OpenBLAS/kernel/power/sgemv_t.c

/***************************************************************************
Copyright (c) 2019, The OpenBLAS Project
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
1. Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
2. Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in
the documentation and/or other materials provided with the
distribution.
3. Neither the name of the OpenBLAS project nor the names of
its contributors may be used to endorse or promote products
derived from this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
ARE DISCLAIMED. IN NO EVENT SHALL THE OPENBLAS PROJECT OR CONTRIBUTORS BE
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE
GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF
THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 *****************************************************************************/
#if !defined(__VEC__) || !defined(__ALTIVEC__)
#include "../arm/gemv_t.c"

#else

#include "common.h"

#define NBMAX 2048

#include <altivec.h>

static void sgemv_kernel_4x8(BLASLONG n, BLASLONG lda, FLOAT *ap, FLOAT *x,
                             FLOAT *y, FLOAT alpha) {
    BLASLONG i;
    FLOAT *a0, *a1, *a2, *a3, *a4, *a5, *a6, *a7;
    register __vector float temp0 = {0, 0, 0, 0};
    register __vector float temp1 = {0, 0, 0, 0};
    register __vector float temp2 = {0, 0, 0, 0};
    register __vector float temp3 = {0, 0, 0, 0};
    register __vector float temp4 = {0, 0, 0, 0};
    register __vector float temp5 = {0, 0, 0, 0};
    register __vector float temp6 = {0, 0, 0, 0};
    register __vector float temp7 = {0, 0, 0, 0};

    a0 = ap;
    a1 = ap + lda;
    a2 = a1 + lda;
    a3 = a2 + lda;
    a4 = a3 + lda;
    a5 = a4 + lda;
    a6 = a5 + lda;
    a7 = a6 + lda;

    for (i = 0; i < n; i += 4) {
        __vector float vx = vec_vsx_ld(0, &x[i]);
        __vector float vva0 = vec_vsx_ld(0, &a0[i]);
        __vector float vva1 = vec_vsx_ld(0, &a1[i]);
        __vector float vva2 = vec_vsx_ld(0, &a2[i]);
        __vector float vva3 = vec_vsx_ld(0, &a3[i]);
        __vector float vva4 = vec_vsx_ld(0, &a4[i]);
        __vector float vva5 = vec_vsx_ld(0, &a5[i]);
        __vector float vva6 = vec_vsx_ld(0, &a6[i]);
        __vector float vva7 = vec_vsx_ld(0, &a7[i]);
        temp0 += vx * vva0;
        temp1 += vx * vva1;
        temp2 += vx * vva2;
        temp3 += vx * vva3;
        temp4 += vx * vva4;
        temp5 += vx * vva5;
        temp6 += vx * vva6;
        temp7 += vx * vva7;
    }


    register __vector float t0, t1, t2, t3;
    register __vector float a = {alpha, alpha, alpha, alpha};
    __vector float vy0 = vec_vsx_ld(0, y);
    __vector float vy1 = vec_vsx_ld(0, &(y[4]));
    t0 = vec_mergeh(temp0, temp2);
    t1 = vec_mergel(temp0, temp2);
    t2 = vec_mergeh(temp1, temp3);
    t3 = vec_mergel(temp1, temp3);
    temp0 = vec_mergeh(t0, t2);
    temp1 = vec_mergel(t0, t2);
    temp2 = vec_mergeh(t1, t3);
    temp3 = vec_mergel(t1, t3);
    temp0 += temp1 + temp2 + temp3;

    t0 = vec_mergeh(temp4, temp6);
    t1 = vec_mergel(temp4, temp6);
    t2 = vec_mergeh(temp5, temp7);
    t3 = vec_mergel(temp5, temp7);
    temp4 = vec_mergeh(t0, t2);
    temp5 = vec_mergel(t0, t2);
    temp6 = vec_mergeh(t1, t3);
    temp7 = vec_mergel(t1, t3);
    temp4 += temp5 + temp6 + temp7;

    vy0 += a * temp0;
    vy1 += a * temp4;
    vec_vsx_st(vy0, 0, y);
    vec_vsx_st(vy1, 0, &(y[4]));

