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
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
|
/*
* Copyright 2011-2013 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.
*/
#ifdef __OSL__
# include "kernel/osl/osl_shader.h"
#endif
// clang-format off
#include "kernel/kernel_random.h"
#include "kernel/kernel_projection.h"
#include "kernel/kernel_montecarlo.h"
#include "kernel/kernel_differential.h"
#include "kernel/kernel_camera.h"
#include "kernel/geom/geom.h"
#include "kernel/bvh/bvh.h"
#include "kernel/kernel_write_passes.h"
#include "kernel/kernel_accumulate.h"
#include "kernel/kernel_shader.h"
#include "kernel/kernel_light.h"
#include "kernel/kernel_adaptive_sampling.h"
#include "kernel/kernel_passes.h"
#if defined(__VOLUME__) || defined(__SUBSURFACE__)
# include "kernel/kernel_volume.h"
#endif
#ifdef __SUBSURFACE__
# include "kernel/kernel_subsurface.h"
#endif
#include "kernel/kernel_path_state.h"
#include "kernel/kernel_shadow.h"
#include "kernel/kernel_emission.h"
#include "kernel/kernel_path_common.h"
#include "kernel/kernel_path_surface.h"
#include "kernel/kernel_path_volume.h"
#include "kernel/kernel_path_subsurface.h"
// clang-format on
CCL_NAMESPACE_BEGIN
ccl_device_forceinline bool kernel_path_scene_intersect(KernelGlobals *kg,
ccl_addr_space PathState *state,
Ray *ray,
Intersection *isect,
PathRadiance *L)
{
PROFILING_INIT(kg, PROFILING_SCENE_INTERSECT);
uint visibility = path_state_ray_visibility(kg, state);
if (path_state_ao_bounce(kg, state)) {
visibility = PATH_RAY_SHADOW;
ray->t = kernel_data.background.ao_distance;
}
bool hit = scene_intersect(kg, ray, visibility, isect);
#ifdef __KERNEL_DEBUG__
if (state->flag & PATH_RAY_CAMERA) {
L->debug_data.num_bvh_traversed_nodes += isect->num_traversed_nodes;
L->debug_data.num_bvh_traversed_instances += isect->num_traversed_instances;
L->debug_data.num_bvh_intersections += isect->num_intersections;
}
L->debug_data.num_ray_bounces++;
#endif /* __KERNEL_DEBUG__ */
return hit;
}
ccl_device_forceinline void kernel_path_lamp_emission(KernelGlobals *kg,
ccl_addr_space PathState *state,
Ray *ray,
float3 throughput,
ccl_addr_space Intersection *isect,
ShaderData *emission_sd,
PathRadiance *L)
{
PROFILING_INIT(kg, PROFILING_INDIRECT_EMISSION);
#ifdef __LAMP_MIS__
if (kernel_data.integrator.use_lamp_mis && !(state->flag & PATH_RAY_CAMERA)) {
/* ray starting from previous non-transparent bounce */
Ray light_ray ccl_optional_struct_init;
light_ray.P = ray->P - state->ray_t * ray->D;
state->ray_t += isect->t;
light_ray.D = ray->D;
light_ray.t = state->ray_t;
light_ray.time = ray->time;
light_ray.dD = ray->dD;
light_ray.dP = ray->dP;
/* intersect with lamp */
indirect_lamp_emission(kg, emission_sd, state, L, &light_ray, throughput);
}
#endif /* __LAMP_MIS__ */
}
ccl_device_forceinline void kernel_path_background(KernelGlobals *kg,
ccl_addr_space PathState *state,
ccl_addr_space Ray *ray,
float3 throughput,
ShaderData *sd,
ccl_global float *buffer,
PathRadiance *L)
{
/* eval background shader if nothing hit */
if (kernel_data.background.transparent && (state->flag & PATH_RAY_TRANSPARENT_BACKGROUND)) {
