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
|
/*
* Copyright (C) 2016 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#include <stdio.h>
#include <stdint.h>
#include <stdbool.h>
#include <assert.h>
#include <math.h>
#include <gallium/auxiliary/util/u_math.h>
#ifdef HAVE_VALGRIND
#include <valgrind.h>
#include <memcheck.h>
#define VG(x) x
#ifndef NDEBUG
#define __gen_validate_value(x) VALGRIND_CHECK_MEM_IS_DEFINED(&(x), sizeof(x))
#endif
#else
#define VG(x)
#endif
#ifndef __gen_validate_value
#define __gen_validate_value(x)
#endif
/*
#ifndef __gen_address_type
#error #define __gen_address_type before including this file
#endif
#ifndef __gen_user_data
#error #define __gen_combine_address before including this file
#endif
*/
union __gen_value {
float f;
uint32_t dw;
};
static inline uint64_t
__gen_mbo(uint32_t start, uint32_t end)
{
return (~0ull >> (64 - (end - start + 1))) << start;
}
static inline uint64_t
__gen_uint(uint64_t v, uint32_t start, uint32_t end)
{
__gen_validate_value(v);
#ifndef NDEBUG
const int width = end - start + 1;
if (width < 64) {
const uint64_t max = (1ull << width) - 1;
assert(v <= max);
}
#endif
return v << start;
}
static inline uint64_t
__gen_sint(int64_t v, uint32_t start, uint32_t end)
{
const int width = end - start + 1;
__gen_validate_value(v);
#ifndef NDEBUG
if (width < 64) {
const int64_t max = (1ll << (width - 1)) - 1;
const int64_t min = -(1ll << (width - 1));
assert(min <= v && v <= max);
}
#endif
const uint64_t mask = ~0ull >> (64 - width);
return (v & mask) << start;
}
static inline uint64_t
__gen_offset(uint64_t v, uint32_t start, uint32_t end)
{
__gen_validate_value(v);
#ifndef NDEBUG
uint64_t mask = (~0ull >> (64 - (end - start + 1))) << start;
assert((v & ~mask) == 0);
#endif
return v;
}
static inline uint32_t
__gen_float(float v)
{
__gen_validate_value(v);
return ((union __gen_value) { .f = (v) }).dw;
}
static inline uint64_t
__gen_sfixed(float v, uint32_t start, uint32_t end, uint32_t fract_bits)
{
__gen_validate_value(v);
const float factor = (1 << fract_bits);
#ifndef NDEBUG
const float max = ((1 << (end - start)) - 1) / factor;
const float min = -(1 << (end - start)) / factor;
assert(min <= v && v <= max);
#endif
const int64_t int_val = llroundf(v * factor);
const uint64_t mask = ~0ull >> (64 - (end - start + 1));
return (int_val & mask) << start;
}
static inline uint64_t
__gen_ufixed(float v, uint32_t start, uint32_t end, uint32_t fract_bits)
{
__gen_validate_value(v);
const float factor = (1 << fract_bits);
#ifndef NDEBUG
const float max = ((1 << (end - start + 1)) - 1) / factor;
const float min = 0.0f;
assert(min <= v && v <= max);
#endif
const uint64_t uint_val = llroundf(v * factor);
return uint_val << start;
}
static inline uint64_t
__gen_unpack_uint(const uint8_t *restrict cl, uint32_t start, uint32_t end)
{
uint64_t val = 0;
const int width = end - start + 1;
const uint32_t mask = (width == 32 ? ~0 : (1 << width) - 1 );
for (int byte = start / 8; byte <= end / 8; byte++) {
val |= cl[byte] << ((byte - start / 8) * 8);
}
return (val >> (start % 8)) & mask;
}
static inline uint64_t
__gen_unpack_sint(const uint8_t *restrict cl, uint32_t start, uint32_t end)
{
int size = end - start + 1;
int64_t val = __gen_unpack_uint(cl, start, end);
/* Get the sign bit extended. */
return (val << (64 - size)) >> (64 - size);
}
static inline float
__gen_unpack_sfixed(const uint8_t *restrict cl, uint32_t start, uint32_t end,
uint32_t fractional_size)
{
int32_t bits = __gen_unpack_sint(cl, start, end);
return (float)bits / (1 << fractional_size);
}
static inline float
__gen_unpack_ufixed(const uint8_t *restrict cl, uint32_t start, uint32_t end,
uint32_t fractional_size)
{
int32_t bits = __gen_unpack_uint(cl, start, end);
return (float)bits / (1 << fractional_size);
}
static inline float
__gen_unpack_float(const uint8_t *restrict cl, uint32_t start, uint32_t end)
{
assert(start % 8 == 0);
assert(end - start == 31);
struct PACKED { float f; } *f = (void *)(cl + (start / 8));
return f->f;
}
static inline float
__gen_unpack_f187(const uint8_t *restrict cl, uint32_t start, uint32_t end)
{
assert(end - start == 15);
uint32_t bits = __gen_unpack_uint(cl, start, end);
return uif(bits << 16);
}
|
