566 lines
13 KiB
C
566 lines
13 KiB
C
/***************************************************************************
|
|
Copyright (c) 2020, 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.
|
|
*****************************************************************************/
|
|
|
|
#include "common.h"
|
|
#include <altivec.h>
|
|
|
|
typedef __vector unsigned char vec_t;
|
|
typedef FLOAT v4sf_t __attribute__ ((vector_size (16)));
|
|
typedef __vector_pair __attribute__((aligned(8))) vecp_t;
|
|
|
|
#include "dgemv_n_microk_power10.c"
|
|
|
|
#define MMA(X, APTR, ACC) \
|
|
rX = (vec_t *) & X; \
|
|
rowA = *((vecp_t*)((void*)&APTR)); \
|
|
__builtin_mma_xvf64gerpp (ACC, rowA, rX[0]);
|
|
|
|
#define SAVE(ACC, Z) \
|
|
rowC = (v4sf_t *) &y[Z]; \
|
|
__builtin_mma_disassemble_acc ((void *)result, ACC); \
|
|
result[0][1] = result[1][0]; \
|
|
result[2][1] = result[3][0]; \
|
|
rowC[0] += valpha * result[0]; \
|
|
rowC[1] += valpha * result[2];
|
|
|
|
void
|
|
dgemv_kernel_4x128 (BLASLONG n, FLOAT * a_ptr, BLASLONG lda, FLOAT * xo,
|
|
FLOAT * y, FLOAT alpha)
|
|
{
|
|
BLASLONG i, j, tmp;
|
|
FLOAT *a0 = a_ptr;
|
|
FLOAT *x1 = xo;
|
|
vector double valpha = { alpha, alpha };
|
|
v4sf_t *rowC;
|
|
__vector_quad acc0, acc1, acc2, acc3, acc4, acc5, acc6, acc7;
|
|
v4sf_t result[4];
|
|
vecp_t rowA;
|
|
vec_t *rX;
|
|
tmp = (n / 32) * 32;
|
|
for (i = 0; i < tmp; i += 32)
|
|
{
|
|
xo = x1;
|
|
a0 = a_ptr;
|
|
__builtin_mma_xxsetaccz (&acc0);
|
|
__builtin_mma_xxsetaccz (&acc1);
|
|
__builtin_mma_xxsetaccz (&acc2);
|
|
__builtin_mma_xxsetaccz (&acc3);
|
|
__builtin_mma_xxsetaccz (&acc4);
|
|
__builtin_mma_xxsetaccz (&acc5);
|
|
__builtin_mma_xxsetaccz (&acc6);
|
|
__builtin_mma_xxsetaccz (&acc7);
|
|
for (j = 0; j < 32; j++)
|
|
{
|
|
__builtin_prefetch (xo+j);
|
|
__builtin_prefetch (a0+i+j+lda);
|
|
MMA (xo[j], a0[i + 0 + j * lda], &acc0);
|
|
MMA (xo[j], a0[i + 4 + j * lda], &acc1);
|
|
MMA (xo[j], a0[i + 8 + j * lda], &acc2);
|
|
MMA (xo[j], a0[i + 12 + j * lda], &acc3);
|
|
MMA (xo[j], a0[i + 16 + j * lda], &acc4);
|
|
MMA (xo[j], a0[i + 20 + j * lda], &acc5);
|
|
MMA (xo[j], a0[i + 24 + j * lda], &acc6);
|
|
MMA (xo[j], a0[i + 28 + j * lda], &acc7);
|
|
}
|
|
xo += 32;
|
|
a0 += lda << 5;
|
|
for (j = 0; j < 32; j++)
|
|
{
|
|
__builtin_prefetch (xo+j);
|
|
__builtin_prefetch (a0+i+j+lda);
|
|
MMA (xo[j], a0[i + 0 + j * lda], &acc0);
|
