tahoma2d/thirdparty/openblas/xianyi-OpenBLAS-e6e87a2/kernel/alpha/dnrm2.S
2016-03-24 02:47:04 +09:00

431 lines
7.7 KiB
ArmAsm

/*********************************************************************/
/* Copyright 2009, 2010 The University of Texas at Austin. */
/* 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. */
/* */
/* THIS SOFTWARE IS PROVIDED BY THE UNIVERSITY OF TEXAS AT */
/* AUSTIN ``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 UNIVERSITY OF TEXAS AT */
/* AUSTIN 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. */
/* */
/* The views and conclusions contained in the software and */
/* documentation are those of the authors and should not be */
/* interpreted as representing official policies, either expressed */
/* or implied, of The University of Texas at Austin. */
/*********************************************************************/
#define ASSEMBLER
#include "common.h"
#include "version.h"
#define PREFETCH_SIZE 80
#define N $16
#define X $17
#define INCX $18
#define XX $19
#define I $0
#define a0 $f0
#define a1 $f1
#define a2 $f10
#define a3 $f11
#define t0 $f12
#define t1 $f13
#define t2 $f14
#define t3 $f15
#define x0 $f16
#define x1 $f17
#define x2 $f18
#define x3 $f19
#define x4 $f20
#define x5 $f21
#define x6 $f22
#define x7 $f23
PROLOGUE
#if defined(EV4) || defined(EV5)
.frame $30,16,$26,0
.mask 0x4000000,-16
ldah $29, 0($27) !gpdisp!1
lda $29, 0($29) !gpdisp!1
lda $sp, -16($sp)
ldq $27, sqrt($29) !literal!2
stq $26, 0($sp)
PROFCODE
.prologue 1
#else
PROFCODE
#endif
fclr a0
SXADDQ INCX, 0, INCX
fclr a1
ble N, $L999
fclr a2
cmpeq INCX, SIZE, $0
fclr a3
beq $0, $L20
fclr t0
sra N, 4, I
fclr t1
ble I, $L15
fclr t2
LD x0, 0 * SIZE(X)
fclr t3
LD x1, 1 * SIZE(X)
LD x2, 2 * SIZE(X)
LD x3, 3 * SIZE(X)
LD x4, 4 * SIZE(X)
LD x5, 5 * SIZE(X)
LD x6, 6 * SIZE(X)
LD x7, 7 * SIZE(X)
lda I, -1(I)
ble I, $L12
.align 4
$L11:
addt a0, t0, a0
ldl $31, (PREFETCH_SIZE) * SIZE(X)
mult x0, x0, t0
LD x0, 8 * SIZE(X)
addt a1, t1, a1
mov X, XX
mult x1, x1, t1
LD x1, 9 * SIZE(X)
addt a2, t2, a2
unop
mult x2, x2, t2
LD x2, 10 * SIZE(X)
addt a3, t3, a3
unop
mult x3, x3, t3
LD x3, 11 * SIZE(X)
addt a0, t0, a0
unop
mult x4, x4, t0
LD x4, 12 * SIZE(X)
addt a1, t1, a1
unop
mult x5, x5, t1
LD x5, 13 * SIZE(X)
addt a2, t2, a2
unop
mult x6, x6, t2
LD x6, 14 * SIZE(X)
addt a3, t3, a3
unop
mult x7, x7, t3
LD x7, 15 * SIZE(X)
addt a0, t0, a0
unop
mult x0, x0, t0
LD x0, 16 * SIZE(X)
addt a1, t1, a1
lda X, 16 * SIZE(X)
mult x1, x1, t1
LD x1, 17 * SIZE(XX)
addt a2, t2, a2
unop
mult x2, x2, t2
LD x2, 18 * SIZE(XX)
addt a3, t3, a3
unop
mult x3, x3, t3
LD x3, 19 * SIZE(XX)
addt a0, t0, a0
unop
mult x4, x4, t0
LD x4, 20 * SIZE(XX)
addt a1, t1, a1
lda I, -1(I)
mult x5, x5, t1
LD x5, 21 * SIZE(XX)
addt a2, t2, a2
unop
mult x6, x6, t2
LD x6, 22 * SIZE(XX)
