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
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
|
/*
* Copyright 2011-2021 Blender Foundation
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#pragma once
#include "kernel/integrator/integrator_state.h"
#include "kernel/kernel_differential.h"
CCL_NAMESPACE_BEGIN
/* Ray */
ccl_device_forceinline void integrator_state_write_ray(KernelGlobals kg,
IntegratorState state,
ccl_private const Ray *ccl_restrict ray)
{
INTEGRATOR_STATE_WRITE(state, ray, P) = ray->P;
INTEGRATOR_STATE_WRITE(state, ray, D) = ray->D;
INTEGRATOR_STATE_WRITE(state, ray, t) = ray->t;
INTEGRATOR_STATE_WRITE(state, ray, time) = ray->time;
INTEGRATOR_STATE_WRITE(state, ray, dP) = ray->dP;
INTEGRATOR_STATE_WRITE(state, ray, dD) = ray->dD;
}
ccl_device_forceinline void integrator_state_read_ray(KernelGlobals kg,
ConstIntegratorState state,
ccl_private Ray *ccl_restrict ray)
{
ray->P = INTEGRATOR_STATE(state, ray, P);
ray->D = INTEGRATOR_STATE(state, ray, D);
ray->t = INTEGRATOR_STATE(state, ray, t);
ray->time = INTEGRATOR_STATE(state, ray, time);
ray->dP = INTEGRATOR_STATE(state, ray, dP);
ray->dD = INTEGRATOR_STATE(state, ray, dD);
}
/* Shadow Ray */
ccl_device_forceinline void integrator_state_write_shadow_ray(
KernelGlobals kg, IntegratorState state, ccl_private const Ray *ccl_restrict ray)
{
INTEGRATOR_STATE_WRITE(state, shadow_ray, P) = ray->P;
INTEGRATOR_STATE_WRITE(state, shadow_ray, D) = ray->D;
INTEGRATOR_STATE_WRITE(state, shadow_ray, t) = ray->t;
INTEGRATOR_STATE_WRITE(state, shadow_ray, time) = ray->time;
INTEGRATOR_STATE_WRITE(state, shadow_ray, dP) = ray->dP;
}
ccl_device_forceinline void integrator_state_read_shadow_ray(KernelGlobals kg,
ConstIntegratorState state,
ccl_private Ray *ccl_restrict ray)
{
ray->P = INTEGRATOR_STATE(state, shadow_ray, P);
ray->D = INTEGRATOR_STATE(state, shadow_ray, D);
ray->t = INTEGRATOR_STATE(state, shadow_ray, t);
ray->time = INTEGRATOR_STATE(state, shadow_ray, time);
ray->dP = INTEGRATOR_STATE(state, shadow_ray, dP);
ray->dD = differential_zero_compact();
}
/* Intersection */
ccl_device_forceinline void integrator_state_write_isect(
KernelGlobals kg, IntegratorState state, ccl_private const Intersection *ccl_restrict isect)
{
INTEGRATOR_STATE_WRITE(state, isect, t) = isect->t;
INTEGRATOR_STATE_WRITE(state, isect, u) = isect->u;
INTEGRATOR_STATE_WRITE(state, isect, v) = isect->v;
INTEGRATOR_STATE_WRITE(state, isect, object) = isect->object;
INTEGRATOR_STATE_WRITE(state, isect, prim) = isect->prim;
INTEGRATOR_STATE_WRITE(state, isect, type) = isect->type;
#ifdef __EMBREE__
INTEGRATOR_STATE_WRITE(state, isect, Ng) = isect->Ng;
#endif
}
ccl_device_forceinline void integrator_state_read_isect(
KernelGlobals kg, ConstIntegratorState state, ccl_private Intersection *ccl_restrict isect)
{
isect->prim = INTEGRATOR_STATE(state, isect, prim);
isect->object = INTEGRATOR_STATE(state, isect, object);
isect->type = INTEGRATOR_STATE(state, isect, type);
isect->u = INTEGRATOR_STATE(state, isect, u);
isect->v = INTEGRATOR_STATE(state, isect, v);
isect->t = INTEGRATOR_STATE(state, isect, t);
#ifdef __EMBREE__
isect->Ng = INTEGRATOR_STATE(state, isect, Ng);
#endif
}
ccl_device_forceinline VolumeStack integrator_state_read_volume_stack(ConstIntegratorState state,
int i)
{
VolumeStack entry = {INTEGRATOR_STATE_ARRAY(state, volume_stack, i, object),
INTEGRATOR_STATE_ARRAY(state, volume_stack, i, shader)};
return entry;
}
ccl_device_forceinline void integrator_state_write_volume_stack(IntegratorState state,
int i,
VolumeStack entry)
{
INTEGRATOR_STATE_ARRAY_WRITE(state, volume_stack, i, object) = entry.object;
INTEGRATOR_STATE_ARRAY_WRITE(state, volume_stack, i, shader) = entry.shader;
}
ccl_device_forceinline bool integrator_state_volume_stack_is_empty(KernelGlobals kg,
ConstIntegratorState state)
{
return (kernel_data.kernel_features & KERNEL_FEATURE_VOLUME) ?