}

static void sgemv_kernel_4x4(BLASLONG n, BLASLONG lda, FLOAT *ap, FLOAT *x,
                             FLOAT *y, FLOAT alpha) {
    BLASLONG i = 0;
    FLOAT *a0, *a1, *a2, *a3;
    a0 = ap;
    a1 = ap + lda;
    a2 = a1 + lda;
    a3 = a2 + lda;
    register __vector float temp0 = {0, 0, 0, 0};
    register __vector float temp1 = {0, 0, 0, 0};
    register __vector float temp2 = {0, 0, 0, 0};
    register __vector float temp3 = {0, 0, 0, 0};

    for (i = 0; i < n; i += 4) {
        __vector float vx = vec_vsx_ld(0, &x[i]);
        __vector float vva0 = vec_vsx_ld(0, &a0[i]);
        __vector float vva1 = vec_vsx_ld(0, &a1[i]);
        __vector float vva2 = vec_vsx_ld(0, &a2[i]);
        __vector float vva3 = vec_vsx_ld(0, &a3[i]);
        temp0 += vx * vva0;
        temp1 += vx * vva1;
        temp2 += vx * vva2;
        temp3 += vx * vva3;
    }

    register __vector float t0, t1, t2, t3;
    register __vector float a = {alpha, alpha, alpha, alpha};
    __vector float vy0 = vec_vsx_ld(0, y);

    t0 = vec_mergeh(temp0, temp2);
    t1 = vec_mergel(temp0, temp2);
    t2 = vec_mergeh(temp1, temp3);
    t3 = vec_mergel(temp1, temp3);
    temp0 = vec_mergeh(t0, t2);
    temp1 = vec_mergel(t0, t2);
    temp2 = vec_mergeh(t1, t3);
    temp3 = vec_mergel(t1, t3);
    temp0 += temp1 + temp2 + temp3;

    vy0 += a * temp0;
    vec_vsx_st(vy0, 0, y);

}

static void sgemv_kernel_4x2(BLASLONG n, BLASLONG lda, FLOAT *ap, FLOAT *x,
                             FLOAT *y, FLOAT alpha, BLASLONG inc_y) {
    BLASLONG i;
    FLOAT *a0, *a1;
    a0 = ap;
    a1 = ap + lda;
    __vector float temp0 = {0, 0, 0, 0};
    __vector float temp1 = {0, 0, 0, 0};
    for (i = 0; i < n; i += 4) {
        __vector float vx = vec_vsx_ld(0, &x[i]);
        __vector float vva0 = vec_vsx_ld(0, &a0[i]);
        __vector float vva1 = vec_vsx_ld(0, &a1[i]);
        temp0 += vx * vva0;
        temp1 += vx * vva1;
    }

    y[0] += alpha * (temp0[0] + temp0[1] + temp0[2] + temp0[3]);
    y[inc_y] += alpha * (temp1[0] + temp1[1] + temp1[2] + temp1[3]);
}

static void sgemv_kernel_4x1(BLASLONG n, FLOAT *ap, FLOAT *x, FLOAT *y,
                             FLOAT alpha) {
    BLASLONG i;
    __vector float temp0 = {0, 0, 0, 0};
    for (i = 0; i < n; i += 4) {
        __vector float vx = vec_vsx_ld(0, &x[i]);
        __vector float vva0 = vec_vsx_ld(0, &ap[i]);
        temp0 += vx * vva0;
    }

    y[0] += alpha * (temp0[0] + temp0[1] + temp0[2] + temp0[3]);
}

static void copy_x(BLASLONG n, FLOAT *src, FLOAT *dest, BLASLONG inc_src) {
    BLASLONG i;
    for (i = 0; i < n; i++) {
        *dest++ = *src;
        src += inc_src;
    }
}

int CNAME(BLASLONG m, BLASLONG n, BLASLONG dummy1, FLOAT alpha, FLOAT *a,
          BLASLONG lda, FLOAT *x, BLASLONG inc_x, FLOAT *y, BLASLONG inc_y,
          FLOAT *buffer) {

    BLASLONG i, j, n1, m1, m2, m3, n2;
    FLOAT *a_ptr, *x_ptr, *y_ptr;
    FLOAT ybuffer[8] __attribute__((aligned(16)));
    FLOAT *xbuffer;
    if (m < 1) return (0);
    if (n < 1) return (0);

    xbuffer = buffer;

    n1 = n >> 3;
    n2 = n & 7;

    m3 = m & 3;
    m1 = m - m3;
    m2 = (m & (NBMAX - 1)) - m3;

    BLASLONG NB = NBMAX;

    while (NB == NBMAX) {
        m1 -= NB;
        if (m1 < 0) {
            if (m2 == 0) break;
            NB = m2;
        }

        y_ptr = y;
        a_ptr = a;
        x_ptr = x;

        if (inc_x != 1)
            copy_x(NB, x_ptr, xbuffer, inc_x);
        else
            xbuffer = x_ptr;

