API tools faq
paste
Advertisement
Guest User

Untitled

a guest
Feb 23rd, 2022
113
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 315.91 KB | None | 0 0
raw download clone embed print report
  1. PS C:\ngp\instant-ngp> cmake . -B build
  2. -- Building for: Visual Studio 16 2019
  3. -- Selecting Windows SDK version 10.0.19041.0 to target Windows 10.0.19044.
  4. -- The C compiler identification is MSVC 19.29.30140.0
  5. -- The CXX compiler identification is MSVC 19.29.30140.0
  6. -- The CUDA compiler identification is NVIDIA 11.6.55
  7. -- Detecting C compiler ABI info
  8. -- Detecting C compiler ABI info - done
  9. -- Check for working C compiler: C:/Program Files (x86)/Microsoft Visual Studio/2019/Community/VC/Tools/MSVC/14.29.30133/bin/Hostx64/x64/cl.exe - skipped
  10. -- Detecting C compile features
  11. -- Detecting C compile features - done
  12. -- Detecting CXX compiler ABI info
  13. -- Detecting CXX compiler ABI info - done
  14. -- Check for working CXX compiler: C:/Program Files (x86)/Microsoft Visual Studio/2019/Community/VC/Tools/MSVC/14.29.30133/bin/Hostx64/x64/cl.exe - skipped
  15. -- Detecting CXX compile features
  16. -- Detecting CXX compile features - done
  17. -- Detecting CUDA compiler ABI info
  18. -- Check for working CUDA compiler: C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v11.6/bin/nvcc.exe - skipped
  19. -- Detecting CUDA compile features
  20. -- Detecting CUDA compile features - done
  21. -- Looking for pthread.h
  22. -- Looking for pthread.h - not found
  23. -- Found Threads: TRUE
  24. -- Using Win32 for window creation
  25. -- Found OpenMP_C: -openmp (found version "2.0")
  26. -- Found OpenMP_CXX: -openmp (found version "2.0")
  27. -- Found OpenMP: TRUE (found version "2.0")
  28. -- OptiX_INSTALL_DIR value: C:\ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0
  29. -- Found Python: C:/Users/alan/AppData/Local/Programs/Python/Python39/python.exe (found suitable version "3.9.10", minimum required is "3.7") found components: Interpreter Development Development.Module Development.Embed
  30. -- pybind11 v2.7.1
  31. CMake Warning (dev) at C:/Program Files/CMake/share/cmake-3.23/Modules/CMakeDependentOption.cmake:84 (message):
  32. Policy CMP0127 is not set: cmake_dependent_option() supports full Condition
  33. Syntax. Run "cmake --help-policy CMP0127" for policy details. Use the
  34. cmake_policy command to set the policy and suppress this warning.
  35. Call Stack (most recent call first):
  36. dependencies/pybind11/CMakeLists.txt:98 (cmake_dependent_option)
  37. This warning is for project developers. Use -Wno-dev to suppress it.
  38.  
  39. -- Performing Test HAS_MSVC_GL_LTCG
  40. -- Performing Test HAS_MSVC_GL_LTCG - Success
  41. -- Targeting GPU architectures: 75
  42. -- Configuring done
  43. -- Generating done
  44. -- Build files have been written to: C:/ngp/instant-ngp/build
  45. PS C:\ngp\instant-ngp> cmake --build build --config RelWithDebInfo -j 16
  46. Microsoft (R) Build Engine version 16.11.2+f32259642 for .NET Framework
  47. Copyright (C) Microsoft Corporation. All rights reserved.
  48.  
  49. Checking Build System
  50. Building Custom Rule C:/ngp/instant-ngp/dependencies/glfw/src/CMakeLists.txt
  51. Building Custom Rule C:/ngp/instant-ngp/CMakeLists.txt
  52. context.c
  53. init.c
  54. input.c
  55. monitor.c
  56. vulkan.c
  57. window.c
  58. win32_init.c
  59. win32_joystick.c
  60. win32_monitor.c
  61. win32_time.c
  62. win32_thread.c
  63. win32_window.c
  64. wgl_context.c
  65. egl_context.c
  66. osmesa_context.c
  67. Generating Code...
  68. glfw_objects.vcxproj -> C:\ngp\instant-ngp\build\dependencies\glfw\src\glfw_objects.dir\RelWithDebInfo\glfw_objects.l
  69. ib
  70. Compiling CUDA source file ..\src\optix\raytrace.cu...
  71. Compiling CUDA source file ..\src\optix\raystab.cu...
  72. Compiling CUDA source file ..\src\optix\pathescape.cu...
  73.  
  74. C:\ngp\instant-ngp\build>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -gencode=arch=comput
  75. e_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\
  76. VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies" -I"C:\ngp\instant-ngp\dependen
  77. cies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\dependencies\glfw\include" -I"C:\ngp
  78. \instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:\ProgramData\NVIDIA Corporatio
  79. n\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"C:\ngp\instant-ngp\dependencie
  80. s\tiny-cuda-nn\dependencies" -I"C:\ngp\instant-ngp\dependencies\tinylogger" -I"C:\ngp\instant-ngp\include" -I"C:\ngp\
  81. instant-ngp\build" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" --keep-dir x64\RelWithDeb
  82. Info -maxrregcount=0 --machine 64 -ptx -cudart static --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob
  83. 1" -D_WINDOWS -DNDEBUG -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -DNGP_GUI -DNGP_OPTIX -D"NGP_VERSION=\"1.0dev\"" -D"CMAK
  84. E_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -o optix_program.dir\RelWithDebInfo\raytrace
  85. .ptx "C:\ngp\instant-ngp\src\optix\raytrace.cu"
  86.  
  87. C:\ngp\instant-ngp\build>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -gencode=arch=comput
  88. e_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\
  89. VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies" -I"C:\ngp\instant-ngp\dependen
  90. cies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\dependencies\glfw\include" -I"C:\ngp
  91. \instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:\ProgramData\NVIDIA Corporatio
  92. n\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"C:\ngp\instant-ngp\dependencie
  93. s\tiny-cuda-nn\dependencies" -I"C:\ngp\instant-ngp\dependencies\tinylogger" -I"C:\ngp\instant-ngp\include" -I"C:\ngp\
  94. instant-ngp\build" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" --keep-dir x64\RelWithDeb
  95. Info -maxrregcount=0 --machine 64 -ptx -cudart static --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob
  96. 1" -D_WINDOWS -DNDEBUG -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -DNGP_GUI -DNGP_OPTIX -D"NGP_VERSION=\"1.0dev\"" -D"CMAK
  97. E_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -o optix_program.dir\RelWithDebInfo\raystab.
  98. ptx "C:\ngp\instant-ngp\src\optix\raystab.cu"
  99.  