L->transparent += average(throughput);
#ifdef __PASSES__
if (!(kernel_data.film.light_pass_flag & PASSMASK(BACKGROUND)))
#endif /* __PASSES__ */
return;
}
/* When using the ao bounces approximation, adjust background
* shader intensity with ao factor. */
if (path_state_ao_bounce(kg, state)) {
throughput *= kernel_data.background.ao_bounces_factor;
}
#ifdef __BACKGROUND__
/* sample background shader */
float3 L_background = indirect_background(kg, sd, state, buffer, ray);
path_radiance_accum_background(kg, L, state, throughput, L_background);
#endif /* __BACKGROUND__ */
}
#ifndef __SPLIT_KERNEL__
# ifdef __VOLUME__
ccl_device_forceinline VolumeIntegrateResult kernel_path_volume(KernelGlobals *kg,
ShaderData *sd,
PathState *state,
Ray *ray,
float3 *throughput,
ccl_addr_space Intersection *isect,
bool hit,
ShaderData *emission_sd,
PathRadiance *L)
{
PROFILING_INIT(kg, PROFILING_VOLUME);
/* Sanitize volume stack. */
if (!hit) {
kernel_volume_clean_stack(kg, state->volume_stack);
}
if (state->volume_stack[0].shader == SHADER_NONE) {
return VOLUME_PATH_ATTENUATED;
}
/* volume attenuation, emission, scatter */
Ray volume_ray = *ray;
volume_ray.t = (hit) ? isect->t : FLT_MAX;
bool heterogeneous = volume_stack_is_heterogeneous(kg, state->volume_stack);
# ifdef __VOLUME_DECOUPLED__
int sampling_method = volume_stack_sampling_method(kg, state->volume_stack);
bool direct = (state->flag & PATH_RAY_CAMERA) != 0;
bool decoupled = kernel_volume_use_decoupled(kg, heterogeneous, direct, sampling_method);
if (decoupled) {
/* cache steps along volume for repeated sampling */
VolumeSegment volume_segment;
shader_setup_from_volume(kg, sd, &volume_ray);
kernel_volume_decoupled_record(kg, state, &volume_ray, sd, &volume_segment, heterogeneous);
volume_segment.sampling_method = sampling_method;
/* emission */
if (volume_segment.closure_flag & SD_EMISSION)
path_radiance_accum_emission(kg, L, state, *throughput, volume_segment.accum_emission);
/* scattering */
VolumeIntegrateResult result = VOLUME_PATH_ATTENUATED;
if (volume_segment.closure_flag & SD_SCATTER) {
int all = kernel_data.integrator.sample_all_lights_indirect;
/* direct light sampling */
kernel_branched_path_volume_connect_light(
kg, sd, emission_sd, *throughput, state, L, all, &volume_ray, &volume_segment);
/* indirect sample. if we use distance sampling and take just
* one sample for direct and indirect light, we could share
* this computation, but makes code a bit complex */
float rphase = path_state_rng_1D(kg, state, PRNG_PHASE_CHANNEL);
float rscatter = path_state_rng_1D(kg, state, PRNG_SCATTER_DISTANCE);
result = kernel_volume_decoupled_scatter(
kg, state, &volume_ray, sd, throughput, rphase, rscatter, &volume_segment, NULL, true);
}
/* free cached steps */
kernel_volume_decoupled_free(kg, &volume_segment);
if (result == VOLUME_PATH_SCATTERED) {
if (kernel_path_volume_bounce(kg, sd, throughput, state, &L->state, ray))
return VOLUME_PATH_SCATTERED;
else
return VOLUME_PATH_MISSED;
}
else {
*throughput *= volume_segment.accum_transmittance;
}
}
else
# endif /* __VOLUME_DECOUPLED__ */
{
