aors_n.asm revision 1.1.1.1 1 dnl PowerPC-32 mpn_add_n and mpn_sub_n.
2
3 dnl Copyright 2002, 2005, 2007 Free Software Foundation, Inc.
4
5 dnl This file is part of the GNU MP Library.
6
7 dnl The GNU MP Library is free software; you can redistribute it and/or modify
8 dnl it under the terms of the GNU Lesser General Public License as published
9 dnl by the Free Software Foundation; either version 3 of the License, or (at
10 dnl your option) any later version.
11
12 dnl The GNU MP Library is distributed in the hope that it will be useful, but
13 dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
14 dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public
15 dnl License for more details.
16
17 dnl You should have received a copy of the GNU Lesser General Public License
18 dnl along with the GNU MP Library. If not, see http://www.gnu.org/licenses/.
19
20 include(`../config.m4')
21
22 C cycles/limb
23 C 603e: ?
24 C 604e: ? old: 3.25
25 C 75x (G3): ? old: 3.5
26 C 7400,7410 (G4): 3.25
27 C 744x,745x (G4+): 4
28 C power4/ppc970: ? old: 2.0
29 C power5: ? old: 2.5
30
31 C INPUT PARAMETERS
32 define(`rp', `r3')
33 define(`up', `r4')
34 define(`vp', `r5')
35 define(`n', `r6')
36 define(`cy', `r7')
37
38 ifdef(`OPERATION_add_n', `
39 define(ADCSBC, adde)
40 define(func, mpn_add_n)
41 define(func_nc, mpn_add_nc)
42 define(IFADD, `$1')
43 define(IFSUB, `')')
44 ifdef(`OPERATION_sub_n', `
45 define(ADCSBC, subfe)
46 define(func, mpn_sub_n)
47 define(func_nc, mpn_sub_nc)
48 define(IFADD, `')
49 define(IFSUB, `$1')')
50
51 MULFUNC_PROLOGUE(mpn_add_n mpn_add_nc mpn_sub_n mpn_sub_nc)
52
53 ASM_START()
54
55 PROLOGUE(func_nc)
56 IFADD(` addic r0, cy, -1') C set carry from argument
57 IFSUB(` subfic r0, cy, 0') C set carry from argument
58 b L(ent)
59 EPILOGUE()
60
61 PROLOGUE(func)
62 IFADD(` addic r0, n, 0') C clear carry
63 IFSUB(` addic r0, n, -1') C set carry
64 L(ent): andi. r0, n, 3
65 addi r3, r3, -12
66 addi n, n, 1
67 cmpwi cr7, r0, 2
68 srwi r0, n, 2
69 sub r4, r4, r3
70 sub r5, r5, r3
71 mtctr r0
72 bne cr0, L(n00)
73
74 lwzx r7, r4, r3 C n = 4, 8, 12, ...
75 lwzx r8, r5, r3
76 addi r3, r3, 4
77 lwzx r9, r4, r3
78 ADCSBC r7, r8, r7
79 lwzx r10, r5, r3
80 addi r3, r3, 4
81 b L(00)
82
83 L(n00): bge cr7, L(n01)
84 cmpwi cr0, r0, 0 C n = 1, 5, 9, 13, ...
85 lwzx r0, r4, r3
86 lwzx r6, r5, r3
87 addi r3, r3, 4
88 ADCSBC r0, r6, r0
89 ble L(ret)
90 L(gt1): lwzx r7, r4, r3
91 lwzx r8, r5, r3
92 addi r3, r3, 4
93 b L(01)
94
95 L(n10):
96 lwzx r9, r4, r3 C n = 3, 7, 11, 15, ...
97 lwzx r10, r5, r3
98 addi r3, r3, 4
99 lwzx r11, r4, r3
100 ADCSBC r9, r10, r9
101 lwzx r12, r5, r3
102 addi r3, r3, 4
103 b L(11)
104
105 L(n01): bne cr7, L(n10)
106 cmpwi cr0, r0, 0 C n = 2, 6, 10, 14, ...
107 lwzx r11, r4, r3
108 lwzx r12, r5, r3
109 addi r3, r3, 4
110 lwzx r0, r4, r3
111 ADCSBC r11, r12, r11
112 lwzx r6, r5, r3
113 addi r3, r3, 4
114 ble cr0, L(end)
115
116
117 L(lp): lwzx r7, r4, r3
118 ADCSBC r0, r6, r0
119 lwzx r8, r5, r3
120 stwu r11, 4(r3)
121 L(01): lwzx r9, r4, r3
122 ADCSBC r7, r8, r7
123 lwzx r10, r5, r3
124 stwu r0, 4(r3)
125 L(00): lwzx r11, r4, r3
126 ADCSBC r9, r10, r9
127 lwzx r12, r5, r3
128 stwu r7, 4(r3)
129 L(11): lwzx r0, r4, r3
130 ADCSBC r11, r12, r11
131 lwzx r6, r5, r3
132 stwu r9, 4(r3)
133 bdnz L(lp)
134
135 L(end): ADCSBC r0, r6, r0
136 stw r11, 4(r3)
137 L(ret): stw r0, 8(r3)
138 IFADD(` li r3, 0 ')
139 IFADD(` addze r3, r3 ')
140 IFSUB(` subfe r3, r0, r0')
141 IFSUB(` neg r3, r3')
142 blr
143 EPILOGUE()
144