  100. C:\ngp\instant-ngp\build>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -gencode=arch=comput
  101. e_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\
  102. VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies" -I"C:\ngp\instant-ngp\dependen
  103. cies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\dependencies\glfw\include" -I"C:\ngp
  104. \instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:\ProgramData\NVIDIA Corporatio
  105. n\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"C:\ngp\instant-ngp\dependencie
  106. s\tiny-cuda-nn\dependencies" -I"C:\ngp\instant-ngp\dependencies\tinylogger" -I"C:\ngp\instant-ngp\include" -I"C:\ngp\
  107. instant-ngp\build" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" --keep-dir x64\RelWithDeb
  108. Info -maxrregcount=0 --machine 64 -ptx -cudart static --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob
  109. 1" -D_WINDOWS -DNDEBUG -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -DNGP_GUI -DNGP_OPTIX -D"NGP_VERSION=\"1.0dev\"" -D"CMAK
  110. E_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -o optix_program.dir\RelWithDebInfo\pathesca
  111. pe.ptx "C:\ngp\instant-ngp\src\optix\pathescape.cu"
  112. raystab.cu
  113. raytrace.cu
  114. pathescape.cu
  115. Building Custom Rule C:/ngp/instant-ngp/dependencies/tiny-cuda-nn/src/CMakeLists.txt
  116. Compiling CUDA source file ..\..\..\..\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu...
  117. Compiling CUDA source file ..\..\..\..\dependencies\tiny-cuda-nn\src\common_device.cu...
  118. Compiling CUDA source file ..\..\..\..\dependencies\tiny-cuda-nn\src\cpp_api.cu...
  119. Compiling CUDA source file ..\..\..\..\dependencies\tiny-cuda-nn\src\common.cu...
  120.  
  121. C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\
  122. nvcc.exe" -gencode=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft
  123. Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies"
  124. -I"C:\ngp\instant-ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\depend
  125. encies\glfw\include" -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:
  126. \ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"
  127. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.
  128. 6\include" --keep-dir x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambd
  129. a --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX
  130. -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"
  131. " -Xcompiler "/EHsc /W1 /nologo /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cud
  132. a-nn.pdb /FS /Zi /MD /GR" -o tiny-cuda-nn.dir\RelWithDebInfo\cutlass_mlp.obj "C:\ngp\instant-ngp\dependencies\tiny-c
  133. uda-nn\src\cutlass_mlp.cu"
  134.  
  135. C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\
  136. nvcc.exe" -gencode=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft
  137. Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies"
  138. -I"C:\ngp\instant-ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\depend
  139. encies\glfw\include" -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:
  140. \ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"
  141. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.
  142. 6\include" --keep-dir x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambd
  143. a --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX
  144. -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"
  145. " -Xcompiler "/EHsc /W1 /nologo /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cud
  146. a-nn.pdb /FS /Zi /MD /GR" -o tiny-cuda-nn.dir\RelWithDebInfo\cpp_api.obj "C:\ngp\instant-ngp\dependencies\tiny-cuda-
  147. nn\src\cpp_api.cu"
  148.  
  149. C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\
  150. nvcc.exe" -gencode=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft
  151. Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies"
  152. -I"C:\ngp\instant-ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\depend
  153. encies\glfw\include" -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:
  154. \ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"
  155. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.
  156. 6\include" --keep-dir x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambd
  157. a --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX
  158. -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"
  159. " -Xcompiler "/EHsc /W1 /nologo /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cud
  160. a-nn.pdb /FS /Zi /MD /GR" -o tiny-cuda-nn.dir\RelWithDebInfo\common_device.obj "C:\ngp\instant-ngp\dependencies\tiny
  161. -cuda-nn\src\common_device.cu"
  162.  
  163. C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\
  164. nvcc.exe" -gencode=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft
  165. Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies"
  166. -I"C:\ngp\instant-ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\depend
  167. encies\glfw\include" -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:
  168. \ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"
  169. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.
  170. 6\include" --keep-dir x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambd
  171. a --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX
  172. -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"
  173. " -Xcompiler "/EHsc /W1 /nologo /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cud
  174. a-nn.pdb /FS /Zi /MD /GR" -o tiny-cuda-nn.dir\RelWithDebInfo\common.obj "C:\ngp\instant-ngp\dependencies\tiny-cuda-n
  175. n\src\common.cu"
  176. Compiling CUDA source file ..\..\..\..\dependencies\tiny-cuda-nn\src\object.cu...
  177. Compiling CUDA source file ..\..\..\..\dependencies\tiny-cuda-nn\src\optimizer.cu...
  178. Compiling CUDA source file ..\..\..\..\dependencies\tiny-cuda-nn\src\network.cu...
  179. Compiling CUDA source file ..\..\..\..\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu...
  180. Compiling CUDA source file ..\..\..\..\dependencies\tiny-cuda-nn\src\encoding.cu...
  181. Compiling CUDA source file ..\..\..\..\dependencies\tiny-cuda-nn\src\reduce_sum.cu...
  182. Compiling CUDA source file ..\..\..\..\dependencies\tiny-cuda-nn\src\loss.cu...
  183. Compiling CUDA source file ..\..\..\..\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu...
  184.  
  185. C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\
  186. nvcc.exe" -gencode=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft
  187. Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies"
  188. -I"C:\ngp\instant-ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\depend
  189. encies\glfw\include" -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:
  190. \ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"
  191. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.
  192. 6\include" --keep-dir x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambd
  193. a --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX
  194. -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"
  195. " -Xcompiler "/EHsc /W1 /nologo /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cud
  196. a-nn.pdb /FS /Zi /MD /GR" -o tiny-cuda-nn.dir\RelWithDebInfo\optimizer.obj "C:\ngp\instant-ngp\dependencies\tiny-cud
  197. a-nn\src\optimizer.cu"
  198.  
  199. C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\
  200. nvcc.exe" -gencode=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft
  201. Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies"
  202. -I"C:\ngp\instant-ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\depend
  203. encies\glfw\include" -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:
  204. \ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"
  205. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.