INTEGRATOR_STATE_ARRAY(state, volume_stack, 0, shader) == SHADER_NONE :
true;
}
/* Shadow Intersection */
ccl_device_forceinline void integrator_state_write_shadow_isect(
IntegratorState state, ccl_private const Intersection *ccl_restrict isect, const int index)
{
INTEGRATOR_STATE_ARRAY_WRITE(state, shadow_isect, index, t) = isect->t;
INTEGRATOR_STATE_ARRAY_WRITE(state, shadow_isect, index, u) = isect->u;
INTEGRATOR_STATE_ARRAY_WRITE(state, shadow_isect, index, v) = isect->v;
INTEGRATOR_STATE_ARRAY_WRITE(state, shadow_isect, index, object) = isect->object;
INTEGRATOR_STATE_ARRAY_WRITE(state, shadow_isect, index, prim) = isect->prim;
INTEGRATOR_STATE_ARRAY_WRITE(state, shadow_isect, index, type) = isect->type;
#ifdef __EMBREE__
INTEGRATOR_STATE_ARRAY_WRITE(state, shadow_isect, index, Ng) = isect->Ng;
#endif
}
ccl_device_forceinline void integrator_state_read_shadow_isect(
ConstIntegratorState state, ccl_private Intersection *ccl_restrict isect, const int index)
{
isect->prim = INTEGRATOR_STATE_ARRAY(state, shadow_isect, index, prim);
isect->object = INTEGRATOR_STATE_ARRAY(state, shadow_isect, index, object);
isect->type = INTEGRATOR_STATE_ARRAY(state, shadow_isect, index, type);
isect->u = INTEGRATOR_STATE_ARRAY(state, shadow_isect, index, u);
isect->v = INTEGRATOR_STATE_ARRAY(state, shadow_isect, index, v);
isect->t = INTEGRATOR_STATE_ARRAY(state, shadow_isect, index, t);
#ifdef __EMBREE__
isect->Ng = INTEGRATOR_STATE_ARRAY(state, shadow_isect, index, Ng);
#endif
}
ccl_device_forceinline void integrator_state_copy_volume_stack_to_shadow(KernelGlobals kg,
IntegratorState state)
{
if (kernel_data.kernel_features & KERNEL_FEATURE_VOLUME) {
int index = 0;
int shader;
do {
shader = INTEGRATOR_STATE_ARRAY(state, volume_stack, index, shader);
INTEGRATOR_STATE_ARRAY_WRITE(state, shadow_volume_stack, index, object) =
INTEGRATOR_STATE_ARRAY(state, volume_stack, index, object);
INTEGRATOR_STATE_ARRAY_WRITE(state, shadow_volume_stack, index, shader) = shader;
++index;
} while (shader != OBJECT_NONE);
}
}
ccl_device_forceinline VolumeStack
integrator_state_read_shadow_volume_stack(ConstIntegratorState state, int i)
{
VolumeStack entry = {INTEGRATOR_STATE_ARRAY(state, shadow_volume_stack, i, object),
INTEGRATOR_STATE_ARRAY(state, shadow_volume_stack, i, shader)};
return entry;
}
ccl_device_forceinline bool integrator_state_shadow_volume_stack_is_empty(
KernelGlobals kg, ConstIntegratorState state)
{
return (kernel_data.kernel_features & KERNEL_FEATURE_VOLUME) ?
