mothost.mot.com (MOT-mothost 2.0) with ESMTP id LAA17936 for ; Wed, 5 Dec 2001 11:22:33 -0700 (MST)] Received: from hyper.wm.sps.mot.com (hyper.wm.sps.mot.com [199.10.246.43]) by mail.wm.sps.mot.com (8.9.3/8.9.3) with ESMTP id NAA06019 for ; Wed, 5 Dec 2001 13:22:27 -0500 Received: by hyper.wm.sps.mot.com (8.11.2) id fB5IMVx20055; Wed, 5 Dec 2001 13:22:31 -0500 Date: Wed, 05 Dec 2001 10:23:00 -0000 Message-Id: <200112051822.fB5IMVx20055@hyper.wm.sps.mot.com> From: Peter Barada To: gcc@gcc.gnu.org Subject: Coldfire V4E addition to gcc X-SW-Source: 2001-12/txt/msg00228.txt.bz2 I'm in the midst of adding in Coldfire v4e support to gcc-2.95.3, and I'm wondering if it isn't time to reconfigure which variant of floating point is used. Sonce I'm adding in the ColdFire FPU, this would bring the variants to four: 1) TARGET_68881 - is an FPU attached, which may be a '881. 2) TARGET_FPA - is a Sun FPA attached. 3) TARGET_68040_ONLY - subset/superset of 68881. 4) TARGET_FPU_V4E - is a ColdFire v4e FPU attached. The first three variants are used in the machine description code, header files, and C code to control the instructions generated. What I'd like to do is to split TARGET_68881 into two parts, TARGET_FPU to indicate that *any* FPU is attached, and TARGET_68881 to indicate that *only* a 68881 is attached. Then code such as: (define_expand "addsf3" [(set (match_operand:SF 0 "general_operand" "") (plus:SF (match_operand:SF 1 "general_operand" "") (match_operand:SF 2 "general_operand" "")))] "TARGET_68881 || TARGET_FPA" "") changes to: (define_expand "addsf3" [(set (match_operand:SF 0 "general_operand" "") (plus:SF (match_operand:SF 1 "general_operand" "") (match_operand:SF 2 "general_operand" "")))] "TARGET_FPU" "") instead of: (define_expand "addsf3" [(set (match_operand:SF 0 "general_operand" "") (plus:SF (match_operand:SF 1 "general_operand" "") (match_operand:SF 2 "general_operand" "")))] "TARGET_68881 || TARGET_FPA || TARGET_FPU_V4E" "") I also understand that I'll have to think of how to deal with the TARGET_68040_ONLY flag, of which the V4e supports a subset. Comments anybody? -- Peter Barada Peter.Barada@motorola.com Wizard 781-852-2768 (direct) WaveMark Solutions(wholly owned by Motorola) 781-270-0193 (fax)