        BLASLONG lda8 = lda << 3;

        if (inc_y == 1) {
            for (i = 0; i < n1; i++) {
                sgemv_kernel_4x8(NB, lda, a_ptr, xbuffer, y_ptr, alpha);

                y_ptr += 8;
                a_ptr += lda8;
            }

        } else {
            for (i = 0; i < n1; i++) {
                ybuffer[0] = 0;
                ybuffer[1] = 0;
                ybuffer[2] = 0;
                ybuffer[3] = 0;
                ybuffer[4] = 0;
                ybuffer[5] = 0;
                ybuffer[6] = 0;
                ybuffer[7] = 0;
                sgemv_kernel_4x8(NB, lda, a_ptr, xbuffer, ybuffer, alpha);

                *y_ptr += ybuffer[0];
                y_ptr += inc_y;
                *y_ptr += ybuffer[1];
                y_ptr += inc_y;
                *y_ptr += ybuffer[2];
                y_ptr += inc_y;
                *y_ptr += ybuffer[3];
                y_ptr += inc_y;

                *y_ptr += ybuffer[4];
                y_ptr += inc_y;
                *y_ptr += ybuffer[5];
                y_ptr += inc_y;
                *y_ptr += ybuffer[6];
                y_ptr += inc_y;
                *y_ptr += ybuffer[7];
                y_ptr += inc_y;

                a_ptr += lda8;
            }
        }

        if (n2 & 4) {
            ybuffer[0] = 0;
            ybuffer[1] = 0;
            ybuffer[2] = 0;
            ybuffer[3] = 0;
            sgemv_kernel_4x4(NB, lda, a_ptr, xbuffer, ybuffer, alpha);

            a_ptr += lda << 2;

            *y_ptr += ybuffer[0];
            y_ptr += inc_y;
            *y_ptr += ybuffer[1];
            y_ptr += inc_y;
            *y_ptr += ybuffer[2];
            y_ptr += inc_y;
            *y_ptr += ybuffer[3];
            y_ptr += inc_y;
        }

        if (n2 & 2) {
            sgemv_kernel_4x2(NB, lda, a_ptr, xbuffer, y_ptr, alpha, inc_y);
            a_ptr += lda << 1;
            y_ptr += 2 * inc_y;
        }

        if (n2 & 1) {
            sgemv_kernel_4x1(NB, a_ptr, xbuffer, y_ptr, alpha);
            a_ptr += lda;
            y_ptr += inc_y;
        }

        a += NB;
        x += NB * inc_x;
    }

    if (m3 == 0) return (0);

    x_ptr = x;
    a_ptr = a;
    if (m3 == 3) {
        FLOAT xtemp0 = *x_ptr * alpha;
        x_ptr += inc_x;
        FLOAT xtemp1 = *x_ptr * alpha;
        x_ptr += inc_x;
        FLOAT xtemp2 = *x_ptr * alpha;

        FLOAT *aj = a_ptr;
        y_ptr = y;

        if (lda == 3 && inc_y == 1) {
            for (j = 0; j < (n & -4); j += 4) {
                y_ptr[j] += aj[0] * xtemp0 + aj[1] * xtemp1 + aj[2] * xtemp2;
                y_ptr[j + 1] +=
                    aj[3] * xtemp0 + aj[4] * xtemp1 + aj[5] * xtemp2;
                y_ptr[j + 2] +=
                    aj[6] * xtemp0 + aj[7] * xtemp1 + aj[8] * xtemp2;
                y_ptr[j + 3] +=
                    aj[9] * xtemp0 + aj[10] * xtemp1 + aj[11] * xtemp2;
                aj += 12;
            }

            for (; j < n; j++) {
                y_ptr[j] += aj[0] * xtemp0 + aj[1] * xtemp1 + aj[2] * xtemp2;
                aj += 3;
            }

        } else {
            if (inc_y == 1) {
                BLASLONG register lda2 = lda << 1;
                BLASLONG register lda4 = lda << 2;
                BLASLONG register lda3 = lda2 + lda;

                for (j = 0; j < (n & -4); j += 4) {
                    y_ptr[j] +=
                        *aj * xtemp0 + *(aj + 1) * xtemp1 + *(aj + 2) * xtemp2;
                    y_ptr[j + 1] += *(aj + lda) * xtemp0 +
                                    *(aj + lda + 1) * xtemp1 +
                                    *(aj + lda + 2) * xtemp2;
                    y_ptr[j + 2] += *(aj + lda2) * xtemp0 +
                                    *(aj + lda2 + 1) * xtemp1 +
                                    *(aj + lda2 + 2) * xtemp2;
                    y_ptr[j + 3] += *(aj + lda3) * xtemp0 +
                                    *(aj + lda3 + 1) * xtemp1 +
                                    *(aj + lda3 + 2) * xtemp2;
                    aj += lda4;
                }

                for (; j < n; j++) {
                    y_ptr[j] +=
                        *aj * xtemp0 + *(aj + 1) * xtemp1 + *(aj + 2) * xtemp2;
                    aj += lda;
                }