/* integrate along volume segment with distance sampling */
VolumeIntegrateResult result = kernel_volume_integrate(
kg, state, sd, &volume_ray, L, throughput, heterogeneous);
# ifdef __VOLUME_SCATTER__
if (result == VOLUME_PATH_SCATTERED) {
/* direct lighting */
kernel_path_volume_connect_light(kg, sd, emission_sd, *throughput, state, L);
/* indirect light bounce */
if (kernel_path_volume_bounce(kg, sd, throughput, state, &L->state, ray))
return VOLUME_PATH_SCATTERED;
else
return VOLUME_PATH_MISSED;
}
# endif /* __VOLUME_SCATTER__ */
}
return VOLUME_PATH_ATTENUATED;
}
# endif /* __VOLUME__ */
#endif /* __SPLIT_KERNEL__ */
ccl_device_forceinline bool kernel_path_shader_apply(KernelGlobals *kg,
ShaderData *sd,
ccl_addr_space PathState *state,
ccl_addr_space Ray *ray,
float3 throughput,
ShaderData *emission_sd,
PathRadiance *L,
ccl_global float *buffer)
{
PROFILING_INIT(kg, PROFILING_SHADER_APPLY);
#ifdef __SHADOW_TRICKS__
if ((sd->object_flag & SD_OBJECT_SHADOW_CATCHER)) {
if (state->flag & PATH_RAY_TRANSPARENT_BACKGROUND) {
state->flag |= (PATH_RAY_SHADOW_CATCHER | PATH_RAY_STORE_SHADOW_INFO);
float3 bg = make_float3(0.0f, 0.0f, 0.0f);
if (!kernel_data.background.transparent) {
bg = indirect_background(kg, emission_sd, state, NULL, ray);
}
path_radiance_accum_shadowcatcher(L, throughput, bg);
}
}
else if (state->flag & PATH_RAY_SHADOW_CATCHER) {
/* Only update transparency after shadow catcher bounce. */
L->shadow_transparency *= average(shader_bsdf_transparency(kg, sd));
}
#endif /* __SHADOW_TRICKS__ */
/* holdout */
#ifdef __HOLDOUT__
if (((sd->flag & SD_HOLDOUT) || (sd->object_flag & SD_OBJECT_HOLDOUT_MASK)) &&
(state->flag & PATH_RAY_TRANSPARENT_BACKGROUND)) {
if (kernel_data.background.transparent) {
float3 holdout_weight;
if (sd->object_flag & SD_OBJECT_HOLDOUT_MASK) {
holdout_weight = make_float3(1.0f, 1.0f, 1.0f);
}
else {
holdout_weight = shader_holdout_eval(kg, sd);
}
/* any throughput is ok, should all be identical here */
L->transparent += average(holdout_weight * throughput);
}
if (sd->object_flag & SD_OBJECT_HOLDOUT_MASK) {
return false;
}
}
#endif /* __HOLDOUT__ */
/* holdout mask objects do not write data passes */
kernel_write_data_passes(kg, buffer, L, sd, state, throughput);
/* blurring of bsdf after bounces, for rays that have a small likelihood
* of following this particular path (diffuse, rough glossy) */
if (kernel_data.integrator.filter_glossy != FLT_MAX) {
float blur_pdf = kernel_data.integrator.filter_glossy * state->min_ray_pdf;
if (blur_pdf < 1.0f) {
float blur_roughness = sqrtf(1.0f - blur_pdf) * 0.5f;
shader_bsdf_blur(kg, sd, blur_roughness);
}
}
#ifdef __EMISSION__
/* emission */
if (sd->flag & SD_EMISSION) {
float3 emission = indirect_primitive_emission(
kg, sd, sd->ray_length, state->flag, state->ray_pdf);
path_radiance_accum_emission(kg, L, state, throughput, emission);
}
#endif /* __EMISSION__ */
return true;
}
#ifdef __KERNEL_OPTIX__
ccl_device_inline /* inline trace calls */
#else
ccl_device_noinline
#endif
void
kernel_path_ao(KernelGlobals *kg,
ShaderData *sd,
ShaderData *emission_sd,