  206. 6\include" --keep-dir x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambd
  207. a --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX
  208. -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"
  209. " -Xcompiler "/EHsc /W1 /nologo /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cud
  210. a-nn.pdb /FS /Zi /MD /GR" -o tiny-cuda-nn.dir\RelWithDebInfo\loss.obj "C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\
  211. src\loss.cu"
  212.  
  213. C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\
  214. nvcc.exe" -gencode=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft
  215. Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies"
  216. -I"C:\ngp\instant-ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\depend
  217. encies\glfw\include" -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:
  218. \ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"
  219. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.
  220. 6\include" --keep-dir x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambd
  221. a --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX
  222. -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"
  223. " -Xcompiler "/EHsc /W1 /nologo /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cud
  224. a-nn.pdb /FS /Zi /MD /GR" -o tiny-cuda-nn.dir\RelWithDebInfo\object.obj "C:\ngp\instant-ngp\dependencies\tiny-cuda-n
  225. n\src\object.cu"
  226.  
  227. C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\
  228. nvcc.exe" -gencode=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft
  229. Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies"
  230. -I"C:\ngp\instant-ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\depend
  231. encies\glfw\include" -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:
  232. \ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"
  233. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.
  234. 6\include" --keep-dir x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambd
  235. a --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX
  236. -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"
  237. " -Xcompiler "/EHsc /W1 /nologo /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cud
  238. a-nn.pdb /FS /Zi /MD /GR" -o tiny-cuda-nn.dir\RelWithDebInfo\encoding.obj "C:\ngp\instant-ngp\dependencies\tiny-cuda
  239. -nn\src\encoding.cu"
  240.  
  241.  
  242.  
  243. C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\
  244. nvcc.exe" -gencode=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft
  245. Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies"
  246. -I"C:\ngp\instant-ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\depend
  247. encies\glfw\include" -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:
  248. \ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"
  249. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.
  250. 6\include" --keep-dir x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambd
  251. a --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX
  252. -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"
  253. " -Xcompiler "/EHsc /W1 /nologo /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cud
  254. a-nn.pdb /FS /Zi /MD /GR" -o tiny-cuda-nn.dir\RelWithDebInfo\cutlass_resnet.obj "C:\ngp\instant-ngp\dependencies\tin
  255. y-cuda-nn\src\cutlass_resnet.cu"
  256. C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\
  257. nvcc.exe" -gencode=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft
  258. Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies"
  259. -I"C:\ngp\instant-ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\depend
  260. encies\glfw\include" -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:
  261. \ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"
  262. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.
  263. 6\include" --keep-dir x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambd
  264. a --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX
  265. -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"
  266. " -Xcompiler "/EHsc /W1 /nologo /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cud
  267. a-nn.pdb /FS /Zi /MD /GR" -o tiny-cuda-nn.dir\RelWithDebInfo\network.obj "C:\ngp\instant-ngp\dependencies\tiny-cuda-
  268. nn\src\network.cu"
  269. C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\
  270. nvcc.exe" -gencode=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft
  271. Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies"
  272. -I"C:\ngp\instant-ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\depend
  273. encies\glfw\include" -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:
  274. \ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"
  275. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.
  276. 6\include" --keep-dir x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambd
  277. a --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX
  278. -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"
  279. " -Xcompiler "/EHsc /W1 /nologo /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cud
  280. a-nn.pdb /FS /Zi /MD /GR" -o tiny-cuda-nn.dir\RelWithDebInfo\reduce_sum.obj "C:\ngp\instant-ngp\dependencies\tiny-cu
  281. da-nn\src\reduce_sum.cu"
  282.  
  283. C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src>"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\
  284. nvcc.exe" -gencode=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft
  285. Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies"
  286. -I"C:\ngp\instant-ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\depend
  287. encies\glfw\include" -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:
  288. \ProgramData\NVIDIA Corporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"
  289. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.
  290. 6\include" --keep-dir x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambd
  291. a --expt-relaxed-constexpr -std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX
  292. -DTCNN_MIN_GPU_ARCH=75 -DTCNN_SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"
  293. " -Xcompiler "/EHsc /W1 /nologo /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cud
  294. a-nn.pdb /FS /Zi /MD /GR" -o tiny-cuda-nn.dir\RelWithDebInfo\fully_fused_mlp.obj "C:\ngp\instant-ngp\dependencies\ti
  295. ny-cuda-nn\src\fully_fused_mlp.cu"
  296. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(74): error : more than one conversion
  297. function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cuda
  298. -nn\src\tiny-cuda-nn.vcxproj]
  299. function "__half::operator float() const"
  300. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  301. function "__half::operator short() const"
  302. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  303. function "__half::operator unsigned short() const"
  304. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  305. function "__half::operator int() const"
  306. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  307. function "__half::operator unsigned int() const"
  308. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  309. function "__half::operator long long() const"
  310. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  311. function "__half::operator unsigned long long() const"
  312. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  313. function "__half::operator __nv_bool() const"
  314. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  315. detected during:
  316. instantiation of "void tcnn::warp_activation<T,fragment_t>(tcnn::Activation, const fragment_t &, fragment
  317. _t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U
  318. >>]"
  319. (244): here
  320. instantiation of "void tcnn::kernel_activation(uint32_t, tcnn::Activation, const T *, T *) [with T=tcnn::
  321. network_precision_t, N=8U]"
  322. (286): here
  323. instantiation of "void tcnn::activation_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *, T *) [wi
  324. th T=tcnn::network_precision_t]"
  325. (295): here
  326. instantiation of "void tcnn::activation_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrixDynamic<
  327. T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  328. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(149): here
  329. instantiation of "__nv_bool tcnn::compute_layer<CutlassLayer,T>(cudaStream_t, __nv_bool, tcnn::Activation
  330. , const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic
  331. <T> &, tcnn::GPUMatrixDynamic<T> &) [with CutlassLayer=tcnn::LastLayer, T=tcnn::network_precision_t]"
  332. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(163): here
  333. instantiation of "__nv_bool tcnn::compute_inference_layer<CutlassLayer,T>(cudaStream_t, tcnn::Activation,
  334. const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<
  335. T> &) [with CutlassLayer=tcnn::LastLayer, T=tcnn::network_precision_t]"
  336. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(183): here
  337. instantiation of "void tcnn::CutlassMLP<T>::inference_mixed_precision(cudaStream_t, const tcnn::GPUMatrix
  338. Dynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t]"
  339. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(117): here
  340.  