INTEGRATOR_STATE_ARRAY(state, shadow_volume_stack, 0, shader) == SHADER_NONE :
true;
}
ccl_device_forceinline void integrator_state_write_shadow_volume_stack(IntegratorState state,
int i,
VolumeStack entry)
{
INTEGRATOR_STATE_ARRAY_WRITE(state, shadow_volume_stack, i, object) = entry.object;
INTEGRATOR_STATE_ARRAY_WRITE(state, shadow_volume_stack, i, shader) = entry.shader;
}
#if defined(__KERNEL_GPU__)
ccl_device_inline void integrator_state_copy_only(const IntegratorState to_state,
const IntegratorState state)
{
int index;
/* Rely on the compiler to optimize out unused assignments and `while(false)`'s. */
# define KERNEL_STRUCT_BEGIN(name) \
index = 0; \
do {
# define KERNEL_STRUCT_MEMBER(parent_struct, type, name, feature) \
if (kernel_integrator_state.parent_struct.name != nullptr) { \
kernel_integrator_state.parent_struct.name[to_state] = \
kernel_integrator_state.parent_struct.name[state]; \
}
# define KERNEL_STRUCT_ARRAY_MEMBER(parent_struct, type, name, feature) \
if (kernel_integrator_state.parent_struct[index].name != nullptr) { \
kernel_integrator_state.parent_struct[index].name[to_state] = \
kernel_integrator_state.parent_struct[index].name[state]; \
}
# define KERNEL_STRUCT_END(name) \
} \
while (false) \
;
# define KERNEL_STRUCT_END_ARRAY(name, cpu_array_size, gpu_array_size) \
++index; \
} \
while (index < gpu_array_size) \
;
# define KERNEL_STRUCT_VOLUME_STACK_SIZE kernel_data.volume_stack_size
# include "kernel/integrator/integrator_state_template.h"
# undef KERNEL_STRUCT_BEGIN
# undef KERNEL_STRUCT_MEMBER
# undef KERNEL_STRUCT_ARRAY_MEMBER
# undef KERNEL_STRUCT_END
# undef KERNEL_STRUCT_END_ARRAY
# undef KERNEL_STRUCT_VOLUME_STACK_SIZE
}
ccl_device_inline void integrator_state_move(const IntegratorState to_state,
const IntegratorState state)
{
integrator_state_copy_only(to_state, state);
INTEGRATOR_STATE_WRITE(state, path, queued_kernel) = 0;
INTEGRATOR_STATE_WRITE(state, shadow_path, queued_kernel) = 0;
}
#endif
/* NOTE: Leaves kernel scheduling information untouched. Use INIT semantic for one of the paths
* after this function. */
ccl_device_inline void integrator_state_shadow_catcher_split(KernelGlobals kg,
IntegratorState state)
{
#if defined(__KERNEL_GPU__)
const IntegratorState to_state = atomic_fetch_and_add_uint32(
&kernel_integrator_state.next_shadow_catcher_path_index[0], 1);
integrator_state_copy_only(to_state, state);
kernel_integrator_state.path.flag[to_state] |= PATH_RAY_SHADOW_CATCHER_PASS;
#else
IntegratorStateCPU *ccl_restrict split_state = state + 1;
/* Only copy the required subset, since shadow intersections are big and irrelevant here. */
split_state->path = state->path;
split_state->ray = state->ray;
split_state->isect = state->isect;
memcpy(split_state->volume_stack, state->volume_stack, sizeof(state->volume_stack));
split_state->shadow_path = state->shadow_path;
split_state->path.flag |= PATH_RAY_SHADOW_CATCHER_PASS;
#endif
}
CCL_NAMESPACE_END
|