            } else {
                for (j = 0; j < n; j++) {
                    *y_ptr +=
                        *aj * xtemp0 + *(aj + 1) * xtemp1 + *(aj + 2) * xtemp2;
                    y_ptr += inc_y;
                    aj += lda;
                }
            }
        }
        return (0);
    }

    if (m3 == 2) {
        FLOAT xtemp0 = *x_ptr * alpha;
        x_ptr += inc_x;
        FLOAT xtemp1 = *x_ptr * alpha;

        FLOAT *aj = a_ptr;
        y_ptr = y;

        if (lda == 2 && inc_y == 1) {
            for (j = 0; j < (n & -4); j += 4) {
                y_ptr[j] += aj[0] * xtemp0 + aj[1] * xtemp1;
                y_ptr[j + 1] += aj[2] * xtemp0 + aj[3] * xtemp1;
                y_ptr[j + 2] += aj[4] * xtemp0 + aj[5] * xtemp1;
                y_ptr[j + 3] += aj[6] * xtemp0 + aj[7] * xtemp1;
                aj += 8;
            }

            for (; j < n; j++) {
                y_ptr[j] += aj[0] * xtemp0 + aj[1] * xtemp1;
                aj += 2;
            }

        } else {
            if (inc_y == 1) {
                BLASLONG register lda2 = lda << 1;
                BLASLONG register lda4 = lda << 2;
                BLASLONG register lda3 = lda2 + lda;

                for (j = 0; j < (n & -4); j += 4) {
                    y_ptr[j] += *aj * xtemp0 + *(aj + 1) * xtemp1;
                    y_ptr[j + 1] +=
                        *(aj + lda) * xtemp0 + *(aj + lda + 1) * xtemp1;
                    y_ptr[j + 2] +=
                        *(aj + lda2) * xtemp0 + *(aj + lda2 + 1) * xtemp1;
                    y_ptr[j + 3] +=
                        *(aj + lda3) * xtemp0 + *(aj + lda3 + 1) * xtemp1;
                    aj += lda4;
                }

                for (; j < n; j++) {
                    y_ptr[j] += *aj * xtemp0 + *(aj + 1) * xtemp1;
                    aj += lda;
                }

            } else {
                for (j = 0; j < n; j++) {
                    *y_ptr += *aj * xtemp0 + *(aj + 1) * xtemp1;
                    y_ptr += inc_y;
                    aj += lda;
                }
            }
        }
        return (0);
    }

    FLOAT xtemp = *x_ptr * alpha;
    FLOAT *aj = a_ptr;
    y_ptr = y;
    if (lda == 1 && inc_y == 1) {
        for (j = 0; j < (n & -4); j += 4) {
            y_ptr[j] += aj[j] * xtemp;
            y_ptr[j + 1] += aj[j + 1] * xtemp;
            y_ptr[j + 2] += aj[j + 2] * xtemp;
            y_ptr[j + 3] += aj[j + 3] * xtemp;
        }
        for (; j < n; j++) {
            y_ptr[j] += aj[j] * xtemp;
        }

    } else {
        if (inc_y == 1) {
            BLASLONG register lda2 = lda << 1;
            BLASLONG register lda4 = lda << 2;
            BLASLONG register lda3 = lda2 + lda;
            for (j = 0; j < (n & -4); j += 4) {
                y_ptr[j] += *aj * xtemp;
                y_ptr[j + 1] += *(aj + lda) * xtemp;
                y_ptr[j + 2] += *(aj + lda2) * xtemp;
                y_ptr[j + 3] += *(aj + lda3) * xtemp;
                aj += lda4;
            }

            for (; j < n; j++) {
                y_ptr[j] += *aj * xtemp;
                aj += lda;
            }

        } else {
            for (j = 0; j < n; j++) {
                *y_ptr += *aj * xtemp;
                y_ptr += inc_y;
                aj += lda;
            }
        }
    }

    return (0);
}

#endif