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
|
/*
* Atomic xchg and cmpxchg operations.
*
* This file is subject to the terms and conditions of the GNU General Public
* License. See the file "COPYING" in the main directory of this archive
* for more details.
*
* Copyright (C) 2001 - 2005 Tensilica Inc.
*/
#ifndef _XTENSA_CMPXCHG_H
#define _XTENSA_CMPXCHG_H
#ifndef __ASSEMBLY__
#include <linux/bits.h>
#include <linux/stringify.h>
/*
* cmpxchg
*/
static inline unsigned long
__cmpxchg_u32(volatile int *p, int old, int new)
{
#if XCHAL_HAVE_EXCLUSIVE
unsigned long tmp, result;
__asm__ __volatile__(
"1: l32ex %[result], %[addr]\n"
" bne %[result], %[cmp], 2f\n"
" mov %[tmp], %[new]\n"
" s32ex %[tmp], %[addr]\n"
" getex %[tmp]\n"
" beqz %[tmp], 1b\n"
"2:\n"
: [result] "=&a" (result), [tmp] "=&a" (tmp)
: [new] "a" (new), [addr] "a" (p), [cmp] "a" (old)
: "memory"
);
return result;
#elif XCHAL_HAVE_S32C1I
__asm__ __volatile__(
" wsr %[cmp], scompare1\n"
" s32c1i %[new], %[mem]\n"
: [new] "+a" (new), [mem] "+m" (*p)
: [cmp] "a" (old)
: "memory"
);
return new;
#else
__asm__ __volatile__(
" rsil a15, "__stringify(TOPLEVEL)"\n"
" l32i %[old], %[mem]\n"
" bne %[old], %[cmp], 1f\n"
" s32i %[new], %[mem]\n"
"1:\n"
" wsr a15, ps\n"
" rsync\n"
: [old] "=&a" (old), [mem] "+m" (*p)
: [cmp] "a" (old), [new] "r" (new)
: "a15", "memory");
return old;
#endif
}
/* This function doesn't exist, so you'll get a linker error
* if something tries to do an invalid cmpxchg(). */
extern void __cmpxchg_called_with_bad_pointer(void);
static __inline__ unsigned long
__cmpxchg(volatile void *ptr, unsigned long old, unsigned long new, int size)
{
switch (size) {
case 4: return __cmpxchg_u32(ptr, old, new);
default: __cmpxchg_called_with_bad_pointer();
return old;
}
}
#define cmpxchg(ptr,o,n) \
({ __typeof__(*(ptr)) _o_ = (o); \
__typeof__(*(ptr)) _n_ = (n); \
(__typeof__(*(ptr))) __cmpxchg((ptr), (unsigned long)_o_, \
(unsigned long)_n_, sizeof (*(ptr))); \
})
#include <asm-generic/cmpxchg-local.h>
static inline unsigned long __cmpxchg_local(volatile void *ptr,
unsigned long old,
unsigned long new, int size)
{
switch (size) {
case 4:
return __cmpxchg_u32(ptr, old, new);
default:
return __generic_cmpxchg_local(ptr, old, new, size);
}
return old;
}
/*
* cmpxchg_local and cmpxchg64_local are atomic wrt current CPU. Always make
* them available.
*/
#define cmpxchg_local(ptr, o, n) \
((__typeof__(*(ptr)))__generic_cmpxchg_local((ptr), (unsigned long)(o),\
(unsigned long)(n), sizeof(*(ptr))))
#define cmpxchg64_local(ptr, o, n) __generic_cmpxchg64_local((ptr), (o), (n))
#define cmpxchg64(ptr, o, n) cmpxchg64_local((ptr), (o), (n))
/*
* xchg_u32
*
* Note that a15 is used here because the register allocation
* done by the compiler is not guaranteed and a window overflow
* may not occur between the rsil and wsr instructions. By using
* a15 in the rsil, the machine is guaranteed to be in a state
* where no register reference will cause an overflow.
*/
static inline unsigned long xchg_u32(volatile int * m, unsigned long val)
{
#if XCHAL_HAVE_EXCLUSIVE
unsigned long tmp, result;
__asm__ __volatile__(
"1: l32ex %[result], %[addr]\n"
" mov %[tmp], %[val]\n"
" s32ex %[tmp], %[addr]\n"
" getex %[tmp]\n"
" beqz %[tmp], 1b\n"
: [result] "=&a" (result), [tmp] "=&a" (tmp)
: [val] "a" (val), [addr] "a" (m)
: "memory"
);
return result;
#elif XCHAL_HAVE_S32C1I
unsigned long tmp, result;
__asm__ __volatile__(
"1: l32i %[tmp], %[mem]\n"
" mov %[result], %[val]\n"
" wsr %[tmp], scompare1\n"
" s32c1i %[result], %[mem]\n"
" bne %[result], %[tmp], 1b\n"
: [result] "=&a" (result), [tmp] "=&a" (tmp),
[mem] "+m" (*m)
: [val] "a" (val)
: "memory"
);
return result;
#else
unsigned long tmp;
__asm__ __volatile__(
" rsil a15, "__stringify(TOPLEVEL)"\n"
" l32i %[tmp], %[mem]\n"
" s32i %[val], %[mem]\n"
" wsr a15, ps\n"
" rsync\n"
: [tmp] "=&a" (tmp), [mem] "+m" (*m)
: [val] "a" (val)
: "a15", "memory");
return tmp;
#endif
}
#define xchg(ptr,x) \
((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
static inline u32 xchg_small(volatile void *ptr, u32 x, int size)
{
int off = (unsigned long)ptr % sizeof(u32);
volatile u32 *p = ptr - off;
#ifdef __BIG_ENDIAN
int bitoff = (sizeof(u32) - size - off) * BITS_PER_BYTE;
#else
int bitoff = off * BITS_PER_BYTE;
#endif
u32 bitmask = ((0x1 << size * BITS_PER_BYTE) - 1) << bitoff;
u32 oldv, newv;
u32 ret;
do {
oldv = READ_ONCE(*p);
ret = (oldv & bitmask) >> bitoff;
newv = (oldv & ~bitmask) | (x << bitoff);
} while (__cmpxchg_u32(p, oldv, newv) != oldv);
return ret;
}
/*
* This only works if the compiler isn't horribly bad at optimizing.
* gcc-2.5.8 reportedly can't handle this, but I define that one to
* be dead anyway.
*/
extern void __xchg_called_with_bad_pointer(void);
static __inline__ unsigned long
__xchg(unsigned long x, volatile void * ptr, int size)
{
switch (size) {
case 1:
return xchg_small(ptr, x, 1);
case 2:
return xchg_small(ptr, x, 2);
case 4:
return xchg_u32(ptr, x);
default:
__xchg_called_with_bad_pointer();
return x;
}
}
#endif /* __ASSEMBLY__ */
#endif /* _XTENSA_CMPXCHG_H */
|
