Debian Codegraph @

gmp4/mpn/x86_64/atom/mul_2.asm

raw
1 dnl AMD64 mpn_mul_2 optimised for Intel Atom.
2
3 dnl Copyright 2008, 2011-2013 Free Software Foundation, Inc.
4
5 dnl This file is part of the GNU MP Library.
6 dnl
7 dnl The GNU MP Library is free software; you can redistribute it and/or modify
8 dnl it under the terms of either:
9 dnl
10 dnl * the GNU Lesser General Public License as published by the Free
11 dnl Software Foundation; either version 3 of the License, or (at your
12 dnl option) any later version.
13 dnl
14 dnl or
15 dnl
16 dnl * the GNU General Public License as published by the Free Software
17 dnl Foundation; either version 2 of the License, or (at your option) any
18 dnl later version.
19 dnl
20 dnl or both in parallel, as here.
21 dnl
22 dnl The GNU MP Library is distributed in the hope that it will be useful, but
23 dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
24 dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
25 dnl for more details.
26 dnl
27 dnl You should have received copies of the GNU General Public License and the
28 dnl GNU Lesser General Public License along with the GNU MP Library. If not,
29 dnl see https://www.gnu.org/licenses/.
30
31 include(`../config.m4')
32
33 C cycles/limb best
34 C AMD K8,K9
35 C AMD K10
36 C AMD bd1
37 C AMD bd2
38 C AMD bobcat
39 C AMD jaguar
40 C Intel P4
41 C Intel PNR
42 C Intel NHM
43 C Intel SBR
44 C Intel IBR
45 C Intel HWL
46 C Intel BWL
47 C Intel atom 17.75 this
48 C VIA nano
49
50 C The loop of this code is the result of running a code generation and
51 C optimisation tool suite written by David Harvey and Torbjorn Granlund.
52
53 define(`rp', `%rdi') C rcx
54 define(`up', `%rsi') C rdx
55 define(`n_param', `%rdx') C r8
56 define(`vp', `%rcx') C r9
57
58 define(`v0', `%r8')
59 define(`v1', `%r9')
60 define(`w0', `%rbx')
61 define(`w1', `%rcx')
62 define(`w2', `%rbp')
63 define(`w3', `%r10')
64 define(`n', `%r11')
65
66 ABI_SUPPORT(DOS64)
67 ABI_SUPPORT(STD64)
68
69 ASM_START()
70 TEXT
71 ALIGN(16)
72 PROLOGUE(mpn_mul_2)
73 FUNC_ENTRY(4)
74 push %rbx
75 push %rbp
76
77 mov (up), %rax
78
79 mov (vp), v0
80 mov 8(vp), v1
81
82 mov n_param, n
83 mul v0
84
85 test $1, R8(n)
86 jnz L(bx1)
87
88 L(bx0): test $2, R8(n)
89 jnz L(b10)
90
91 L(b00): mov %rax, w0
92 mov (up), %rax
93 mov %rdx, w1
94 xor R32(w2), R32(w2)
95 lea -8(rp), rp
96 jmp L(lo0)
97
98 L(b10): mov %rax, w2
99 mov (up), %rax
100 mov %rdx, w3
101 xor R32(w0), R32(w0)
102 lea -16(up), up
103 lea -24(rp), rp
104 jmp L(lo2)
105
106 L(bx1): test $2, R8(n)
107 jnz L(b11)
108
109 L(b01): mov %rax, w3
110 mov %rdx, w0
111 mov (up), %rax
112 xor R32(w1), R32(w1)
113 lea 8(up), up
114 dec n
115 jmp L(lo1)
116
117 L(b11): mov %rax, w1
118 mov (up), %rax
119 mov %rdx, w2
120 xor R32(w3), R32(w3)
121 lea -8(up), up
122 lea -16(rp), rp
123 jmp L(lo3)
124
125 ALIGN(16)
126 L(top):
127 L(lo1): mul v1
128 add %rax, w0
129 mov (up), %rax
130 mov $0, R32(w2)
131 mov w3, (rp)
132 adc %rdx, w1
133 mul v0
134 add %rax, w0
135 mov (up), %rax
136 adc %rdx, w1
137 adc $0, R32(w2)
138 L(lo0): mul v1
139 add %rax, w1
140 mov 8(up), %rax
141 mov w0, 8(rp)
142 adc %rdx, w2
143 mul v0
144 add %rax, w1
145 mov 8(up), %rax
146 adc %rdx, w2
147 mov $0, R32(w3)
148 adc $0, R32(w3)
149 L(lo3): mul v1
150 add %rax, w2
151 mov 16(up), %rax
152 mov w1, 16(rp)
153 mov $0, R32(w0)
154 adc %rdx, w3
155 mul v0
156 add %rax, w2
157 mov 16(up), %rax
158 adc %rdx, w3
159 L(lo2): mov $0, R32(w1)
160 mov w2, 24(rp)
161 adc $0, R32(w0)
162 mul v1
163 add %rax, w3
164 mov 24(up), %rax
165 lea 32(up), up
166 adc %rdx, w0
167 mul v0
168 lea 32(rp), rp
169 add %rax, w3
170 adc %rdx, w0
171 mov -8(up), %rax
172 adc $0, R32(w1)
173 sub $4, n
174 ja L(top)
175
176 L(end): mul v1
177 mov w3, (rp)
178 add %rax, w0
179 adc %rdx, w1
180 mov w0, 8(rp)
181 mov w1, %rax
182 pop %rbp
183 pop %rbx
184 FUNC_EXIT()
185 ret
186 EPILOGUE()