summaryrefslogtreecommitdiffstats
path: root/arch/sparc/include/asm/spinlock_64.h
blob: 073936a8b2755e1c15e637ff28c5bc607876f4dc (plain)
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
/* spinlock.h: 64-bit Sparc spinlock support.
 *
 * Copyright (C) 1997 David S. Miller (davem@caip.rutgers.edu)
 */

#ifndef __SPARC64_SPINLOCK_H
#define __SPARC64_SPINLOCK_H

#ifndef __ASSEMBLY__

/* To get debugging spinlocks which detect and catch
 * deadlock situations, set CONFIG_DEBUG_SPINLOCK
 * and rebuild your kernel.
 */

/* Because we play games to save cycles in the non-contention case, we
 * need to be extra careful about branch targets into the "spinning"
 * code.  They live in their own section, but the newer V9 branches
 * have a shorter range than the traditional 32-bit sparc branch
 * variants.  The rule is that the branches that go into and out of
 * the spinner sections must be pre-V9 branches.
 */

#define arch_spin_is_locked(lp)	((lp)->lock != 0)

#define arch_spin_unlock_wait(lp)	\
	do {	rmb();			\
	} while((lp)->lock)

static inline void arch_spin_lock(arch_spinlock_t *lock)
{
	unsigned long tmp;

	__asm__ __volatile__(
"1:	ldstub		[%1], %0\n"
"	brnz,pn		%0, 2f\n"
"	 nop\n"
"	.subsection	2\n"
"2:	ldub		[%1], %0\n"
"	brnz,pt		%0, 2b\n"
"	 nop\n"
"	ba,a,pt		%%xcc, 1b\n"
"	.previous"
	: "=&r" (tmp)
	: "r" (lock)
	: "memory");
}

static inline int arch_spin_trylock(arch_spinlock_t *lock)
{
	unsigned long result;

	__asm__ __volatile__(
"	ldstub		[%1], %0\n"
	: "=r" (result)
	: "r" (lock)
	: "memory");

	return (result == 0UL);
}

static inline void arch_spin_unlock(arch_spinlock_t *lock)
{
	__asm__ __volatile__(
"	stb		%%g0, [%0]"
	: /* No outputs */
	: "r" (lock)
	: "memory");
}

static inline void arch_spin_lock_flags(arch_spinlock_t *lock, unsigned long flags)
{
	unsigned long tmp1, tmp2;

	__asm__ __volatile__(
"1:	ldstub		[%2], %0\n"
"	brnz,pn		%0, 2f\n"
"	 nop\n"
"	.subsection	2\n"
"2:	rdpr		%%pil, %1\n"
"	wrpr		%3, %%pil\n"
"3:	ldub		[%2], %0\n"
"	brnz,pt		%0, 3b\n"
"	 nop\n"
"	ba,pt		%%xcc, 1b\n"
"	 wrpr		%1, %%pil\n"
"	.previous"
	: "=&r" (tmp1), "=&r" (tmp2)
	: "r"(lock), "r"(flags)
	: "memory");
}

/* Multi-reader locks, these are much saner than the 32-bit Sparc ones... */

static void inline arch_read_lock(arch_rwlock_t *lock)
{
	unsigned long tmp1, tmp2;

	__asm__ __volatile__ (
"1:	ldsw		[%2], %0\n"
"	brlz,pn		%0, 2f\n"
"4:	 add		%0, 1, %1\n"
"	cas		[%2], %0, %1\n"
"	cmp		%0, %1\n"
"	bne,pn		%%icc, 1b\n"
"	 nop\n"
"	.subsection	2\n"
"2:	ldsw		[%2], %0\n"
"	brlz,pt		%0, 2b\n"
"	 nop\n"
"	ba,a,pt		%%xcc, 4b\n"
"	.previous"
	: "=&r" (tmp1), "=&r" (tmp2)
	: "r" (lock)
	: "memory");
}

static int inline arch_read_trylock(arch_rwlock_t *lock)
{
	int tmp1, tmp2;