  341. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(74): error : more than one conversion
  342. function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cuda
  343. -nn\src\tiny-cuda-nn.vcxproj]
  344. function "__half::operator float() const"
  345. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  346. function "__half::operator short() const"
  347. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  348. function "__half::operator unsigned short() const"
  349. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  350. function "__half::operator int() const"
  351. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  352. function "__half::operator unsigned int() const"
  353. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  354. function "__half::operator long long() const"
  355. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  356. function "__half::operator unsigned long long() const"
  357. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  358. function "__half::operator __nv_bool() const"
  359. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  360. detected during:
  361. instantiation of "void tcnn::warp_activation<T,fragment_t>(tcnn::Activation, const fragment_t &, fragment
  362. _t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U
  363. >>]"
  364. (244): here
  365. instantiation of "void tcnn::kernel_activation(uint32_t, tcnn::Activation, const T *, T *) [with T=tcnn::
  366. network_precision_t, N=8U]"
  367. (286): here
  368. instantiation of "void tcnn::activation_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *, T *) [wi
  369. th T=tcnn::network_precision_t]"
  370. (295): here
  371. instantiation of "void tcnn::activation_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrixDynamic<
  372. T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  373. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(149): here
  374. instantiation of "__nv_bool tcnn::compute_layer<CutlassLayer,T>(cudaStream_t, __nv_bool, tcnn::Activation
  375. , const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic
  376. <T> &, tcnn::GPUMatrixDynamic<T> &) [with CutlassLayer=tcnn::LastLayer, T=tcnn::network_precision_t]"
  377. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(163): here
  378. instantiation of "__nv_bool tcnn::compute_inference_layer<CutlassLayer,T>(cudaStream_t, tcnn::Activation,
  379. const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<
  380. T> &) [with CutlassLayer=tcnn::LastLayer, T=tcnn::network_precision_t]"
  381. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(183): here
  382. instantiation of "void tcnn::CutlassMLP<T>::inference_mixed_precision(cudaStream_t, const tcnn::GPUMatrix
  383. Dynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t]"
  384. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(117): here
  385.  
  386. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(187): error : more than one conversio
  387. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  388. a-nn\src\tiny-cuda-nn.vcxproj]
  389. function "__half::operator float() const"
  390. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  391. function "__half::operator short() const"
  392. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  393. function "__half::operator unsigned short() const"
  394. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  395. function "__half::operator int() const"
  396. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  397. function "__half::operator unsigned int() const"
  398. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  399. function "__half::operator long long() const"
  400. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  401. function "__half::operator unsigned long long() const"
  402. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  403. function "__half::operator __nv_bool() const"
  404. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  405. detected during:
  406. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  407. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  408. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  409. nn::network_precision_t, 8U>>]"
  410. (269): here
  411. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  412. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  413. (334): here
  414. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  415. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  416. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(300): here
  417. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  418. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  419. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  420. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  421.  
  422. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(187): error : more than one conversio
  423. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  424. a-nn\src\tiny-cuda-nn.vcxproj]
  425. function "__half::operator float() const"
  426. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  427. function "__half::operator short() const"
  428. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  429. function "__half::operator unsigned short() const"
  430. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  431. function "__half::operator int() const"
  432. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  433. function "__half::operator unsigned int() const"
  434. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  435. function "__half::operator long long() const"
  436. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  437. function "__half::operator unsigned long long() const"
  438. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  439. function "__half::operator __nv_bool() const"
  440. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  441. detected during:
  442. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  443. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  444. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  445. nn::network_precision_t, 8U>>]"
  446. (269): here
  447. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  448. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  449. (334): here
  450. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  451. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  452. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(300): here
  453. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  454. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  455. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  456. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  457.  
  458. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(193): error : more than one conversio
  459. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  460. a-nn\src\tiny-cuda-nn.vcxproj]
  461. function "__half::operator float() const"
  462. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  463. function "__half::operator short() const"
  464. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  465. function "__half::operator unsigned short() const"
  466. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  467. function "__half::operator int() const"
  468. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  469. function "__half::operator unsigned int() const"
  470. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  471. function "__half::operator long long() const"
  472. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  473. function "__half::operator unsigned long long() const"
  474. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  475. function "__half::operator __nv_bool() const"
  476. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  477. detected during:
  478. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  479. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  480. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  481. nn::network_precision_t, 8U>>]"
  482. (269): here
  483. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  484. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  485. (334): here
  486. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  487. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  488. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(300): here
  489. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  490. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  491. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  492. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  493.  
  494. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(193): error : more than one conversio
  495. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  496. a-nn\src\tiny-cuda-nn.vcxproj]
  497. function "__half::operator float() const"
  498. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  499. function "__half::operator short() const"
  500. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  501. function "__half::operator unsigned short() const"
  502. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  503. function "__half::operator int() const"
  504. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  505. function "__half::operator unsigned int() const"
  506. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  507. function "__half::operator long long() const"
  508. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  509. function "__half::operator unsigned long long() const"
  510. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  511. function "__half::operator __nv_bool() const"
  512. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  513. detected during:
  514. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  515. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  516. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  517. nn::network_precision_t, 8U>>]"
  518. (269): here
  519. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  520. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  521. (334): here
  522. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  523. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  524. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(300): here
  525. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  526. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  527. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  528. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  529.  
  530. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(204): error : more than one conversio
  531. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  532. a-nn\src\tiny-cuda-nn.vcxproj]
  533. function "__half::operator float() const"
  534. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  535. function "__half::operator short() const"
  536. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  537. function "__half::operator unsigned short() const"
  538. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  539. function "__half::operator int() const"
  540. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  541. function "__half::operator unsigned int() const"
  542. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  543. function "__half::operator long long() const"
  544. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  545. function "__half::operator unsigned long long() const"
  546. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  547. function "__half::operator __nv_bool() const"
  548. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  549. detected during:
  550. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  551. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  552. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  553. nn::network_precision_t, 8U>>]"
  554. (269): here
  555. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  556. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  557. (334): here
  558. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  559. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  560. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(300): here
  561. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  562. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  563. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  564. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  565.  