addt a3, t3, a3
mult x7, x7, t3
LD x7, 23 * SIZE(XX)
bgt I, $L11
.align 4
$L12:
addt a0, t0, a0
mov X, XX
mult x0, x0, t0
LD x0, 8 * SIZE(X)
addt a1, t1, a1
unop
mult x1, x1, t1
LD x1, 9 * SIZE(X)
addt a2, t2, a2
unop
mult x2, x2, t2
LD x2, 10 * SIZE(X)
addt a3, t3, a3
unop
mult x3, x3, t3
LD x3, 11 * SIZE(X)
addt a0, t0, a0
unop
mult x4, x4, t0
LD x4, 12 * SIZE(XX)
addt a1, t1, a1
unop
mult x5, x5, t1
LD x5, 13 * SIZE(XX)
addt a2, t2, a2
unop
mult x6, x6, t2
LD x6, 14 * SIZE(XX)
addt a3, t3, a3
lda X, 16 * SIZE(X)
mult x7, x7, t3
LD x7, 15 * SIZE(XX)
addt a0, t0, a0
mult x0, x0, t0
addt a1, t1, a1
mult x1, x1, t1
addt a2, t2, a2
mult x2, x2, t2
addt a3, t3, a3
mult x3, x3, t3
addt a0, t0, a0
mult x4, x4, t0
addt a1, t1, a1
mult x5, x5, t1
addt a2, t2, a2
mult x6, x6, t2
addt a3, t3, a3
mult x7, x7, t3
addt a1, t1, a1
addt a2, t2, a2
addt a3, t3, a3
.align 4
$L15:
and N, 15, I
ble I, $L998
.align 4
$L16:
LD x0, 0 * SIZE(X)
lda X, 1 * SIZE(X)
addt a0, t0, a0
mult x0, x0, t0
lda I, -1(I)
bgt I, $L16
bsr $31, $L998
.align 4
$L20:
fclr t0
sra N, 3, I
fclr t1
ble I, $L25
fclr t2
fclr t3
LD x0, 0 * SIZE(X)
addq X, INCX, X
LD x1, 0 * SIZE(X)
addq X, INCX, X
LD x2, 0 * SIZE(X)
addq X, INCX, X
LD x3, 0 * SIZE(X)
addq X, INCX, X
LD x4, 0 * SIZE(X)
addq X, INCX, X
LD x5, 0 * SIZE(X)
addq X, INCX, X
LD x6, 0 * SIZE(X)
addq X, INCX, X
lda I, -1(I)
ble I, $L22
.align 4
$L21:
addt a0, t0, a0
LD x7, 0 * SIZE(X)
mult x0, x0, t0
addq X, INCX, X
addt a1, t1, a1
LD x0, 0 * SIZE(X)
mult x1, x1, t1
addq X, INCX, X
addt a2, t2, a2
LD x1, 0 * SIZE(X)
mult x2, x2, t2
addq X, INCX, X
addt a3, t3, a3
LD x2, 0 * SIZE(X)
mult x3, x3, t3
addq X, INCX, X
addt a0, t0, a0
LD x3, 0 * SIZE(X)
mult x4, x4, t0
addq X, INCX, X
addt a1, t1, a1
LD x4, 0 * SIZE(X)
mult x5, x5, t1
addq X, INCX, X
addt a2, t2, a2
LD x5, 0 * SIZE(X)
mult x6, x6, t2
addq X, INCX, X
addt a3, t3, a3
LD x6, 0 * SIZE(X)
mult x7, x7, t3
addq X, INCX, X
lda I, -1(I)
bgt I, $L21
.align 4
$L22:
addt a0, t0, a0
LD x7, 0 * SIZE(X)
mult x0, x0, t0
addq X, INCX, X
addt a1, t1, a1
unop
mult x1, x1, t1
unop
addt a2, t2, a2
mult x2, x2, t2
addt a3, t3, a3
mult x3, x3, t3
addt a0, t0, a0
mult x4, x4, t0
addt a1, t1, a1
mult x5, x5, t1
addt a2, t2, a2
mult x6, x6, t2
addt a3, t3, a3
mult x7, x7, t3
addt a1, t1, a1
addt a2, t2, a2
addt a3, t3, a3
.align 4
$L25:
and N, 7, I
ble I, $L998
.align 4
$L26:
LD x0, 0 * SIZE(X)
addq X, INCX, X
addt a0, t0, a0
mult x0, x0, t0
lda I, -1(I)
bgt I, $L26
.align 4
$L998:
addt a0, t0, a0
addt a0, a1, a0
addt a2, a3, a2
#if defined(EV4) || defined(EV5)
addt a0, a2, $f16
jsr $26, ($27), sqrt !lituse_jsr!2
ldah $29, 0($26) !gpdisp!3
lda $29, 0($29) !gpdisp!3
#else
addt a0, a2, a0
sqrtt a0, a0
#endif
.align 4
$L999:
#if defined(EV4) || defined(EV5)
ldq $26, 0($sp)
lda $sp, 16($sp)
#endif
ret
EPILOGUE