OpenBLAS/kernel/loongarch64/camax_lasx.S

213 lines
4.9 KiB
ArmAsm

/***************************************************************************
Copyright (c) 2023, 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.
*****************************************************************************/
#define ASSEMBLER
#include "common.h"
#define N $r4
#define X $r5
#define INCX $r6
#define I $r12
#define t1 $f14
#define t2 $f18
#define t3 $f15
#define t4 $f17
#define s1 $f22
#define s2 $f9
#define s3 $f10
#define s4 $f11
#define TEMP $r16
#define a0 $f20
#define a1 $f21
#define x1 $xr9
#define x2 $xr10
#define x3 $xr11
#define x4 $xr12
#define VT0 $xr13
#define VT1 $xr14
#define res0 $xr18
#define neg1 $xr19
#define VX0 $xr20
#define VX1 $xr21
#define VM0 $xr22
#define VM1 $xr23
PROLOGUE
xvxor.v VM0, VM0, VM0
xvxor.v res0, res0, res0
bge $r0, N, .L999
bge $r0, INCX, .L999
li.d TEMP, 1
slli.d TEMP, TEMP, ZBASE_SHIFT
slli.d INCX, INCX, ZBASE_SHIFT
srai.d I, N, 3
bne INCX, TEMP, .L20
bge $r0, I, .L23
.align 3
.L10:
xvld VX0, X, 0
xvld VX1, X, 32
#ifdef DOUBLE
xvpickev.d x1, VX1, VX0
xvpickod.d x2, VX1, VX0
#else
xvpickev.w x1, VX1, VX0
xvpickod.w x2, VX1, VX0
#endif
XVFSUB x3, res0, x1
XVFSUB x4, res0, x2
XVFMAX x1, x1, x3
XVFMAX x2, x2, x4
XVFADD VM1, x1, x2
XVFMAX VM0, VM0, VM1
#ifdef DOUBLE
xvld VX0, X, 64
xvld VX1, X, 96
xvpickev.d x1, VX1, VX0
xvpickod.d x2, VX1, VX0
XVFSUB x3, res0, x1
XVFSUB x4, res0, x2
XVFMAX x1, x1, x3
XVFMAX x2, x2, x4
XVFADD VM1, x1, x2
XVFMAX VM0, VM0, VM1
#endif
addi.d I, I, -1
addi.d X, X, 16 * SIZE
blt $r0, I, .L10
.align 3
.L11:
#ifdef DOUBLE
xvpickve.d x1, VM0, 0
xvpickve.d x2, VM0, 1
XVFMAX VM0, x1, x2
#else
xvpickve.w x1, VM0, 0
xvpickve.w x2, VM0, 1
xvpickve.w x3, VM0, 2
xvpickve.w x4, VM0, 3
XVFMAX VM0, x1, x2
XVFMAX VM1, x3, x4
XVFMAX VM0, VM0, VM1
#endif
b .L23
.align 3
.L20: // INCX!=1
bge $r0, I, .L23
.align 3
.L21:
LD t1, X, 0 * SIZE
LD t2, X, 1 * SIZE
add.d X, X, INCX
LD t3, X, 0 * SIZE
LD t4, X, 1 * SIZE
add.d X, X, INCX
FABS t1, t1
FABS t2, t2
FABS t3, t3
FABS t4, t4
ADD t1, t1, t2
ADD t3, t3, t4
FMAX s1, t1, t3
LD t1, X, 0 * SIZE
LD t2, X, 1 * SIZE
add.d X, X, INCX
LD t3, X, 0 * SIZE
LD t4, X, 1 * SIZE
add.d X, X, INCX
FABS t1, t1
FABS t2, t2
FABS t3, t3
FABS t4, t4
ADD t1, t1, t2
ADD t3, t3, t4
FMAX s1, t1, t3
LD t1, X, 0 * SIZE
LD t2, X, 1 * SIZE
add.d X, X, INCX
LD t3, X, 0 * SIZE
LD t4, X, 1 * SIZE
add.d X, X, INCX
FABS t1, t1
FABS t2, t2
FABS t3, t3
FABS t4, t4
addi.d I, I, -1
ADD t1, t1, t2
ADD t3, t3, t4
FMAX s3, t1, t3
LD t1, X, 0 * SIZE
LD t2, X, 1 * SIZE
add.d X, X, INCX
LD t3, X, 0 * SIZE
LD t4, X, 1 * SIZE
add.d X, X, INCX
FABS t1, t1
FABS t2, t2
FABS t3, t3
FABS t4, t4
ADD t1, t1, t2
ADD t3, t3, t4
FMAX s4, t1, t3
blt $r0, I, .L21
.align 3
.L22:
FMAX s1, s1, s2
FMAX s3, s3, s4
FMAX s1, s1, s3
.align 3
.L23: //N<8
andi I, N, 7
bge $r0, I, .L999
.align 3
.L24:
LD a0, X, 0 * SIZE
LD a1, X, 1 * SIZE
addi.d I, I, -1
FABS a0, a0
FABS a1, a1
ADD a0, a0, a1
add.d X, X, INCX
FMAX s1, a0, s1
blt $r0, I, .L24
.align 3
.L999:
MOV $f0, $f22
jirl $r0, $r1, 0x0
.align 3
EPILOGUE