|
MMA (xo[j], a0[i + 4 + j * lda], &acc1);
|
|
MMA (xo[j], a0[i + 8 + j * lda], &acc2);
|
|
MMA (xo[j], a0[i + 12 + j * lda], &acc3);
|
|
MMA (xo[j], a0[i + 16 + j * lda], &acc4);
|
|
MMA (xo[j], a0[i + 20 + j * lda], &acc5);
|
|
MMA (xo[j], a0[i + 24 + j * lda], &acc6);
|
|
MMA (xo[j], a0[i + 28 + j * lda], &acc7);
|
|
}
|
|
xo += 32;
|
|
a0 += lda << 5;
|
|
for (j = 0; j < 32; j++)
|
|
{
|
|
__builtin_prefetch (xo+j);
|
|
__builtin_prefetch (a0+i+j+lda);
|
|
MMA (xo[j], a0[i + 0 + j * lda], &acc0);
|
|
MMA (xo[j], a0[i + 4 + j * lda], &acc1);
|
|
MMA (xo[j], a0[i + 8 + j * lda], &acc2);
|
|
MMA (xo[j], a0[i + 12 + j * lda], &acc3);
|
|
MMA (xo[j], a0[i + 16 + j * lda], &acc4);
|
|
MMA (xo[j], a0[i + 20 + j * lda], &acc5);
|
|
MMA (xo[j], a0[i + 24 + j * lda], &acc6);
|
|
MMA (xo[j], a0[i + 28 + j * lda], &acc7);
|
|
}
|
|
xo += 32;
|
|
a0 += lda << 5;
|
|
for (j = 0; j < 32; j++)
|
|
{
|
|
__builtin_prefetch (xo+j);
|
|
__builtin_prefetch (a0+i+j+lda);
|
|
MMA (xo[j], a0[i + 0 + j * lda], &acc0);
|
|
MMA (xo[j], a0[i + 4 + j * lda], &acc1);
|
|
MMA (xo[j], a0[i + 8 + j * lda], &acc2);
|
|
MMA (xo[j], a0[i + 12 + j * lda], &acc3);
|
|
MMA (xo[j], a0[i + 16 + j * lda], &acc4);
|
|
MMA (xo[j], a0[i + 20 + j * lda], &acc5);
|
|
MMA (xo[j], a0[i + 24 + j * lda], &acc6);
|
|
MMA (xo[j], a0[i + 28 + j * lda], &acc7);
|
|
}
|
|
xo += 32;
|
|
a0 += lda << 5;
|
|
SAVE (&acc0, i + 0);
|
|
SAVE (&acc1, i + 4);
|
|
SAVE (&acc2, i + 8);
|
|
SAVE (&acc3, i + 12);
|
|
SAVE (&acc4, i + 16);
|
|
SAVE (&acc5, i + 20);
|
|
SAVE (&acc6, i + 24);
|
|
SAVE (&acc7, i + 28);
|
|
|
|
}
|
|
for (i = tmp; i < n; i += 4)
|
|
{
|
|
xo = x1;
|
|
a0 = a_ptr;
|
|
__builtin_mma_xxsetaccz (&acc0);
|
|
for (j = 0; j < 32; j++)
|
|
{
|
|
__builtin_prefetch (xo+j);
|
|
__builtin_prefetch (a0+i+j+lda);
|
|
MMA (xo[j], a0[i + j * lda], &acc0);
|
|
}
|
|
xo += 32;
|
|
a0 += lda << 5;
|
|
for (j = 0; j < 32; j++)
|
|
{
|
|
__builtin_prefetch (xo+j);
|
|
__builtin_prefetch (a0+i+j+lda);
|
|
MMA (xo[j], a0[i + j * lda], &acc0);
|
|
}
|
|
xo += 32;
|
|
a0 += lda << 5;
|
|
for (j = 0; j < 32; j++)
|
|
{
|
|
__builtin_prefetch (xo+j);
|
|
__builtin_prefetch (a0+i+j+lda);
|
|
MMA (xo[j], a0[i + j * lda], &acc0);
|
|
}
|
|
xo += 32;
|
|
a0 += lda << 5;
|
|
for (j = 0; j < 32; j++)
|
|
{
|
|
__builtin_prefetch (xo+j);
|
|
__builtin_prefetch (a0+i+j+lda);
|
|
MMA (xo[j], a0[i + j * lda], &acc0);