  566. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(204): error : more than one conversio
  567. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  568. a-nn\src\tiny-cuda-nn.vcxproj]
  569. function "__half::operator float() const"
  570. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  571. function "__half::operator short() const"
  572. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  573. function "__half::operator unsigned short() const"
  574. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  575. function "__half::operator int() const"
  576. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  577. function "__half::operator unsigned int() const"
  578. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  579. function "__half::operator long long() const"
  580. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  581. function "__half::operator unsigned long long() const"
  582. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  583. function "__half::operator __nv_bool() const"
  584. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  585. detected during:
  586. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  587. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  588. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  589. nn::network_precision_t, 8U>>]"
  590. (269): here
  591. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  592. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  593. (334): here
  594. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  595. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  596. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(300): here
  597. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  598. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  599. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  600. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  601.  
  602. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(211): error : more than one conversio
  603. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  604. a-nn\src\tiny-cuda-nn.vcxproj]
  605. function "__half::operator float() const"
  606. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  607. function "__half::operator short() const"
  608. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  609. function "__half::operator unsigned short() const"
  610. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  611. function "__half::operator int() const"
  612. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  613. function "__half::operator unsigned int() const"
  614. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  615. function "__half::operator long long() const"
  616. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  617. function "__half::operator unsigned long long() const"
  618. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  619. function "__half::operator __nv_bool() const"
  620. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  621. detected during:
  622. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  623. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  624. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  625. nn::network_precision_t, 8U>>]"
  626. (269): here
  627. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  628. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  629. (334): here
  630. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  631. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  632. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(300): here
  633. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  634. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  635. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  636. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  637.  
  638. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(211): error : more than one conversio
  639. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  640. a-nn\src\tiny-cuda-nn.vcxproj]
  641. function "__half::operator float() const"
  642. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  643. function "__half::operator short() const"
  644. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  645. function "__half::operator unsigned short() const"
  646. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  647. function "__half::operator int() const"
  648. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  649. function "__half::operator unsigned int() const"
  650. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  651. function "__half::operator long long() const"
  652. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  653. function "__half::operator unsigned long long() const"
  654. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  655. function "__half::operator __nv_bool() const"
  656. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  657. detected during:
  658. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  659. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  660. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  661. nn::network_precision_t, 8U>>]"
  662. (269): here
  663. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  664. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  665. (334): here
  666. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  667. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  668. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(300): here
  669. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  670. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  671. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  672. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  673.  
  674. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(217): error : more than one conversio
  675. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  676. a-nn\src\tiny-cuda-nn.vcxproj]
  677. function "__half::operator float() const"
  678. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  679. function "__half::operator short() const"
  680. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  681. function "__half::operator unsigned short() const"
  682. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  683. function "__half::operator int() const"
  684. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  685. function "__half::operator unsigned int() const"
  686. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  687. function "__half::operator long long() const"
  688. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  689. function "__half::operator unsigned long long() const"
  690. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  691. function "__half::operator __nv_bool() const"
  692. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  693. detected during:
  694. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  695. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  696. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  697. nn::network_precision_t, 8U>>]"
  698. (269): here
  699. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  700. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  701. (334): here
  702. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  703. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  704. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(300): here
  705. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  706. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  707. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  708. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  709.  
  710. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(217): error : more than one conversio
  711. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  712. a-nn\src\tiny-cuda-nn.vcxproj]
  713. function "__half::operator float() const"
  714. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  715. function "__half::operator short() const"
  716. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  717. function "__half::operator unsigned short() const"
  718. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  719. function "__half::operator int() const"
  720. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  721. function "__half::operator unsigned int() const"
  722. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  723. function "__half::operator long long() const"
  724. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  725. function "__half::operator unsigned long long() const"
  726. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  727. function "__half::operator __nv_bool() const"
  728. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  729. detected during:
  730. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  731. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  732. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  733. nn::network_precision_t, 8U>>]"
  734. (269): here
  735. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  736. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  737. (334): here
  738. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  739. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  740. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(300): here
  741. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  742. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  743. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  744. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  745.  
  746. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(129): error : more than one conversio
  747. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  748. a-nn\src\tiny-cuda-nn.vcxproj]
  749. function "__half::operator float() const"
  750. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  751. function "__half::operator short() const"
  752. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  753. function "__half::operator unsigned short() const"
  754. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  755. function "__half::operator int() const"
  756. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  757. function "__half::operator unsigned int() const"
  758. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  759. function "__half::operator long long() const"
  760. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  761. function "__half::operator unsigned long long() const"
  762. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  763. function "__half::operator __nv_bool() const"
  764. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  765. detected during:
  766. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  767. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  768. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  769. <tcnn::network_precision_t, 8U>>]"
  770. (256): here
  771. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  772. T *) [with T=tcnn::network_precision_t, N=8U]"
  773. (310): here
  774. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  775. const T *, T *) [with T=tcnn::network_precision_t]"
  776. (319): here
  777. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  778. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  779. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(353): here
  780. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  781. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  782. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  783. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  784.  
  785. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(129): error : more than one conversio
  786. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  787. a-nn\src\tiny-cuda-nn.vcxproj]
  788. function "__half::operator float() const"
  789. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  790. function "__half::operator short() const"
  791. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  792. function "__half::operator unsigned short() const"
  793. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  794. function "__half::operator int() const"
  795. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  796. function "__half::operator unsigned int() const"
  797. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  798. function "__half::operator long long() const"
  799. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  800. function "__half::operator unsigned long long() const"
  801. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  802. function "__half::operator __nv_bool() const"
  803. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  804. detected during:
  805. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  806. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  807. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  808. <tcnn::network_precision_t, 8U>>]"
  809. (256): here
  810. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  811. T *) [with T=tcnn::network_precision_t, N=8U]"
  812. (310): here
  813. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  814. const T *, T *) [with T=tcnn::network_precision_t]"
  815. (319): here
  816. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  817. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  818. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(353): here
  819. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  820. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  821. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  822. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  823.  
  824. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(135): error : more than one conversio
  825. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  826. a-nn\src\tiny-cuda-nn.vcxproj]
  827. function "__half::operator float() const"
  828. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  829. function "__half::operator short() const"
  830. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  831. function "__half::operator unsigned short() const"
  832. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  833. function "__half::operator int() const"
  834. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  835. function "__half::operator unsigned int() const"
  836. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  837. function "__half::operator long long() const"
  838. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  839. function "__half::operator unsigned long long() const"
  840. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  841. function "__half::operator __nv_bool() const"
  842. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  843. detected during:
  844. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  845. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  846. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  847. <tcnn::network_precision_t, 8U>>]"
  848. (256): here
  849. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  850. T *) [with T=tcnn::network_precision_t, N=8U]"
  851. (310): here
  852. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  853. const T *, T *) [with T=tcnn::network_precision_t]"
  854. (319): here
  855. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  856. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  857. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(353): here
  858. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  859. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  860. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  861. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  862.  
