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
|
/*
* Performance events support for SH-4A performance counters
*
* Copyright (C) 2009 Paul Mundt
*
* 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.
*/
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/io.h>
#include <linux/irq.h>
#include <linux/perf_event.h>
#include <asm/processor.h>
#define PPC_CCBR(idx) (0xff200800 + (sizeof(u32) * idx))
#define PPC_PMCTR(idx) (0xfc100000 + (sizeof(u32) * idx))
#define CCBR_CIT_MASK (0x7ff << 6)
#define CCBR_DUC (1 << 3)
#define CCBR_CMDS (1 << 1)
#define CCBR_PPCE (1 << 0)
#define PPC_PMCAT 0xfc100080
#define PMCAT_OVF3 (1 << 27)
#define PMCAT_CNN3 (1 << 26)
#define PMCAT_CLR3 (1 << 25)
#define PMCAT_OVF2 (1 << 19)
#define PMCAT_CLR2 (1 << 17)
#define PMCAT_OVF1 (1 << 11)
#define PMCAT_CNN1 (1 << 10)
#define PMCAT_CLR1 (1 << 9)
#define PMCAT_OVF0 (1 << 3)
#define PMCAT_CLR0 (1 << 1)
static struct sh_pmu sh4a_pmu;
/*
* Special reserved bits used by hardware emulators, read values will
* vary, but writes must always be 0.
*/
#define PMCAT_EMU_CLR_MASK ((1 << 24) | (1 << 16) | (1 << 8) | (1 << 0))
static const int sh4a_general_events[] = {
[PERF_COUNT_HW_CPU_CYCLES] = 0x0000,
[PERF_COUNT_HW_INSTRUCTIONS] = 0x0202,
[PERF_COUNT_HW_CACHE_REFERENCES] = 0x0029, /* I-cache */
[PERF_COUNT_HW_CACHE_MISSES] = 0x002a, /* I-cache */
[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x0204,
[PERF_COUNT_HW_BRANCH_MISSES] = -1,
[PERF_COUNT_HW_BUS_CYCLES] = -1,
};
#define C(x) PERF_COUNT_HW_CACHE_##x
static const int sh4a_cache_events
[PERF_COUNT_HW_CACHE_MAX]
[PERF_COUNT_HW_CACHE_OP_MAX]
[PERF_COUNT_HW_CACHE_RESULT_MAX] =
{
[ C(L1D) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x0031,
[ C(RESULT_MISS) ] = 0x0032,
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = 0x0039,
[ C(RESULT_MISS) ] = 0x003a,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0,
},
},
[ C(L1I) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x0029,
[ C(RESULT_MISS) ] = 0x002a,
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0,
},
},
[ C(LL) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x0030,
[ C(RESULT_MISS) ] = 0,
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = 0x0038,
[ C(RESULT_MISS) ] = 0,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0,
},
},
[ C(DTLB) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0x0222,
[ C(RESULT_MISS) ] = 0x0220,
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0,
},
},
[ C(ITLB) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = 0,
[ C(RESULT_MISS) ] = 0x02a0,
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
},
[ C(BPU) ] = {
[ C(OP_READ) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
[ C(OP_WRITE) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
[ C(OP_PREFETCH) ] = {
[ C(RESULT_ACCESS) ] = -1,
[ C(RESULT_MISS) ] = -1,
},
},
};
static int sh4a_event_map(int event)
{
return sh4a_general_events[event];
}
static u64 sh4a_pmu_read(int idx)
{
return __raw_readl(PPC_PMCTR(idx));
}
static void sh4a_pmu_disable(struct hw_perf_event *hwc, int idx)
{
unsigned int tmp;
tmp = __raw_readl(PPC_CCBR(idx));
tmp &= ~(CCBR_CIT_MASK | CCBR_DUC);
__raw_writel(tmp, PPC_CCBR(idx));
}
static void sh4a_pmu_enable(struct hw_perf_event *hwc, int idx)
{
unsigned int tmp;
tmp = __raw_readl(PPC_PMCAT);
tmp &= ~PMCAT_EMU_CLR_MASK;
tmp |= idx ? PMCAT_CLR1 : PMCAT_CLR0;
__raw_writel(tmp, PPC_PMCAT);
tmp = __raw_readl(PPC_CCBR(idx));
tmp |= (hwc->config << 6) | CCBR_CMDS | CCBR_PPCE;
__raw_writel(tmp, PPC_CCBR(idx));
__raw_writel(__raw_readl(PPC_CCBR(idx)) | CCBR_DUC, PPC_CCBR(idx));
}
static void sh4a_pmu_disable_all(void)
{
int i;
for (i = 0; i < sh4a_pmu.num_events; i++)
__raw_writel(__raw_readl(PPC_CCBR(i)) & ~CCBR_DUC, PPC_CCBR(i));
}
static void sh4a_pmu_enable_all(void)
{
int i;
for (i = 0; i < sh4a_pmu.num_events; i++)
__raw_writel(__raw_readl(PPC_CCBR(i)) | CCBR_DUC, PPC_CCBR(i));
}
static struct sh_pmu sh4a_pmu = {
.name = "SH-4A",
.num_events = 2,
.event_map = sh4a_event_map,
.max_events = ARRAY_SIZE(sh4a_general_events),
.raw_event_mask = 0x3ff,
.cache_events = &sh4a_cache_events,
.read = sh4a_pmu_read,
.disable = sh4a_pmu_disable,
.enable = sh4a_pmu_enable,
.disable_all = sh4a_pmu_disable_all,
.enable_all = sh4a_pmu_enable_all,
};
static int __init sh4a_pmu_init(void)
{
/*
* Make sure this CPU actually has perf counters.
*/
if (!(boot_cpu_data.flags & CPU_HAS_PERF_COUNTER)) {
pr_notice("HW perf events unsupported, software events only.\n");
return -ENODEV;
}
return register_sh_pmu(&sh4a_pmu);
}
arch_initcall(sh4a_pmu_init);
|