	__asm__ __volatile__ (
"1:	ldsw		[%2], %0\n"
"	brlz,a,pn	%0, 2f\n"
"	 mov		0, %0\n"
"	add		%0, 1, %1\n"
"	cas		[%2], %0, %1\n"
"	cmp		%0, %1\n"
"	bne,pn		%%icc, 1b\n"
"	 mov		1, %0\n"
"2:"
	: "=&r" (tmp1), "=&r" (tmp2)
	: "r" (lock)
	: "memory");

	return tmp1;
}

static void inline arch_read_unlock(arch_rwlock_t *lock)
{
	unsigned long tmp1, tmp2;

	__asm__ __volatile__(
"1:	lduw	[%2], %0\n"
"	sub	%0, 1, %1\n"
"	cas	[%2], %0, %1\n"
"	cmp	%0, %1\n"
"	bne,pn	%%xcc, 1b\n"
"	 nop"
	: "=&r" (tmp1), "=&r" (tmp2)
	: "r" (lock)
	: "memory");
}

static void inline arch_write_lock(arch_rwlock_t *lock)
{
	unsigned long mask, tmp1, tmp2;

	mask = 0x80000000UL;

	__asm__ __volatile__(
"1:	lduw		[%2], %0\n"
"	brnz,pn		%0, 2f\n"
"4:	 or		%0, %3, %1\n"
"	cas		[%2], %0, %1\n"
"	cmp		%0, %1\n"
"	bne,pn		%%icc, 1b\n"
"	 nop\n"
"	.subsection	2\n"
"2:	lduw		[%2], %0\n"
"	brnz,pt		%0, 2b\n"
"	 nop\n"
"	ba,a,pt		%%xcc, 4b\n"
"	.previous"
	: "=&r" (tmp1), "=&r" (tmp2)
	: "r" (lock), "r" (mask)
	: "memory");
}

static void inline arch_write_unlock(arch_rwlock_t *lock)
{
	__asm__ __volatile__(
"	stw		%%g0, [%0]"
	: /* no outputs */
	: "r" (lock)
	: "memory");
}

static int inline arch_write_trylock(arch_rwlock_t *lock)
{
	unsigned long mask, tmp1, tmp2, result;

	mask = 0x80000000UL;

	__asm__ __volatile__(
"	mov		0, %2\n"
"1:	lduw		[%3], %0\n"
"	brnz,pn		%0, 2f\n"
"	 or		%0, %4, %1\n"
"	cas		[%3], %0, %1\n"
"	cmp		%0, %1\n"
"	bne,pn		%%icc, 1b\n"
"	 nop\n"
"	mov		1, %2\n"
"2:"
	: "=&r" (tmp1), "=&r" (tmp2), "=&r" (result)
	: "r" (lock), "r" (mask)
	: "memory");

	return result;
}

#define arch_read_lock(p)	arch_read_lock(p)
#define arch_read_lock_flags(p, f) arch_read_lock(p)
#define arch_read_trylock(p)	arch_read_trylock(p)
#define arch_read_unlock(p)	arch_read_unlock(p)
#define arch_write_lock(p)	arch_write_lock(p)
#define arch_write_lock_flags(p, f) arch_write_lock(p)
#define arch_write_unlock(p)	arch_write_unlock(p)
#define arch_write_trylock(p)	arch_write_trylock(p)

#define arch_read_can_lock(rw)		(!((rw)->lock & 0x80000000UL))
#define arch_write_can_lock(rw)	(!(rw)->lock)

#define arch_spin_relax(lock)	cpu_relax()
#define arch_read_relax(lock)	cpu_relax()
#define arch_write_relax(lock)	cpu_relax()

#endif /* !(__ASSEMBLY__) */

#endif /* !(__SPARC64_SPINLOCK_H) */