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
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
|
/* SPDX-License-Identifier: Apache-2.0
* Copyright 2011-2022 Blender Foundation */
#pragma once
#include "kernel/integrator/state.h"
#include "kernel/util/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, tmin) = ray->tmin;
INTEGRATOR_STATE_WRITE(state, ray, tmax) = ray->tmax;
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->tmin = INTEGRATOR_STATE(state, ray, tmin);
ray->tmax = INTEGRATOR_STATE(state, ray, tmax);
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, IntegratorShadowState 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, tmin) = ray->tmin;
INTEGRATOR_STATE_WRITE(state, shadow_ray, tmax) = ray->tmax;
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,
ConstIntegratorShadowState state,
ccl_private Ray *ccl_restrict ray)
{
ray->P = INTEGRATOR_STATE(state, shadow_ray, P);
ray->D = INTEGRATOR_STATE(state, shadow_ray, D);
ray->tmin = INTEGRATOR_STATE(state, shadow_ray, tmin);
ray->tmax = INTEGRATOR_STATE(state, shadow_ray, tmax);
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;
}
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);
}
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(
IntegratorShadowState 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;
}
ccl_device_forceinline void integrator_state_read_shadow_isect(
ConstIntegratorShadowState 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);
}
ccl_device_forceinline void integrator_state_copy_volume_stack_to_shadow(
KernelGlobals kg, IntegratorShadowState shadow_state, ConstIntegratorState 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(shadow_state, shadow_volume_stack, index, object) =
INTEGRATOR_STATE_ARRAY(state, volume_stack, index, object);
INTEGRATOR_STATE_ARRAY_WRITE(shadow_state, shadow_volume_stack, index, shader) = shader;
++index;
} while (shader != OBJECT_NONE);
}
}
ccl_device_forceinline void integrator_state_copy_volume_stack(KernelGlobals kg,
IntegratorState to_state,
ConstIntegratorState 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(to_state, volume_stack, index, object) = INTEGRATOR_STATE_ARRAY(
state, volume_stack, index, object);
INTEGRATOR_STATE_ARRAY_WRITE(to_state, volume_stack, index, shader) = shader;
++index;
} while (shader != OBJECT_NONE);
}
}
ccl_device_forceinline VolumeStack
integrator_state_read_shadow_volume_stack(ConstIntegratorShadowState 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, ConstIntegratorShadowState 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(IntegratorShadowState 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(KernelGlobals kg,
ConstIntegratorState to_state,
ConstIntegratorState 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/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(KernelGlobals kg,
ConstIntegratorState to_state,
ConstIntegratorState state)
{
integrator_state_copy_only(kg, to_state, state);
INTEGRATOR_STATE_WRITE(state, path, queued_kernel) = 0;
}
ccl_device_inline void integrator_shadow_state_copy_only(KernelGlobals kg,
ConstIntegratorShadowState to_state,
ConstIntegratorShadowState 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/shadow_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_shadow_state_move(KernelGlobals kg,
ConstIntegratorState to_state,
ConstIntegratorState state)
{
integrator_shadow_state_copy_only(kg, to_state, state);
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 IntegratorState integrator_state_shadow_catcher_split(KernelGlobals kg,
IntegratorState state)
{
#if defined(__KERNEL_GPU__)
ConstIntegratorState to_state = atomic_fetch_and_add_uint32(
&kernel_integrator_state.next_main_path_index[0], 1);
integrator_state_copy_only(kg, to_state, state);
#else
IntegratorStateCPU *ccl_restrict to_state = state + 1;
/* Only copy the required subset for performance. */
to_state->path = state->path;
to_state->ray = state->ray;
to_state->isect = state->isect;
integrator_state_copy_volume_stack(kg, to_state, state);
#endif
return to_state;
}
#ifdef __KERNEL_CPU__
ccl_device_inline int integrator_state_bounce(ConstIntegratorState state, const int)
{
return INTEGRATOR_STATE(state, path, bounce);
}
ccl_device_inline int integrator_state_bounce(ConstIntegratorShadowState state, const int)
{
return INTEGRATOR_STATE(state, shadow_path, bounce);
}
ccl_device_inline int integrator_state_diffuse_bounce(ConstIntegratorState state, const int)
{
return INTEGRATOR_STATE(state, path, diffuse_bounce);
}
ccl_device_inline int integrator_state_diffuse_bounce(ConstIntegratorShadowState state, const int)
{
return INTEGRATOR_STATE(state, shadow_path, diffuse_bounce);
}
ccl_device_inline int integrator_state_glossy_bounce(ConstIntegratorState state, const int)
{
return INTEGRATOR_STATE(state, path, glossy_bounce);
}
ccl_device_inline int integrator_state_glossy_bounce(ConstIntegratorShadowState state, const int)
{
return INTEGRATOR_STATE(state, shadow_path, glossy_bounce);
}
ccl_device_inline int integrator_state_transmission_bounce(ConstIntegratorState state, const int)
{
return INTEGRATOR_STATE(state, path, transmission_bounce);
}
ccl_device_inline int integrator_state_transmission_bounce(ConstIntegratorShadowState state,
const int)
{
return INTEGRATOR_STATE(state, shadow_path, transmission_bounce);
}
ccl_device_inline int integrator_state_transparent_bounce(ConstIntegratorState state, const int)
{
return INTEGRATOR_STATE(state, path, transparent_bounce);
}
ccl_device_inline int integrator_state_transparent_bounce(ConstIntegratorShadowState state,
const int)
{
return INTEGRATOR_STATE(state, shadow_path, transparent_bounce);
}
#else
ccl_device_inline int integrator_state_bounce(ConstIntegratorShadowState state,
const uint32_t path_flag)
{
return (path_flag & PATH_RAY_SHADOW) ? INTEGRATOR_STATE(state, shadow_path, bounce) :
INTEGRATOR_STATE(state, path, bounce);
}
ccl_device_inline int integrator_state_diffuse_bounce(ConstIntegratorShadowState state,
const uint32_t path_flag)
{
return (path_flag & PATH_RAY_SHADOW) ? INTEGRATOR_STATE(state, shadow_path, diffuse_bounce) :
INTEGRATOR_STATE(state, path, diffuse_bounce);
}
ccl_device_inline int integrator_state_glossy_bounce(ConstIntegratorShadowState state,
const uint32_t path_flag)
{
return (path_flag & PATH_RAY_SHADOW) ? INTEGRATOR_STATE(state, shadow_path, glossy_bounce) :
INTEGRATOR_STATE(state, path, glossy_bounce);
}
ccl_device_inline int integrator_state_transmission_bounce(ConstIntegratorShadowState state,
const uint32_t path_flag)
{
return (path_flag & PATH_RAY_SHADOW) ?
INTEGRATOR_STATE(state, shadow_path, transmission_bounce) :
INTEGRATOR_STATE(state, path, transmission_bounce);
}
ccl_device_inline int integrator_state_transparent_bounce(ConstIntegratorShadowState state,
const uint32_t path_flag)
{
return (path_flag & PATH_RAY_SHADOW) ? INTEGRATOR_STATE(state, shadow_path, transparent_bounce) :
INTEGRATOR_STATE(state, path, transparent_bounce);
}
#endif
CCL_NAMESPACE_END
|