|
|
}
|
|
xo += 32;
|
|
a0 += lda << 5;
|
|
SAVE (&acc0, i);
|
|
}
|
|
}
|
|
|
|
|
|
#define NBMAX 4096
|
|
|
|
#ifndef HAVE_KERNEL_4x4
|
|
|
|
static void dgemv_kernel_4x4(BLASLONG n, FLOAT *a_ptr, BLASLONG lda, FLOAT *xo, FLOAT *y, FLOAT alpha)
|
|
{
|
|
BLASLONG i;
|
|
FLOAT x[4] __attribute__ ((aligned (16)));;
|
|
FLOAT *a0 = a_ptr;
|
|
FLOAT *a1 = a0 + lda;
|
|
FLOAT *a2 = a1 + lda;
|
|
FLOAT *a3 = a2 + lda;
|
|
|
|
|
|
for ( i=0; i<4; i++)
|
|
x[i] = xo[i] * alpha;
|
|
|
|
for ( i=0; i< n; i+=4 )
|
|
{
|
|
y[i] += a0[i]*x[0] + a1[i]*x[1] + a2[i]*x[2] + a3[i]*x[3];
|
|
y[i+1] += a0[i+1]*x[0] + a1[i+1]*x[1] + a2[i+1]*x[2] + a3[i+1]*x[3];
|
|
y[i+2] += a0[i+2]*x[0] + a1[i+2]*x[1] + a2[i+2]*x[2] + a3[i+2]*x[3];
|
|
y[i+3] += a0[i+3]*x[0] + a1[i+3]*x[1] + a2[i+3]*x[2] + a3[i+3]*x[3];
|
|
}
|
|
}
|
|
|
|
#endif
|
|
|
|
#ifndef HAVE_KERNEL_4x2
|
|
|
|
static void dgemv_kernel_4x2(BLASLONG n, FLOAT *a0, FLOAT *a1, FLOAT *xo, FLOAT *y, FLOAT alpha)
|
|
{
|
|
BLASLONG i;
|
|
FLOAT x[4] __attribute__ ((aligned (16)));;
|
|
|
|
for ( i=0; i<2; i++)
|
|
x[i] = xo[i] * alpha;
|
|
|
|
for ( i=0; i< n; i+=4 )
|
|
{
|
|
y[i] += a0[i]*x[0] + a1[i]*x[1];
|
|
y[i+1] += a0[i+1]*x[0] + a1[i+1]*x[1];
|
|
y[i+2] += a0[i+2]*x[0] + a1[i+2]*x[1];
|
|
y[i+3] += a0[i+3]*x[0] + a1[i+3]*x[1];
|
|
}
|
|
}
|
|
|
|
|
|
#endif
|
|
|
|
#ifndef HAVE_KERNEL_4x1
|
|
|
|
static void dgemv_kernel_4x1(BLASLONG n, FLOAT *a0, FLOAT *xo, FLOAT *y, FLOAT alpha)
|
|
{
|
|
BLASLONG i;
|
|
FLOAT x[4] __attribute__ ((aligned (16)));;
|
|
|
|
for ( i=0; i<1; i++)
|
|
x[i] = xo[i] * alpha;
|
|
|
|
for ( i=0; i< n; i+=4 )
|
|
{
|
|
y[i] += a0[i]*x[0];
|
|
y[i+1] += a0[i+1]*x[0];
|
|
y[i+2] += a0[i+2]*x[0];
|
|
y[i+3] += a0[i+3]*x[0];
|
|
}
|
|
}
|
|
|
|
|
|
#endif
|
|
|
|
|
|
static void add_y(BLASLONG n, FLOAT *src, FLOAT *dest, BLASLONG inc_dest)
|
|
{
|
|
BLASLONG i;
|
|
if ( inc_dest != 1 )
|
|
{
|
|
for ( i=0; i<n; i++ )
|
|
{
|
|
*dest += *src;
|
|
src++;
|
|
dest += inc_dest;
|
|
}
|
|
return;
|
|
}
|
|
|
|
}
|
|
|
|
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;
|
|
FLOAT *a_ptr;
|
|
FLOAT *x_ptr;
|
|
FLOAT *y_ptr;
|
|
BLASLONG n1;
|
|
BLASLONG m1;
|
|
BLASLONG m2;
|
|
BLASLONG m3;
|
|
BLASLONG n2;
|
|
BLASLONG lda4 = lda << 2;
|
|
BLASLONG lda128 = lda << 7;
|
|
|
|
FLOAT xbuffer[8] __attribute__ ((aligned (16)));
|
|
FLOAT *ybuffer;
|
|
|
|
if ( m < 1 ) return(0);
