From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-143248-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 14375 invoked by alias); 26 Dec 2007 15:00:18 -0000
Received: (qmail 14363 invoked by uid 22791); 26 Dec 2007 15:00:17 -0000
X-Spam-Check-By: sourceware.org
Received: from rv-out-0910.google.com (HELO rv-out-0910.google.com) (209.85.198.185)     by sourceware.org (qpsmtpd/0.31) with ESMTP; Wed, 26 Dec 2007 15:00:05 +0000
Received: by rv-out-0910.google.com with SMTP id g11so1780064rvb.56         for <gcc@gcc.gnu.org>; Wed, 26 Dec 2007 07:00:03 -0800 (PST)
Received: by 10.140.147.18 with SMTP id u18mr3336426rvd.267.1198681203395;         Wed, 26 Dec 2007 07:00:03 -0800 (PST)
Received: from ?192.168.0.123? ( [222.244.188.69])         by mx.google.com with ESMTPS id c3sm5389175rvf.32.2007.12.26.06.59.50         (version=TLSv1/SSLv3 cipher=RC4-MD5);         Wed, 26 Dec 2007 07:00:01 -0800 (PST)
Message-ID: <47726C5B.4050701@gmail.com>
Date: Wed, 26 Dec 2007 17:38:00 -0000
From: Qing Wei <dragonylffly@gmail.com>
User-Agent: Thunderbird 2.0.0.9 (Windows/20071031)
MIME-Version: 1.0
To: Tim Prince <tprince@computer.org>, gcc@gcc.gnu.org
Subject: Re: How to describe a FMAC insn
References: <4771EDE8.9030709@gmail.com> <477201FF.10802@computer.org>
In-Reply-To: <477201FF.10802@computer.org>
Content-Type: text/plain; charset=GB2312
Content-Transfer-Encoding: 7bit
X-IsSubscribed: yes
Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc/>
List-Post: <mailto:gcc@gcc.gnu.org>
List-Help: <http://gcc.gnu.org/ml/>
Sender: gcc-owner@gcc.gnu.org
X-SW-Source: 2007-12/txt/msg00688.txt.bz2

I tried by referring the ia64.md, unfortunately it does not work. The
insn I wrote for FMAC is as follows,

(define_insn "maddsi4"
[(set (match_operand:SI 0 "register_operand" "=r")
(plus:SI (mult:SI (match_operand:SI 1 "register_operand" "r")
(match_operand:SI 2 "register_operand" "r"))
(match_operand:SI 3 "register_operand" "r")))]
""
"fma %0, %1, %2, %3")

And besides this, I defined other two insns for dedicated add and mult
operations as follows,

(define_insn "addsi3"
[(set (match_operand:SI 0 "register_operand" "=r")
(plus:SI (match_operand:SI 1 "register_operand" "r")
(match_operand:SI 2 "register_operand" "r"))
)]
""
"add %0, %1, %2")

(define_insn "mulsi3"
[(set (match_operand:SI 0 "register_operand" "=r")
(mult:SI (match_operand:SI 1 "register_operand" "r")
(match_operand:SI 2 "register_operand" "r"))
)]
""
"mul %0, %1, %2")

It seems trivial. But after I rebuilt GCC for this new target, I found
that no optabs entry is initialized for maddsi4 in insn-opinit.c which
is generated by genopinit. However, the add_optab and smul_optab do be
initialized with Code_for_addsi3/mulsi3. As a result, when I test the
following simple program, cc1 produces separate add and mul instructions
rather than fma, where the problem is? Thanks.

void f(int s1[], int s2[], int s3[], int s4[])
{
int j;

for (j = 0; j < 16; j++)
s4[j] = s1[j]*s2[j]+s3[j];
}

Qing
>> Could someone give some hints of how to describe a FMAC (float mult and
>> add) insn in machine description, it matches d = b*c+a, which is a four
>> operands float instrution. 
>>     
> There are plenty of examples in ia64.md and rs6000.md.
>