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
|
/* SPDX-License-Identifier: Apache-2.0
* Copyright 2011-2022 Blender Foundation */
#include "device/device.h"
#include "integrator/pass_accessor_cpu.h"
#include "session/buffers.h"
#include "util/log.h"
#include "util/tbb.h"
// clang-format off
#include "kernel/device/cpu/compat.h"
#include "kernel/device/cpu/globals.h"
#include "kernel/types.h"
#include "kernel/film/read.h"
// clang-format on
CCL_NAMESPACE_BEGIN
/* --------------------------------------------------------------------
* Kernel processing.
*/
inline void PassAccessorCPU::run_get_pass_kernel_processor_float(
const KernelFilmConvert *kfilm_convert,
const RenderBuffers *render_buffers,
const BufferParams &buffer_params,
const Destination &destination,
const CPUKernels::FilmConvertFunction func) const
{
/* NOTE: No overlays are applied since they are not used for final renders.
* Can be supported via some sort of specialization to avoid code duplication. */
DCHECK_EQ(destination.stride, 0) << "Custom stride for float destination is not implemented.";
const int64_t pass_stride = buffer_params.pass_stride;
const int64_t buffer_row_stride = buffer_params.stride * buffer_params.pass_stride;
const float *window_data = render_buffers->buffer.data() + buffer_params.window_x * pass_stride +
buffer_params.window_y * buffer_row_stride;
const int pixel_stride = destination.pixel_stride ? destination.pixel_stride :
destination.num_components;
tbb::parallel_for(0, buffer_params.window_height, [&](int64_t y) {
const float *buffer = window_data + y * buffer_row_stride;
float *pixel = destination.pixels +
(y * buffer_params.width + destination.offset) * pixel_stride;
func(kfilm_convert, buffer, pixel, buffer_params.window_width, pass_stride, pixel_stride);
});
}
inline void PassAccessorCPU::run_get_pass_kernel_processor_half_rgba(
const KernelFilmConvert *kfilm_convert,
const RenderBuffers *render_buffers,
const BufferParams &buffer_params,
const Destination &destination,
const CPUKernels::FilmConvertHalfRGBAFunction func) const
{
const int64_t pass_stride = buffer_params.pass_stride;
const int64_t buffer_row_stride = buffer_params.stride * buffer_params.pass_stride;
const float *window_data = render_buffers->buffer.data() + buffer_params.window_x * pass_stride +
buffer_params.window_y * buffer_row_stride;
half4 *dst_start = destination.pixels_half_rgba + destination.offset;
const int destination_stride = destination.stride != 0 ? destination.stride :
buffer_params.width;
tbb::parallel_for(0, buffer_params.window_height, [&](int64_t y) {
const float *buffer = window_data + y * buffer_row_stride;
half4 *pixel = dst_start + y * destination_stride;
func(kfilm_convert, buffer, pixel, buffer_params.window_width, pass_stride);
});
}
/* --------------------------------------------------------------------
* Pass accessors.
*/
#define DEFINE_PASS_ACCESSOR(pass) \
void PassAccessorCPU::get_pass_##pass(const RenderBuffers *render_buffers, \
const BufferParams &buffer_params, \
const Destination &destination) const \
{ \
const CPUKernels &kernels = Device::get_cpu_kernels(); \
KernelFilmConvert kfilm_convert; \
init_kernel_film_convert(&kfilm_convert, buffer_params, destination); \
\
if (destination.pixels) { \
run_get_pass_kernel_processor_float(&kfilm_convert, \
render_buffers, \
buffer_params, \
destination, \
kernels.film_convert_##pass); \
} \
\
if (destination.pixels_half_rgba) { \
run_get_pass_kernel_processor_half_rgba(&kfilm_convert, \
render_buffers, \
buffer_params, \
destination, \
kernels.film_convert_half_rgba_##pass); \
} \
}
/* Float (scalar) passes. */
DEFINE_PASS_ACCESSOR(depth)
DEFINE_PASS_ACCESSOR(mist)
DEFINE_PASS_ACCESSOR(sample_count)
DEFINE_PASS_ACCESSOR(float)
/* Float3 passes. */
DEFINE_PASS_ACCESSOR(light_path)
DEFINE_PASS_ACCESSOR(shadow_catcher)
DEFINE_PASS_ACCESSOR(float3)
/* Float4 passes. */
DEFINE_PASS_ACCESSOR(motion)
DEFINE_PASS_ACCESSOR(cryptomatte)
DEFINE_PASS_ACCESSOR(shadow_catcher_matte_with_shadow)
DEFINE_PASS_ACCESSOR(combined)
DEFINE_PASS_ACCESSOR(float4)
#undef DEFINE_PASS_ACCESSOR
CCL_NAMESPACE_END
|