  863. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(135): error : more than one conversio
  864. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  865. a-nn\src\tiny-cuda-nn.vcxproj]
  866. function "__half::operator float() const"
  867. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  868. function "__half::operator short() const"
  869. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  870. function "__half::operator unsigned short() const"
  871. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  872. function "__half::operator int() const"
  873. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  874. function "__half::operator unsigned int() const"
  875. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  876. function "__half::operator long long() const"
  877. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  878. function "__half::operator unsigned long long() const"
  879. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  880. function "__half::operator __nv_bool() const"
  881. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  882. detected during:
  883. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  884. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  885. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  886. <tcnn::network_precision_t, 8U>>]"
  887. (256): here
  888. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  889. T *) [with T=tcnn::network_precision_t, N=8U]"
  890. (310): here
  891. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  892. const T *, T *) [with T=tcnn::network_precision_t]"
  893. (319): here
  894. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  895. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  896. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(353): here
  897. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  898. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  899. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  900. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  901.  
  902. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(141): error : more than one conversio
  903. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  904. a-nn\src\tiny-cuda-nn.vcxproj]
  905. function "__half::operator float() const"
  906. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  907. function "__half::operator short() const"
  908. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  909. function "__half::operator unsigned short() const"
  910. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  911. function "__half::operator int() const"
  912. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  913. function "__half::operator unsigned int() const"
  914. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  915. function "__half::operator long long() const"
  916. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  917. function "__half::operator unsigned long long() const"
  918. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  919. function "__half::operator __nv_bool() const"
  920. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  921. detected during:
  922. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  923. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  924. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  925. <tcnn::network_precision_t, 8U>>]"
  926. (256): here
  927. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  928. T *) [with T=tcnn::network_precision_t, N=8U]"
  929. (310): here
  930. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  931. const T *, T *) [with T=tcnn::network_precision_t]"
  932. (319): here
  933. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  934. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  935. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(353): here
  936. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  937. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  938. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  939. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  940.  
  941. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(141): error : more than one conversio
  942. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  943. a-nn\src\tiny-cuda-nn.vcxproj]
  944. function "__half::operator float() const"
  945. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  946. function "__half::operator short() const"
  947. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  948. function "__half::operator unsigned short() const"
  949. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  950. function "__half::operator int() const"
  951. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  952. function "__half::operator unsigned int() const"
  953. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  954. function "__half::operator long long() const"
  955. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  956. function "__half::operator unsigned long long() const"
  957. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  958. function "__half::operator __nv_bool() const"
  959. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  960. detected during:
  961. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  962. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  963. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  964. <tcnn::network_precision_t, 8U>>]"
  965. (256): here
  966. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  967. T *) [with T=tcnn::network_precision_t, N=8U]"
  968. (310): here
  969. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  970. const T *, T *) [with T=tcnn::network_precision_t]"
  971. (319): here
  972. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  973. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  974. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(353): here
  975. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  976. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  977. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  978. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  979.  
  980. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(148): error : more than one conversio
  981. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  982. a-nn\src\tiny-cuda-nn.vcxproj]
  983. function "__half::operator float() const"
  984. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  985. function "__half::operator short() const"
  986. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  987. function "__half::operator unsigned short() const"
  988. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  989. function "__half::operator int() const"
  990. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  991. function "__half::operator unsigned int() const"
  992. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  993. function "__half::operator long long() const"
  994. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  995. function "__half::operator unsigned long long() const"
  996. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  997. function "__half::operator __nv_bool() const"
  998. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  999. detected during:
  1000. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  1001. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  1002. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  1003. <tcnn::network_precision_t, 8U>>]"
  1004. (256): here
  1005. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  1006. T *) [with T=tcnn::network_precision_t, N=8U]"
  1007. (310): here
  1008. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  1009. const T *, T *) [with T=tcnn::network_precision_t]"
  1010. (319): here
  1011. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  1012. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  1013. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(353): here
  1014. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  1015. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  1016. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  1017. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  1018.  
  1019. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(148): error : more than one conversio
  1020. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  1021. a-nn\src\tiny-cuda-nn.vcxproj]
  1022. function "__half::operator float() const"
  1023. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  1024. function "__half::operator short() const"
  1025. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  1026. function "__half::operator unsigned short() const"
  1027. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  1028. function "__half::operator int() const"
  1029. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  1030. function "__half::operator unsigned int() const"
  1031. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  1032. function "__half::operator long long() const"
  1033. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  1034. function "__half::operator unsigned long long() const"
  1035. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  1036. function "__half::operator __nv_bool() const"
  1037. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  1038. detected during:
  1039. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  1040. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  1041. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  1042. <tcnn::network_precision_t, 8U>>]"
  1043. (256): here
  1044. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  1045. T *) [with T=tcnn::network_precision_t, N=8U]"
  1046. (310): here
  1047. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  1048. const T *, T *) [with T=tcnn::network_precision_t]"
  1049. (319): here
  1050. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  1051. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  1052. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(353): here
  1053. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  1054. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  1055. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  1056. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  1057.  
  1058. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(156): error : more than one conversio
  1059. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  1060. a-nn\src\tiny-cuda-nn.vcxproj]
  1061. function "__half::operator float() const"
  1062. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  1063. function "__half::operator short() const"
  1064. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  1065. function "__half::operator unsigned short() const"
  1066. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  1067. function "__half::operator int() const"
  1068. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  1069. function "__half::operator unsigned int() const"
  1070. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  1071. function "__half::operator long long() const"
  1072. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  1073. function "__half::operator unsigned long long() const"
  1074. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  1075. function "__half::operator __nv_bool() const"
  1076. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  1077. detected during:
  1078. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  1079. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  1080. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  1081. <tcnn::network_precision_t, 8U>>]"
  1082. (256): here
  1083. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  1084. T *) [with T=tcnn::network_precision_t, N=8U]"
  1085. (310): here
  1086. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  1087. const T *, T *) [with T=tcnn::network_precision_t]"
  1088. (319): here
  1089. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  1090. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  1091. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(353): here
  1092. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  1093. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  1094. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  1095. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  1096.  