|
|
if ( n < 1 ) return(0);
|
|
|
|
ybuffer = buffer;
|
|
BLASLONG n128 = n >> 7;
|
|
n1 = (n - (n128 * 128)) >> 2;
|
|
n2 = (n - (n128 * 128)) & 3;
|
|
|
|
m3 = m & 3 ;
|
|
m1 = m & -4 ;
|
|
m2 = (m & (NBMAX-1)) - m3 ;
|
|
|
|
y_ptr = y;
|
|
|
|
BLASLONG NB = NBMAX;
|
|
|
|
while ( NB == NBMAX )
|
|
{
|
|
|
|
m1 -= NB;
|
|
if ( m1 < 0)
|
|
{
|
|
if ( m2 == 0 ) break;
|
|
NB = m2;
|
|
}
|
|
|
|
a_ptr = a;
|
|
x_ptr = x;
|
|
|
|
if ( inc_y != 1 )
|
|
memset(ybuffer,0,NB*8);
|
|
else
|
|
ybuffer = y_ptr;
|
|
|
|
if ( inc_x == 1 )
|
|
{
|
|
|
|
for( i = 0; i < n128 ; i++)
|
|
{
|
|
dgemv_kernel_4x128(NB,a_ptr,lda,x_ptr,ybuffer,alpha);
|
|
a_ptr += lda128;
|
|
x_ptr += 128;
|
|
}
|
|
|
|
for( i = 0; i < n1 ; i++)
|
|
{
|
|
dgemv_kernel_4x4(NB,a_ptr,lda,x_ptr,ybuffer,alpha);
|
|
a_ptr += lda4;
|
|
x_ptr += 4;
|
|
}
|
|
|
|
if ( n2 & 2 )
|
|
{
|
|
dgemv_kernel_4x2(NB,a_ptr,a_ptr+lda,x_ptr,ybuffer,alpha);
|
|
a_ptr += lda*2;
|
|
x_ptr += 2;
|
|
}
|
|
|
|
|
|
if ( n2 & 1 )
|
|
{
|
|
dgemv_kernel_4x1(NB,a_ptr,x_ptr,ybuffer,alpha);
|
|
a_ptr += lda;
|
|
x_ptr += 1;
|
|
|
|
}
|
|
|
|
|
|
}
|
|
else
|
|
{
|
|
for( i = 0; i < n128 ; i++)
|
|
{
|
|
FLOAT xbuffer[128] __attribute__ ((aligned (16)));
|
|
BLASLONG j;
|
|
for ( j = 0; j < 128 ; j++)
|
|
{
|
|
xbuffer[j] = x_ptr[0];
|
|
x_ptr += inc_x;
|
|
}
|
|
dgemv_kernel_4x128(NB,a_ptr,lda,xbuffer,ybuffer,alpha);
|
|
a_ptr += lda128;
|
|
}
|
|
|
|
for( i = 0; i < n1 ; i++)
|
|
{
|
|
xbuffer[0] = x_ptr[0];
|
|
x_ptr += inc_x;
|
|
xbuffer[1] = x_ptr[0];
|
|
x_ptr += inc_x;
|
|
xbuffer[2] = x_ptr[0];
|
|
x_ptr += inc_x;
|
|
xbuffer[3] = x_ptr[0];
|
|
x_ptr += inc_x;
|
|
dgemv_kernel_4x4(NB,a_ptr,lda,xbuffer,ybuffer,alpha);
|
|
a_ptr += lda4;
|
|
}
|
|
|
|
for( i = 0; i < n2 ; i++)
|
|
{
|
|
xbuffer[0] = x_ptr[0];
|
|
x_ptr += inc_x;
|
|
dgemv_kernel_4x1(NB,a_ptr,xbuffer,ybuffer,alpha);
|
|
a_ptr += lda;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
a += NB;
|
|
if ( inc_y != 1 )
|
|
{
|
|
add_y(NB,ybuffer,y_ptr,inc_y);
|
|
y_ptr += NB * inc_y;
|
|
}
|
|
else
|
|
y_ptr += NB ;
|
|
|
|
}
|
|
|
|
if ( m3 == 0 ) return(0);
|
|
|
|
if ( m3 == 3 )
|
|
{
|
|
a_ptr = a;
|
|
x_ptr = x;
|
|
FLOAT temp0 = 0.0;
|
|
FLOAT temp1 = 0.0;
|
|
FLOAT temp2 = 0.0;
|
|
if ( lda == 3 && inc_x ==1 )
|
|
{
|
|
|
|
for( i = 0; i < ( n & -4 ); i+=4 )
|
|
{
|
|
|
|
temp0 += a_ptr[0] * x_ptr[0] + a_ptr[3] * x_ptr[1];
|
|