PathRadiance *L,
ccl_addr_space PathState *state,
float3 throughput,
float3 ao_alpha)
{
PROFILING_INIT(kg, PROFILING_AO);
/* todo: solve correlation */
float bsdf_u, bsdf_v;
path_state_rng_2D(kg, state, PRNG_BSDF_U, &bsdf_u, &bsdf_v);
float ao_factor = kernel_data.background.ao_factor;
float3 ao_N;
float3 ao_bsdf = shader_bsdf_ao(kg, sd, ao_factor, &ao_N);
float3 ao_D;
float ao_pdf;
sample_cos_hemisphere(ao_N, bsdf_u, bsdf_v, &ao_D, &ao_pdf);
if (dot(sd->Ng, ao_D) > 0.0f && ao_pdf != 0.0f) {
Ray light_ray;
float3 ao_shadow;
light_ray.P = ray_offset(sd->P, sd->Ng);
light_ray.D = ao_D;
light_ray.t = kernel_data.background.ao_distance;
light_ray.time = sd->time;
light_ray.dP = sd->dP;
light_ray.dD = differential3_zero();
if (!shadow_blocked(kg, sd, emission_sd, state, &light_ray, &ao_shadow)) {
path_radiance_accum_ao(kg, L, state, throughput, ao_alpha, ao_bsdf, ao_shadow);
}
else {
path_radiance_accum_total_ao(L, state, throughput, ao_bsdf);
}
}
}
#ifndef __SPLIT_KERNEL__
# if defined(__BRANCHED_PATH__) || defined(__BAKING__)
ccl_device void kernel_path_indirect(KernelGlobals *kg,
ShaderData *sd,
ShaderData *emission_sd,
Ray *ray,
float3 throughput,
PathState *state,
PathRadiance *L)
{
# ifdef __SUBSURFACE__
SubsurfaceIndirectRays ss_indirect;
kernel_path_subsurface_init_indirect(&ss_indirect);
for (;;) {
# endif /* __SUBSURFACE__ */
/* path iteration */
for (;;) {
/* Find intersection with objects in scene. */
Intersection isect;
bool hit = kernel_path_scene_intersect(kg, state, ray, &isect, L);
/* Find intersection with lamps and compute emission for MIS. */
kernel_path_lamp_emission(kg, state, ray, throughput, &isect, sd, L);
# ifdef __VOLUME__
/* Volume integration. */
VolumeIntegrateResult result = kernel_path_volume(
kg, sd, state, ray, &throughput, &isect, hit, emission_sd, L);
if (result == VOLUME_PATH_SCATTERED) {
continue;
}
else if (result == VOLUME_PATH_MISSED) {
break;
}
# endif /* __VOLUME__*/
/* Shade background. */
if (!hit) {
kernel_path_background(kg, state, ray, throughput, sd, NULL, L);
break;
}
else if (path_state_ao_bounce(kg, state)) {
break;
}
/* Setup shader data. */
shader_setup_from_ray(kg, sd, &isect, ray);
/* Skip most work for volume bounding surface. */
# ifdef __VOLUME__
if (!(sd->flag & SD_HAS_ONLY_VOLUME)) {
# endif
/* Evaluate shader. */
shader_eval_surface(kg, sd, state, NULL, state->flag);
shader_prepare_closures(sd, state);
/* Apply shadow catcher, holdout, emission. */
if (!kernel_path_shader_apply(kg, sd, state, ray, throughput, emission_sd, L, NULL)) {
break;
}
/* path termination. this is a strange place to put the termination, it's
* mainly due to the mixed in MIS that we use. gives too many unneeded
* shader evaluations, only need emission if we are going to terminate */
float probability = path_state_continuation_probability(kg, state, throughput);
if (probability == 0.0f) {
break;
}
else if (probability != 1.0f) {
float terminate = path_state_rng_1D(kg, state, PRNG_TERMINATE);
if (terminate >= probability)
break;
throughput /= probability;
}
# ifdef __DENOISING_FEATURES__
kernel_update_denoising_features(kg, sd, state, L);
