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
|
/*
* Copyright 2019, NVIDIA Corporation.
* Copyright 2019, 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
#define OPTIX_DONT_INCLUDE_CUDA
#include <optix.h>
#define __KERNEL_GPU__
#define __KERNEL_CUDA__ /* OptiX kernels are implicitly CUDA kernels too */
#define __KERNEL_OPTIX__
#define CCL_NAMESPACE_BEGIN
#define CCL_NAMESPACE_END
#ifndef ATTR_FALLTHROUGH
# define ATTR_FALLTHROUGH
#endif
/* Manual definitions so we can compile without CUDA toolkit. */
#ifdef __CUDACC_RTC__
typedef unsigned int uint32_t;
typedef unsigned long long uint64_t;
#else
# include <stdint.h>
#endif
#ifdef CYCLES_CUBIN_CC
# define FLT_MIN 1.175494350822287507969e-38f
# define FLT_MAX 340282346638528859811704183484516925440.0f
# define FLT_EPSILON 1.192092896e-07F
#endif
#define ccl_device \
__device__ __forceinline__ // Function calls are bad for OptiX performance, so inline everything
#define ccl_device_inline ccl_device
#define ccl_device_forceinline ccl_device
#define ccl_device_noinline __device__ __noinline__
#define ccl_device_noinline_cpu ccl_device
#define ccl_global
#define ccl_static_constant __constant__
#define ccl_device_constant __constant__ __device__
#define ccl_constant const
#define ccl_gpu_shared __shared__
#define ccl_private
#define ccl_may_alias
#define ccl_addr_space
#define ccl_restrict __restrict__
#define ccl_loop_no_unroll
#define ccl_align(n) __align__(n)
/* Zero initialize structs to help the compiler figure out scoping */
#define ccl_optional_struct_init = {}
/* No assert supported for CUDA */
#define kernel_assert(cond)
/* GPU thread, block, grid size and index */
#define ccl_gpu_thread_idx_x (threadIdx.x)
#define ccl_gpu_block_dim_x (blockDim.x)
#define ccl_gpu_block_idx_x (blockIdx.x)
#define ccl_gpu_grid_dim_x (gridDim.x)
#define ccl_gpu_warp_size (warpSize)
#define ccl_gpu_global_id_x() (ccl_gpu_block_idx_x * ccl_gpu_block_dim_x + ccl_gpu_thread_idx_x)
#define ccl_gpu_global_size_x() (ccl_gpu_grid_dim_x * ccl_gpu_block_dim_x)
/* GPU warp synchronizaton */
#define ccl_gpu_syncthreads() __syncthreads()
#define ccl_gpu_ballot(predicate) __ballot_sync(0xFFFFFFFF, predicate)
#define ccl_gpu_shfl_down_sync(mask, var, detla) __shfl_down_sync(mask, var, detla)
#define ccl_gpu_popc(x) __popc(x)
/* GPU texture objects */
typedef unsigned long long CUtexObject;
typedef CUtexObject ccl_gpu_tex_object;
template<typename T>
ccl_device_forceinline T ccl_gpu_tex_object_read_2D(const ccl_gpu_tex_object texobj,
const float x,
const float y)
{
return tex2D<T>(texobj, x, y);
}
template<typename T>
ccl_device_forceinline T ccl_gpu_tex_object_read_3D(const ccl_gpu_tex_object texobj,
const float x,
const float y,
const float z)
{
return tex3D<T>(texobj, x, y, z);
}
/* Half */
typedef unsigned short half;
__device__ half __float2half(const float f)
{
half val;
asm("{ cvt.rn.f16.f32 %0, %1;}\n" : "=h"(val) : "f"(f));
return val;
}
/* Types */
#include "util/util_half.h"
#include "util/util_types.h"
|