temp1 += a_ptr[1] * x_ptr[0] + a_ptr[4] * x_ptr[1];
|
|
temp2 += a_ptr[2] * x_ptr[0] + a_ptr[5] * x_ptr[1];
|
|
|
|
temp0 += a_ptr[6] * x_ptr[2] + a_ptr[9] * x_ptr[3];
|
|
temp1 += a_ptr[7] * x_ptr[2] + a_ptr[10] * x_ptr[3];
|
|
temp2 += a_ptr[8] * x_ptr[2] + a_ptr[11] * x_ptr[3];
|
|
|
|
a_ptr += 12;
|
|
x_ptr += 4;
|
|
}
|
|
|
|
for( ; i < n; i++ )
|
|
{
|
|
temp0 += a_ptr[0] * x_ptr[0];
|
|
temp1 += a_ptr[1] * x_ptr[0];
|
|
temp2 += a_ptr[2] * x_ptr[0];
|
|
a_ptr += 3;
|
|
x_ptr ++;
|
|
}
|
|
|
|
}
|
|
else
|
|
{
|
|
|
|
for( i = 0; i < n; i++ )
|
|
{
|
|
temp0 += a_ptr[0] * x_ptr[0];
|
|
temp1 += a_ptr[1] * x_ptr[0];
|
|
temp2 += a_ptr[2] * x_ptr[0];
|
|
a_ptr += lda;
|
|
x_ptr += inc_x;
|
|
|
|
|
|
}
|
|
|
|
}
|
|
y_ptr[0] += alpha * temp0;
|
|
y_ptr += inc_y;
|
|
y_ptr[0] += alpha * temp1;
|
|
y_ptr += inc_y;
|
|
y_ptr[0] += alpha * temp2;
|
|
return(0);
|
|
}
|
|
|
|
|
|
if ( m3 == 2 )
|
|
{
|
|
a_ptr = a;
|
|
x_ptr = x;
|
|
FLOAT temp0 = 0.0;
|
|
FLOAT temp1 = 0.0;
|
|
if ( lda == 2 && inc_x ==1 )
|
|
{
|
|
|
|
for( i = 0; i < (n & -4) ; i+=4 )
|
|
{
|
|
temp0 += a_ptr[0] * x_ptr[0] + a_ptr[2] * x_ptr[1];
|
|
temp1 += a_ptr[1] * x_ptr[0] + a_ptr[3] * x_ptr[1];
|
|
temp0 += a_ptr[4] * x_ptr[2] + a_ptr[6] * x_ptr[3];
|
|
temp1 += a_ptr[5] * x_ptr[2] + a_ptr[7] * x_ptr[3];
|
|
a_ptr += 8;
|
|
x_ptr += 4;
|
|
|
|
}
|
|
|
|
|
|
for( ; i < n; i++ )
|
|
{
|
|
temp0 += a_ptr[0] * x_ptr[0];
|
|
temp1 += a_ptr[1] * x_ptr[0];
|
|
a_ptr += 2;
|
|
x_ptr ++;
|
|
}
|
|
|
|
}
|
|
else
|
|
{
|
|
|
|
for( i = 0; i < n; i++ )
|
|
{
|
|
temp0 += a_ptr[0] * x_ptr[0];
|
|
temp1 += a_ptr[1] * x_ptr[0];
|
|
a_ptr += lda;
|
|
x_ptr += inc_x;
|
|
|
|
|
|
}
|
|
|
|
}
|
|
y_ptr[0] += alpha * temp0;
|
|
y_ptr += inc_y;
|
|
y_ptr[0] += alpha * temp1;
|
|
return(0);
|
|
}
|
|
|
|
if ( m3 == 1 )
|
|
{
|
|
a_ptr = a;
|
|
x_ptr = x;
|
|
FLOAT temp = 0.0;
|
|
if ( lda == 1 && inc_x ==1 )
|
|
{
|
|
|
|
for( i = 0; i < (n & -4); i+=4 )
|
|
{
|
|
temp += a_ptr[i] * x_ptr[i] + a_ptr[i+1] * x_ptr[i+1] + a_ptr[i+2] * x_ptr[i+2] + a_ptr[i+3] * x_ptr[i+3];
|
|
|
|
}
|
|
|
|
for( ; i < n; i++ )
|
|
{
|
|
temp += a_ptr[i] * x_ptr[i];
|
|
}
|
|
|
|
}
|
|
else
|
|
{
|
|
|
|
for( i = 0; i < n; i++ )
|
|
{
|
|
temp += a_ptr[0] * x_ptr[0];
|
|
a_ptr += lda;
|
|
x_ptr += inc_x;
|
|
}
|
|
|
|
}
|
|
y_ptr[0] += alpha * temp;
|
|
return(0);
|
|
}
|
|
|
|
|
|
return(0);
|
|
}
|
|
|
|
|