  1097. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(156): error : more than one conversio
  1098. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  1099. a-nn\src\tiny-cuda-nn.vcxproj]
  1100. function "__half::operator float() const"
  1101. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  1102. function "__half::operator short() const"
  1103. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  1104. function "__half::operator unsigned short() const"
  1105. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  1106. function "__half::operator int() const"
  1107. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  1108. function "__half::operator unsigned int() const"
  1109. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  1110. function "__half::operator long long() const"
  1111. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  1112. function "__half::operator unsigned long long() const"
  1113. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  1114. function "__half::operator __nv_bool() const"
  1115. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  1116. detected during:
  1117. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  1118. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  1119. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  1120. <tcnn::network_precision_t, 8U>>]"
  1121. (256): here
  1122. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  1123. T *) [with T=tcnn::network_precision_t, N=8U]"
  1124. (310): here
  1125. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  1126. const T *, T *) [with T=tcnn::network_precision_t]"
  1127. (319): here
  1128. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  1129. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  1130. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(353): here
  1131. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  1132. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  1133. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  1134. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  1135.  
  1136. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(163): error : more than one conversio
  1137. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  1138. a-nn\src\tiny-cuda-nn.vcxproj]
  1139. function "__half::operator float() const"
  1140. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  1141. function "__half::operator short() const"
  1142. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  1143. function "__half::operator unsigned short() const"
  1144. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  1145. function "__half::operator int() const"
  1146. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  1147. function "__half::operator unsigned int() const"
  1148. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  1149. function "__half::operator long long() const"
  1150. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  1151. function "__half::operator unsigned long long() const"
  1152. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  1153. function "__half::operator __nv_bool() const"
  1154. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  1155. detected during:
  1156. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  1157. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  1158. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  1159. <tcnn::network_precision_t, 8U>>]"
  1160. (256): here
  1161. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  1162. T *) [with T=tcnn::network_precision_t, N=8U]"
  1163. (310): here
  1164. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  1165. const T *, T *) [with T=tcnn::network_precision_t]"
  1166. (319): here
  1167. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  1168. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  1169. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(353): here
  1170. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  1171. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  1172. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  1173. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  1174.  
  1175. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(163): error : more than one conversio
  1176. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  1177. a-nn\src\tiny-cuda-nn.vcxproj]
  1178. function "__half::operator float() const"
  1179. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  1180. function "__half::operator short() const"
  1181. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  1182. function "__half::operator unsigned short() const"
  1183. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  1184. function "__half::operator int() const"
  1185. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  1186. function "__half::operator unsigned int() const"
  1187. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  1188. function "__half::operator long long() const"
  1189. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  1190. function "__half::operator unsigned long long() const"
  1191. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  1192. function "__half::operator __nv_bool() const"
  1193. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  1194. detected during:
  1195. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  1196. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  1197. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  1198. <tcnn::network_precision_t, 8U>>]"
  1199. (256): here
  1200. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  1201. T *) [with T=tcnn::network_precision_t, N=8U]"
  1202. (310): here
  1203. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  1204. const T *, T *) [with T=tcnn::network_precision_t]"
  1205. (319): here
  1206. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  1207. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  1208. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(353): here
  1209. instantiation of "void tcnn::CutlassMLP<T>::backward(cudaStream_t, const tcnn::Context &, const tcnn::GPU
  1210. MatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *
  1211. , __nv_bool, __nv_bool) [with T=tcnn::network_precision_t]"
  1212. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu(452): here
  1213.  
  1214. 24 errors detected in the compilation of "C:/ngp/instant-ngp/dependencies/tiny-cuda-nn/src/cutlass_mlp.cu".
  1215. cutlass_mlp.cu
  1216. C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\BuildCustomizations\CUDA 11.6.t
  1217. argets(790,9): error MSB3721: The command ""C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -gen
  1218. code=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft Visual Studio\2
  1219. 019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies" -I"C:\ngp\instant-
  1220. ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\dependencies\glfw\include"
  1221. -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:\ProgramData\NVIDIA Co
  1222. rporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"C:\ngp\instant-ngp\depen
  1223. dencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" --keep-dir
  1224. x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambda --expt-relaxed-constexpr -
  1225. std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX -DTCNN_MIN_GPU_ARCH=75 -DTCNN
  1226. _SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -Xcompiler "/EHsc /W1 /nologo
  1227. /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cuda-nn.pdb /FS /Zi /MD /GR" -o tiny
  1228. -cuda-nn.dir\RelWithDebInfo\cutlass_mlp.obj "C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_mlp.cu"" exited w
  1229. ith code 1. [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1230. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(415): error : name followed by "::" must be a class
  1231. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1232.  
  1233. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(493): error : name followed by "::" must be a class
  1234. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1235.  
  1236. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(493): error : name followed by "::" must be a class
  1237. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1238.  
  1239. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/grid.h(249): error : no operator "+=" match
  1240. es these operands [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1241. operand types are: __half += __half
  1242. detected during:
  1243. instantiation of "void tcnn::kernel_grid<T,N_POS_DIMS,N_FEATURES_PER_LEVEL>(uint32_t, uint32_t, const uin
  1244. t32_t *, uint32_t, float, float, float, const float *, tcnn::InterpolationType, tcnn::GridType, const T *, const floa
  1245. t *, T *, float *) [with T=__half, N_POS_DIMS=2U, N_FEATURES_PER_LEVEL=1U]"
  1246. (679): here
  1247. instantiation of "void tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::encode(cudaStrea
  1248. m_t, uint32_t, tcnn::PitchedPtr<const float>, tcnn::PitchedPtr<T>, float *, __nv_bool) const [with T=__half, N_POS_DI
  1249. MS=2U, N_FEATURES_PER_LEVEL=1U]"
  1250. (608): here
  1251. implicit generation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::~GridEncodingTe
  1252. mplated() [with T=__half, N_POS_DIMS=2U, N_FEATURES_PER_LEVEL=1U]"
  1253. (608): here
  1254. instantiation of class "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL> [with T=__half,
  1255. N_POS_DIMS=2U, N_FEATURES_PER_LEVEL=1U]"
  1256. (608): here
  1257. instantiation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::GridEncodingTemplated
  1258. (uint32_t, uint32_t, uint32_t, float, __nv_bool, tcnn::InterpolationType, tcnn::GridType) [with T=__half, N_POS_DIMS=
  1259. 2U, N_FEATURES_PER_LEVEL=1U]"
  1260. (932): here
  1261. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding_templated<T,N_FEATURES_PER_LEVEL>(uin
  1262. t32_t, const tcnn::json &) [with T=__half, N_FEATURES_PER_LEVEL=1U]"
  1263. (944): here
  1264. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding<T>(uint32_t, const tcnn::json &) [wit
  1265. h T=__half]"
  1266. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\encoding.cu(126): here
  1267. instantiation of "tcnn::Encoding<T> *tcnn::create_encoding<T>(uint32_t, const tcnn::json &, uint32_t) [wi
  1268. th T=__half]"
  1269. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/composite.h(84): here
  1270.  