# endif
# ifdef __AO__
/* ambient occlusion */
if (kernel_data.integrator.use_ambient_occlusion) {
kernel_path_ao(kg, sd, emission_sd, L, state, throughput, make_float3(0.0f, 0.0f, 0.0f));
}
# endif /* __AO__ */
# ifdef __SUBSURFACE__
/* bssrdf scatter to a different location on the same object, replacing
* the closures with a diffuse BSDF */
if (sd->flag & SD_BSSRDF) {
if (kernel_path_subsurface_scatter(
kg, sd, emission_sd, L, state, ray, &throughput, &ss_indirect)) {
break;
}
}
# endif /* __SUBSURFACE__ */
# if defined(__EMISSION__)
int all = (kernel_data.integrator.sample_all_lights_indirect) ||
(state->flag & PATH_RAY_SHADOW_CATCHER);
kernel_branched_path_surface_connect_light(
kg, sd, emission_sd, state, throughput, 1.0f, L, all);
# endif /* defined(__EMISSION__) */
# ifdef __VOLUME__
}
# endif
if (!kernel_path_surface_bounce(kg, sd, &throughput, state, &L->state, ray))
break;
}
# ifdef __SUBSURFACE__
/* Trace indirect subsurface rays by restarting the loop. this uses less
* stack memory than invoking kernel_path_indirect.
*/
if (ss_indirect.num_rays) {
kernel_path_subsurface_setup_indirect(kg, &ss_indirect, state, ray, L, &throughput);
}
else {
break;
}
}
# endif /* __SUBSURFACE__ */
}
# endif /* defined(__BRANCHED_PATH__) || defined(__BAKING__) */
ccl_device_forceinline void kernel_path_integrate(KernelGlobals *kg,
PathState *state,
float3 throughput,
Ray *ray,
PathRadiance *L,
ccl_global float *buffer,
ShaderData *emission_sd)
{
PROFILING_INIT(kg, PROFILING_PATH_INTEGRATE);
/* Shader data memory used for both volumes and surfaces, saves stack space. */
ShaderData sd;
# ifdef __SUBSURFACE__
SubsurfaceIndirectRays ss_indirect;
kernel_path_subsurface_init_indirect(&ss_indirect);
for (;;) {
# endif /* __SUBSURFACE__ */
/* path iteration */
for (;;) {
/* Find intersection with objects in scene. */
Intersection isect;
bool hit = kernel_path_scene_intersect(kg, state, ray, &isect, L);
/* Find intersection with lamps and compute emission for MIS. */
kernel_path_lamp_emission(kg, state, ray, throughput, &isect, &sd, L);
# ifdef __VOLUME__
/* Volume integration. */
VolumeIntegrateResult result = kernel_path_volume(
kg, &sd, state, ray, &throughput, &isect, hit, emission_sd, L);
if (result == VOLUME_PATH_SCATTERED) {
continue;
}
else if (result == VOLUME_PATH_MISSED) {
break;
}
# endif /* __VOLUME__*/
/* Shade background. */
if (!hit) {
kernel_path_background(kg, state, ray, throughput, &sd, buffer, L);
break;
}
else if (path_state_ao_bounce(kg, state)) {
break;
}
/* Setup shader data. */
shader_setup_from_ray(kg, &sd, &isect, ray);
/* Skip most work for volume bounding surface. */
# ifdef __VOLUME__
if (!(sd.flag & SD_HAS_ONLY_VOLUME)) {
# endif
/* Evaluate shader. */
shader_eval_surface(kg, &sd, state, buffer, state->flag);
shader_prepare_closures(&sd, state);
/* Apply shadow catcher, holdout, emission. */
if (!kernel_path_shader_apply(kg, &sd, state, ray, throughput, emission_sd, L, buffer)) {
break;
}
/* path termination. this is a strange place to put the termination, it's
* mainly due to the mixed in MIS that we use. gives too many unneeded
