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
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
|
// Copyright 2015 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.
// +build ppc64 ppc64le
#include "textflag.h"
// bool cas(uint32 *ptr, uint32 old, uint32 new)
// Atomically:
// if(*val == old){
// *val = new;
// return 1;
// } else
// return 0;
TEXT runtime∕internal∕atomic·Cas(SB), NOSPLIT, $0-17
MOVD ptr+0(FP), R3
MOVWZ old+8(FP), R4
MOVWZ new+12(FP), R5
SYNC
cas_again:
LWAR (R3), R6
CMPW R6, R4
BNE cas_fail
STWCCC R5, (R3)
BNE cas_again
MOVD $1, R3
ISYNC
MOVB R3, ret+16(FP)
RET
cas_fail:
MOVB R0, ret+16(FP)
RET
// bool runtime∕internal∕atomic·Cas64(uint64 *ptr, uint64 old, uint64 new)
// Atomically:
// if(*val == *old){
// *val = new;
// return 1;
// } else {
// return 0;
// }
TEXT runtime∕internal∕atomic·Cas64(SB), NOSPLIT, $0-25
MOVD ptr+0(FP), R3
MOVD old+8(FP), R4
MOVD new+16(FP), R5
SYNC
cas64_again:
LDAR (R3), R6
CMP R6, R4
BNE cas64_fail
STDCCC R5, (R3)
BNE cas64_again
MOVD $1, R3
ISYNC
MOVB R3, ret+24(FP)
RET
cas64_fail:
MOVB R0, ret+24(FP)
RET
TEXT runtime∕internal∕atomic·Casuintptr(SB), NOSPLIT, $0-25
BR runtime∕internal∕atomic·Cas64(SB)
TEXT runtime∕internal∕atomic·Loaduintptr(SB), NOSPLIT|NOFRAME, $0-16
BR runtime∕internal∕atomic·Load64(SB)
TEXT runtime∕internal∕atomic·Loaduint(SB), NOSPLIT|NOFRAME, $0-16
BR runtime∕internal∕atomic·Load64(SB)
TEXT runtime∕internal∕atomic·Storeuintptr(SB), NOSPLIT, $0-16
BR runtime∕internal∕atomic·Store64(SB)
TEXT runtime∕internal∕atomic·Xadduintptr(SB), NOSPLIT, $0-24
BR runtime∕internal∕atomic·Xadd64(SB)
TEXT runtime∕internal∕atomic·Loadint64(SB), NOSPLIT, $0-16
BR runtime∕internal∕atomic·Load64(SB)
TEXT runtime∕internal∕atomic·Xaddint64(SB), NOSPLIT, $0-24
BR runtime∕internal∕atomic·Xadd64(SB)
// bool casp(void **val, void *old, void *new)
// Atomically:
// if(*val == old){
// *val = new;
// return 1;
// } else
// return 0;
TEXT runtime∕internal∕atomic·Casp1(SB), NOSPLIT, $0-25
BR runtime∕internal∕atomic·Cas64(SB)
// uint32 xadd(uint32 volatile *ptr, int32 delta)
// Atomically:
// *val += delta;
// return *val;
TEXT runtime∕internal∕atomic·Xadd(SB), NOSPLIT, $0-20
MOVD ptr+0(FP), R4
MOVW delta+8(FP), R5
SYNC
LWAR (R4), R3
ADD R5, R3
STWCCC R3, (R4)
BNE -3(PC)
ISYNC
MOVW R3, ret+16(FP)
RET
TEXT runtime∕internal∕atomic·Xadd64(SB), NOSPLIT, $0-24
MOVD ptr+0(FP), R4
MOVD delta+8(FP), R5
SYNC
LDAR (R4), R3
ADD R5, R3
STDCCC R3, (R4)
BNE -3(PC)
ISYNC
MOVD R3, ret+16(FP)
RET
TEXT runtime∕internal∕atomic·Xchg(SB), NOSPLIT, $0-20
MOVD ptr+0(FP), R4
MOVW new+8(FP), R5
SYNC
LWAR (R4), R3
STWCCC R5, (R4)
BNE -2(PC)
ISYNC
MOVW R3, ret+16(FP)
RET
TEXT runtime∕internal∕atomic·Xchg64(SB), NOSPLIT, $0-24
MOVD ptr+0(FP), R4
MOVD new+8(FP), R5
SYNC
LDAR (R4), R3
STDCCC R5, (R4)
BNE -2(PC)
ISYNC
MOVD R3, ret+16(FP)
RET
TEXT runtime∕internal∕atomic·Xchguintptr(SB), NOSPLIT, $0-24
BR runtime∕internal∕atomic·Xchg64(SB)
TEXT runtime∕internal∕atomic·StorepNoWB(SB), NOSPLIT, $0-16
BR runtime∕internal∕atomic·Store64(SB)
TEXT runtime∕internal∕atomic·Store(SB), NOSPLIT, $0-12
MOVD ptr+0(FP), R3
MOVW val+8(FP), R4
SYNC
MOVW R4, 0(R3)
RET
TEXT runtime∕internal∕atomic·Store64(SB), NOSPLIT, $0-16
MOVD ptr+0(FP), R3
MOVD val+8(FP), R4
SYNC
MOVD R4, 0(R3)
RET
// void runtime∕internal∕atomic·Or8(byte volatile*, byte);
TEXT runtime∕internal∕atomic·Or8(SB), NOSPLIT, $0-9
MOVD ptr+0(FP), R3
MOVBZ val+8(FP), R4
#ifdef GOARCH_ppc64
// Align ptr down to 4 bytes so we can use 32-bit load/store.
// R5 = (R3 << 0) & ~3
RLDCR $0, R3, $~3, R5
// Compute val shift.
// Big endian. ptr = ptr ^ 3
XOR $3, R3
// R6 = ((ptr & 3) * 8) = (ptr << 3) & (3*8)
RLDC $3, R3, $(3*8), R6
// Shift val for aligned ptr. R4 = val << R6
SLD R6, R4, R4
SYNC
again:
LWAR (R5), R6
OR R4, R6
STWCCC R6, (R5)
BNE again
#else
SYNC
again:
LBAR (R3), R6
OR R4, R6
STBCCC R6, (R3)
BNE again
#endif
ISYNC
RET
// void runtime∕internal∕atomic·And8(byte volatile*, byte);
TEXT runtime∕internal∕atomic·And8(SB), NOSPLIT, $0-9
MOVD ptr+0(FP), R3
MOVBZ val+8(FP), R4
#ifdef GOARCH_ppc64
// Align ptr down to 4 bytes so we can use 32-bit load/store.
// R5 = (R3 << 0) & ~3
RLDCR $0, R3, $~3, R5
// Compute val shift.
// Big endian. ptr = ptr ^ 3
XOR $3, R3
// R6 = ((ptr & 3) * 8) = (ptr << 3) & (3*8)
RLDC $3, R3, $(3*8), R6
// Shift val for aligned ptr. R4 = val << R6 | ^(0xFF << R6)
MOVD $0xFF, R7
SLD R6, R4
SLD R6, R7
XOR $-1, R7
OR R7, R4
SYNC
again:
LWAR (R5), R6
AND R4, R6
STWCCC R6, (R5)
BNE again
#else
SYNC
again:
LBAR (R3),R6
AND R4,R6
STBCCC R6,(R3)
BNE again
#endif
ISYNC
RET
|
