github.com/aergoio/aergo@v1.3.1/libtool/src/gmp-6.1.2/mpn/pa32/hppa1_1/pa7100/lshift.asm (about)

     1  dnl  HP-PA  mpn_lshift -- Shift a number left.
     2  dnl  Optimized for the PA7100, where is runs at 3.25 cycles/limb.
     3  
     4  dnl  Copyright 1992, 1994, 2000-2003 Free Software Foundation, Inc.
     5  
     6  dnl  This file is part of the GNU MP Library.
     7  dnl
     8  dnl  The GNU MP Library is free software; you can redistribute it and/or modify
     9  dnl  it under the terms of either:
    10  dnl
    11  dnl    * the GNU Lesser General Public License as published by the Free
    12  dnl      Software Foundation; either version 3 of the License, or (at your
    13  dnl      option) any later version.
    14  dnl
    15  dnl  or
    16  dnl
    17  dnl    * the GNU General Public License as published by the Free Software
    18  dnl      Foundation; either version 2 of the License, or (at your option) any
    19  dnl      later version.
    20  dnl
    21  dnl  or both in parallel, as here.
    22  dnl
    23  dnl  The GNU MP Library is distributed in the hope that it will be useful, but
    24  dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
    25  dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
    26  dnl  for more details.
    27  dnl
    28  dnl  You should have received copies of the GNU General Public License and the
    29  dnl  GNU Lesser General Public License along with the GNU MP Library.  If not,
    30  dnl  see https://www.gnu.org/licenses/.
    31  
    32  include(`../config.m4')
    33  
    34  C INPUT PARAMETERS
    35  C res_ptr	r26
    36  C s_ptr		r25
    37  C size		r24
    38  C cnt		r23
    39  
    40  ASM_START()
    41  PROLOGUE(mpn_lshift)
    42  	sh2add		%r24,%r25,%r25
    43  	sh2add		%r24,%r26,%r26
    44  	ldws,mb		-4(0,%r25),%r22
    45  	subi		32,%r23,%r1
    46  	mtsar		%r1
    47  	addib,=		-1,%r24,L(0004)
    48  	vshd		%r0,%r22,%r28		C compute carry out limb
    49  	ldws,mb		-4(0,%r25),%r29
    50  	addib,<=	-5,%r24,L(rest)
    51  	vshd		%r22,%r29,%r20
    52  
    53  LDEF(loop)
    54  	ldws,mb		-4(0,%r25),%r22
    55  	stws,mb		%r20,-4(0,%r26)
    56  	vshd		%r29,%r22,%r20
    57  	ldws,mb		-4(0,%r25),%r29
    58  	stws,mb		%r20,-4(0,%r26)
    59  	vshd		%r22,%r29,%r20
    60  	ldws,mb		-4(0,%r25),%r22
    61  	stws,mb		%r20,-4(0,%r26)
    62  	vshd		%r29,%r22,%r20
    63  	ldws,mb		-4(0,%r25),%r29
    64  	stws,mb		%r20,-4(0,%r26)
    65  	addib,>		-4,%r24,L(loop)
    66  	vshd		%r22,%r29,%r20
    67  
    68  LDEF(rest)
    69  	addib,=		4,%r24,L(end1)
    70  	nop
    71  
    72  LDEF(eloop)
    73  	ldws,mb		-4(0,%r25),%r22
    74  	stws,mb		%r20,-4(0,%r26)
    75  	addib,<=	-1,%r24,L(end2)
    76  	vshd		%r29,%r22,%r20
    77  	ldws,mb		-4(0,%r25),%r29
    78  	stws,mb		%r20,-4(0,%r26)
    79  	addib,>		-1,%r24,L(eloop)
    80  	vshd		%r22,%r29,%r20
    81  
    82  LDEF(end1)
    83  	stws,mb		%r20,-4(0,%r26)
    84  	vshd		%r29,%r0,%r20
    85  	bv		0(%r2)
    86  	stw		%r20,-4(0,%r26)
    87  
    88  LDEF(end2)
    89  	stws,mb		%r20,-4(0,%r26)
    90  
    91  LDEF(0004)
    92  	vshd		%r22,%r0,%r20
    93  	bv		0(%r2)
    94  	stw		%r20,-4(0,%r26)
    95  EPILOGUE()