From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <cgen-return-1788-listarch-cgen=sources.redhat.com@sources.redhat.com>
Received: (qmail 16083 invoked by alias); 9 Sep 2005 09:08:56 -0000
Mailing-List: contact cgen-help@sources.redhat.com; run by ezmlm
Precedence: bulk
List-Subscribe: <mailto:cgen-subscribe@sources.redhat.com>
List-Archive: <http://sources.redhat.com/ml/cgen/>
List-Post: <mailto:cgen@sources.redhat.com>
List-Help: <mailto:cgen-help@sources.redhat.com>, <http://sources.redhat.com/lists.html#faqs>
Sender: cgen-owner@sources.redhat.com
Received: (qmail 15427 invoked by uid 22791); 9 Sep 2005 09:08:28 -0000
Received: from multi.imgtec.com (HELO multi.imgtec.com) (194.200.65.239)
    by sourceware.org (qpsmtpd/0.30-dev) with ESMTP; Fri, 09 Sep 2005 09:08:28 +0000
X-SEF-Processed: 5_0_0_812__2005_09_09_10_08_18
Received: from Unknown [192.168.5.239] by multi.imgtec.com - SurfControl E-mail Filter (5.0); Fri, 09 Sep 2005 10:08:18 +0100
Received: from lemail1.le.imgtec.org ([192.168.152.65]) by klmail1.kl.imgtec.org with Microsoft SMTPSVC(6.0.3790.1830);
	 Fri, 9 Sep 2005 10:08:18 +0100
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Subject: RE: Constraints between operands
Date: Fri, 09 Sep 2005 09:08:00 -0000
Message-ID: <0D107966AF6D79418315B7C5549F4B5104DFDF@lemail1.le.imgtec.org>
From: "Will Newton" <will.newton@imgtec.com>
To: "Frank Ch. Eigler" <fche@redhat.com>
Cc: <cgen@sources.redhat.com>
X-SW-Source: 2005-q3/txt/msg00021.txt.bz2


> > [...]
> > I'm using cgen to write a binutils port for a processor. I've come=20
> > across a problem I haven't been able to solve yet with expressing a=20
> > constraint between a number of fields in an instruction.
> > [...]
> > ADD D0.1,D0.2 ; Data unit 0 for both regs, OK ADD D0.1,D1.2 ; Data=20
> > unit mismatch, error!
> > [...]
>=20
> One way may be to write a custom operand parser for the=20
> second D slot, which would enforce this constraint.  It would=20
> signal a parse error.

I could not see an easy way to reference a previous field. The prototype
of parse handlers is like:

static const char *
parse_hi16 (cd, strp, opindex, valuep)
     CGEN_CPU_DESC cd;                  /* CPU description */
     const char **strp;                 /* Current position in input
text */
     int opindex;                       /* ??? */
     unsigned long *valuep;             /* Result */

In order to find what the previous operand's data unit was the only way
I can see is to rewind strp and parse the input text, which may or may
not work and seems quite nasty. Or is there an easier way?