| 12
 3
 4
 5
 6
 7
 8
 9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
 100
 
 | /*
 * Copyright (c) 2015, 2023, Oracle and/or its affiliates. All rights reserved.
 * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
 *
 * This code is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License version 2 only, as
 * published by the Free Software Foundation.
 *
 * This code is distributed in the hope that it will be useful, but WITHOUT
 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
 * version 2 for more details (a copy is included in the LICENSE file that
 * accompanied this code).
 *
 * You should have received a copy of the GNU General Public License version
 * 2 along with this work; if not, write to the Free Software Foundation,
 * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
 *
 * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
 * or visit www.oracle.com if you need additional information or have any
 * questions.
 */
/**
 * @test
 * @bug 8074981
 * @summary Add C2 x86 Superword support for scalar product reduction optimizations : float test
 * @library /test/lib /
 * @run driver compiler.loopopts.superword.ProdRed_Float
 */
package compiler.loopopts.superword;
import compiler.lib.ir_framework.*;
public class ProdRed_Float {
    public static void main(String[] args) throws Exception {
        TestFramework framework = new TestFramework();
        framework.addFlags("-XX:+IgnoreUnrecognizedVMOptions",
                           "-XX:LoopUnrollLimit=250",
                           "-XX:CompileThresholdScaling=0.1");
        int i = 0;
        Scenario[] scenarios = new Scenario[6];
        for (String reductionSign : new String[] {"+", "-"}) {
            for (int maxUnroll : new int[] {4, 8, 16}) {
                scenarios[i] = new Scenario(i, "-XX:" + reductionSign + "SuperWordReductions",
                                               "-XX:LoopMaxUnroll=" + maxUnroll);
                i++;
            }
        }
        framework.addScenarios(scenarios);
        framework.start();
    }
    @Run(test = {"prodReductionImplement"},
         mode = RunMode.STANDALONE)
    public void runTests() throws Exception {
        float[] a = new float[256 * 1024];
        float[] b = new float[256 * 1024];
        prodReductionInit(a, b);
        float valid = 2000;
        float total = 0;
        for (int j = 0; j < 2000; j++) {
            total = j + 1;
            total = prodReductionImplement(a, b, total);
        }
        if (total == valid) {
            System.out.println("Success");
        } else {
            System.out.println("Invalid sum of elements variable in total: " + total);
            System.out.println("Expected value = " + valid);
            throw new Exception("Failed");
        }
    }
    public static void prodReductionInit(float[] a, float[] b) {
        for (int i = 0; i < a.length; i++) {
            a[i] = i + 2;
            b[i] = i + 1;
        }
    }
    @Test
    @IR(applyIf = {"SuperWordReductions", "false"},
        failOn = {IRNode.MUL_REDUCTION_VF})
    @IR(applyIfCPUFeature = {"sse4.1", "true"},
        applyIfAnd = {"SuperWordReductions", "true", "LoopMaxUnroll", ">= 8"},
        counts = {IRNode.MUL_REDUCTION_VF, ">= 1"})
    // There is no efficient way to implement strict-ordered version on riscv64.
    @IR(applyIfCPUFeature = {"rvv", "true"},
        applyIf = {"SuperWordReductions", "true"},
        failOn = {IRNode.MUL_REDUCTION_VF})
    public static float prodReductionImplement(float[] a, float[] b, float total) {
        for (int i = 0; i < a.length; i++) {
            total *= a[i] - b[i];
        }
        return total;
    }
}
 |