1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
|
###################################-
#
# Copyright (C) 2009-2014 Free Software Foundation, Inc.
#
# Contributed by Michael Eager <eager@eagercon.com>.
#
# This file is free software; you can redistribute it and/or modify it
# under the terms of the GNU General Public License as published by the
# Free Software Foundation; either version 3, or (at your option) any
# later version.
#
# GCC is distributed in the hope that it will be useful, but WITHOUT
# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
# or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
# License for more details.
#
# Under Section 7 of GPL version 3, you are granted additional
# permissions described in the GCC Runtime Library Exception, version
# 3.1, as published by the Free Software Foundation.
#
# You should have received a copy of the GNU General Public License and
# a copy of the GCC Runtime Library Exception along with this program;
# see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
# <http://www.gnu.org/licenses/>.
#
# muldi3_hard.S
#
# Multiply operation for 64 bit integers, for devices with hard multiply
# Input : Operand1[H] in Reg r5
# Operand1[L] in Reg r6
# Operand2[H] in Reg r7
# Operand2[L] in Reg r8
# Output: Result[H] in Reg r3
# Result[L] in Reg r4
#
# Explaination:
#
# Both the input numbers are divided into 16 bit number as follows
# op1 = A B C D
# op2 = E F G H
# result = D * H
# + (C * H + D * G) << 16
# + (B * H + C * G + D * F) << 32
# + (A * H + B * G + C * F + D * E) << 48
#
# Only 64 bits of the output are considered
#
#######################################
.globl muldi3_hardproc
.ent muldi3_hardproc
muldi3_hardproc:
addi r1,r1,-40
# Save the input operands on the caller's stack
swi r5,r1,44
swi r6,r1,48
swi r7,r1,52
swi r8,r1,56
# Store all the callee saved registers
sw r20,r1,r0
swi r21,r1,4
swi r22,r1,8
swi r23,r1,12
swi r24,r1,16
swi r25,r1,20
swi r26,r1,24
swi r27,r1,28
# Load all the 16 bit values for A through H
lhui r20,r1,44 # A
lhui r21,r1,46 # B
lhui r22,r1,48 # C
lhui r23,r1,50 # D
lhui r24,r1,52 # E
lhui r25,r1,54 # F
lhui r26,r1,56 # G
lhui r27,r1,58 # H
# D * H ==> LSB of the result on stack ==> Store1
mul r9,r23,r27
swi r9,r1,36 # Pos2 and Pos3
# Hi (Store1) + C * H + D * G ==> Store2 ==> Pos1 and Pos2
# Store the carry generated in position 2 for Pos 3
lhui r11,r1,36 # Pos2
mul r9,r22,r27 # C * H
mul r10,r23,r26 # D * G
add r9,r9,r10
addc r12,r0,r0
add r9,r9,r11
addc r12,r12,r0 # Store the Carry
shi r9,r1,36 # Store Pos2
swi r9,r1,32
lhui r11,r1,32
shi r11,r1,34 # Store Pos1
# Hi (Store2) + B * H + C * G + D * F ==> Store3 ==> Pos0 and Pos1
mul r9,r21,r27 # B * H
mul r10,r22,r26 # C * G
mul r7,r23,r25 # D * F
add r9,r9,r11
add r9,r9,r10
add r9,r9,r7
swi r9,r1,32 # Pos0 and Pos1
# Hi (Store3) + A * H + B * G + C * F + D * E ==> Store3 ==> Pos0
lhui r11,r1,32 # Pos0
mul r9,r20,r27 # A * H
mul r10,r21,r26 # B * G
mul r7,r22,r25 # C * F
mul r8,r23,r24 # D * E
add r9,r9,r11
add r9,r9,r10
add r9,r9,r7
add r9,r9,r8
sext16 r9,r9 # Sign extend the MSB
shi r9,r1,32
# Move results to r3 and r4
lhui r3,r1,32
add r3,r3,r12
shi r3,r1,32
lwi r3,r1,32 # Hi Part
lwi r4,r1,36 # Lo Part
# Restore Callee saved registers
lw r20,r1,r0
lwi r21,r1,4
lwi r22,r1,8
lwi r23,r1,12
lwi r24,r1,16
lwi r25,r1,20
lwi r26,r1,24
lwi r27,r1,28
# Restore Frame and return
rtsd r15,8
addi r1,r1,40
.end muldi3_hardproc
|
