From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-bugs-return-128001-listarch-gcc-bugs=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 26739 invoked by alias); 24 Jan 2005 09:43:28 -0000
Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Archive: <http://gcc.gnu.org/ml/gcc-bugs/>
List-Post: <mailto:gcc-bugs@gcc.gnu.org>
List-Help: <mailto:gcc-bugs-help@gcc.gnu.org>
Sender: gcc-bugs-owner@gcc.gnu.org
Received: (qmail 26688 invoked by uid 48); 24 Jan 2005 09:43:23 -0000
Date: Mon, 24 Jan 2005 09:43:00 -0000
Message-ID: <20050124094323.26687.qmail@sourceware.org>
From: "rguenth at tat dot physik dot uni-tuebingen dot de" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
In-Reply-To: <20050112160800.19401.rguenth@tat.physik.uni-tuebingen.de>
References: <20050112160800.19401.rguenth@tat.physik.uni-tuebingen.de>
Reply-To: gcc-bugzilla@gcc.gnu.org
Subject: [Bug tree-optimization/19401] Trivial loop not unrolled
X-Bugzilla-Reason: CC
X-SW-Source: 2005-01/txt/msg03482.txt.bz2
List-Id: <gcc-bugs.sourceware.org>


------- Additional Comments From rguenth at tat dot physik dot uni-tuebingen dot de  2005-01-24 09:43 -------
Another one - matrix multiplication:

/* A [NxM], B [MxP] */
#define DOLOOP(N, M, P) \
void matmul ## N ## M ## P(double *res, const double *A, const double *B) \
{ \
        int i,j,k; \
        for (k=0; k<P; ++k) \
                for (i=0; i<N; ++i) { \
                        double s = 0.0; \
                        for (j=0; j<M; ++j) \
                                s += A[i*M+j] * B[j*P+k]; \
                        res[i*P+k] = s; \
                } \
}

DOLOOP(1, 1, 1)
DOLOOP(2, 1, 2)
DOLOOP(1, 2, 1)
DOLOOP(2, 2, 2)
DOLOOP(1, 3, 1)
DOLOOP(1, 1024, 1)

all up to 2x2 should be profitable to completely unroll.  Be sure to unroll
one-time rolling loops like for the last case.

Zdeneks patch only does not unroll the DOLOOP(2, 2, 2) case at -O2. Good.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19401