* shader evaluations, only need emission if we are going to terminate */
float probability = path_state_continuation_probability(kg, state, throughput);
if (probability == 0.0f) {
break;
}
else if (probability != 1.0f) {
float terminate = path_state_rng_1D(kg, state, PRNG_TERMINATE);
if (terminate >= probability)
break;
throughput /= probability;
}
# ifdef __DENOISING_FEATURES__
kernel_update_denoising_features(kg, &sd, state, L);
# endif
# ifdef __AO__
/* ambient occlusion */
if (kernel_data.integrator.use_ambient_occlusion) {
kernel_path_ao(kg, &sd, emission_sd, L, state, throughput, shader_bsdf_alpha(kg, &sd));
}
# endif /* __AO__ */
# ifdef __SUBSURFACE__
/* bssrdf scatter to a different location on the same object, replacing
* the closures with a diffuse BSDF */
if (sd.flag & SD_BSSRDF) {
if (kernel_path_subsurface_scatter(
kg, &sd, emission_sd, L, state, ray, &throughput, &ss_indirect)) {
break;
}
}
# endif /* __SUBSURFACE__ */
# ifdef __EMISSION__
/* direct lighting */
kernel_path_surface_connect_light(kg, &sd, emission_sd, throughput, state, L);
# endif /* __EMISSION__ */
# ifdef __VOLUME__
}
# endif
/* compute direct lighting and next bounce */
if (!kernel_path_surface_bounce(kg, &sd, &throughput, state, &L->state, ray))
break;
}
# ifdef __SUBSURFACE__
/* Trace indirect subsurface rays by restarting the loop. this uses less
* stack memory than invoking kernel_path_indirect.
*/
if (ss_indirect.num_rays) {
kernel_path_subsurface_setup_indirect(kg, &ss_indirect, state, ray, L, &throughput);
}
else {
break;
}
}
# endif /* __SUBSURFACE__ */
}
ccl_device void kernel_path_trace(
KernelGlobals *kg, ccl_global float *buffer, int sample, int x, int y, int offset, int stride)
{
PROFILING_INIT(kg, PROFILING_RAY_SETUP);
/* buffer offset */
int index = offset + x + y * stride;
int pass_stride = kernel_data.film.pass_stride;
buffer += index * pass_stride;
if (kernel_data.film.pass_adaptive_aux_buffer) {
ccl_global float4 *aux = (ccl_global float4 *)(buffer +
kernel_data.film.pass_adaptive_aux_buffer);
if (aux->w > 0.0f) {
return;
}
}
/* Initialize random numbers and sample ray. */
uint rng_hash;
Ray ray;
kernel_path_trace_setup(kg, sample, x, y, &rng_hash, &ray);
# ifndef __KERNEL_OPTIX__
if (ray.t == 0.0f) {
return;
}
# endif
/* Initialize state. */
float3 throughput = make_float3(1.0f, 1.0f, 1.0f);
PathRadiance L;
path_radiance_init(kg, &L);
ShaderDataTinyStorage emission_sd_storage;
ShaderData *emission_sd = AS_SHADER_DATA(&emission_sd_storage);
PathState state;
path_state_init(kg, emission_sd, &state, rng_hash, sample, &ray);
# ifdef __KERNEL_OPTIX__
/* Force struct into local memory to avoid costly spilling on trace calls. */
if (pass_stride < 0) /* This is never executed and just prevents the compiler from doing SROA. */
for (int i = 0; i < sizeof(L); ++i)
reinterpret_cast<unsigned char *>(&L)[-pass_stride + i] = 0;
# endif
/* Integrate. */
kernel_path_integrate(kg, &state, throughput, &ray, &L, buffer, emission_sd);
kernel_write_result(kg, buffer, sample, &L);
}
#endif /* __SPLIT_KERNEL__ */
CCL_NAMESPACE_END
|