  1271. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/grid.h(249): error : no operator "+=" match
  1272. es these operands [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1273. common.cu
  1274. operand types are: __half += __half
  1275. detected during:
  1276. instantiation of "void tcnn::kernel_grid<T,N_POS_DIMS,N_FEATURES_PER_LEVEL>(uint32_t, uint32_t, const uin
  1277. t32_t *, uint32_t, float, float, float, const float *, tcnn::InterpolationType, tcnn::GridType, const T *, const floa
  1278. t *, T *, float *) [with T=__half, N_POS_DIMS=3U, N_FEATURES_PER_LEVEL=1U]"
  1279. (679): here
  1280. instantiation of "void tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::encode(cudaStrea
  1281. m_t, uint32_t, tcnn::PitchedPtr<const float>, tcnn::PitchedPtr<T>, float *, __nv_bool) const [with T=__half, N_POS_DI
  1282. MS=3U, N_FEATURES_PER_LEVEL=1U]"
  1283. (608): here
  1284. implicit generation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::~GridEncodingTe
  1285. mplated() [with T=__half, N_POS_DIMS=3U, N_FEATURES_PER_LEVEL=1U]"
  1286. (608): here
  1287. instantiation of class "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL> [with T=__half,
  1288. N_POS_DIMS=3U, N_FEATURES_PER_LEVEL=1U]"
  1289. (608): here
  1290. instantiation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::GridEncodingTemplated
  1291. (uint32_t, uint32_t, uint32_t, float, __nv_bool, tcnn::InterpolationType, tcnn::GridType) [with T=__half, N_POS_DIMS=
  1292. 3U, N_FEATURES_PER_LEVEL=1U]"
  1293. (933): here
  1294. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding_templated<T,N_FEATURES_PER_LEVEL>(uin
  1295. t32_t, const tcnn::json &) [with T=__half, N_FEATURES_PER_LEVEL=1U]"
  1296. (944): here
  1297. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding<T>(uint32_t, const tcnn::json &) [wit
  1298. h T=__half]"
  1299. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\encoding.cu(126): here
  1300. instantiation of "tcnn::Encoding<T> *tcnn::create_encoding<T>(uint32_t, const tcnn::json &, uint32_t) [wi
  1301. th T=__half]"
  1302. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/composite.h(84): here
  1303.  
  1304. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/grid.h(249): error : no operator "+=" match
  1305. es these operands [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1306. operand types are: __half += __half
  1307. detected during:
  1308. instantiation of "void tcnn::kernel_grid<T,N_POS_DIMS,N_FEATURES_PER_LEVEL>(uint32_t, uint32_t, const uin
  1309. t32_t *, uint32_t, float, float, float, const float *, tcnn::InterpolationType, tcnn::GridType, const T *, const floa
  1310. t *, T *, float *) [with T=__half, N_POS_DIMS=2U, N_FEATURES_PER_LEVEL=2U]"
  1311. (679): here
  1312. instantiation of "void tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::encode(cudaStrea
  1313. m_t, uint32_t, tcnn::PitchedPtr<const float>, tcnn::PitchedPtr<T>, float *, __nv_bool) const [with T=__half, N_POS_DI
  1314. MS=2U, N_FEATURES_PER_LEVEL=2U]"
  1315. (608): here
  1316. implicit generation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::~GridEncodingTe
  1317. mplated() [with T=__half, N_POS_DIMS=2U, N_FEATURES_PER_LEVEL=2U]"
  1318. (608): here
  1319. instantiation of class "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL> [with T=__half,
  1320. N_POS_DIMS=2U, N_FEATURES_PER_LEVEL=2U]"
  1321. (608): here
  1322. instantiation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::GridEncodingTemplated
  1323. (uint32_t, uint32_t, uint32_t, float, __nv_bool, tcnn::InterpolationType, tcnn::GridType) [with T=__half, N_POS_DIMS=
  1324. 2U, N_FEATURES_PER_LEVEL=2U]"
  1325. (932): here
  1326. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding_templated<T,N_FEATURES_PER_LEVEL>(uin
  1327. t32_t, const tcnn::json &) [with T=__half, N_FEATURES_PER_LEVEL=2U]"
  1328. (945): here
  1329. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding<T>(uint32_t, const tcnn::json &) [wit
  1330. h T=__half]"
  1331. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\encoding.cu(126): here
  1332. instantiation of "tcnn::Encoding<T> *tcnn::create_encoding<T>(uint32_t, const tcnn::json &, uint32_t) [wi
  1333. th T=__half]"
  1334. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/composite.h(84): here
  1335.  
  1336. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/grid.h(249): error : no operator "+=" match
  1337. es these operands [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1338. operand types are: __half += __half
  1339. detected during:
  1340. instantiation of "void tcnn::kernel_grid<T,N_POS_DIMS,N_FEATURES_PER_LEVEL>(uint32_t, uint32_t, const uin
  1341. t32_t *, uint32_t, float, float, float, const float *, tcnn::InterpolationType, tcnn::GridType, const T *, const floa
  1342. t *, T *, float *) [with T=__half, N_POS_DIMS=3U, N_FEATURES_PER_LEVEL=2U]"
  1343. (679): here
  1344. instantiation of "void tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::encode(cudaStrea
  1345. m_t, uint32_t, tcnn::PitchedPtr<const float>, tcnn::PitchedPtr<T>, float *, __nv_bool) const [with T=__half, N_POS_DI
  1346. MS=3U, N_FEATURES_PER_LEVEL=2U]"
  1347. (608): here
  1348. implicit generation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::~GridEncodingTe
  1349. mplated() [with T=__half, N_POS_DIMS=3U, N_FEATURES_PER_LEVEL=2U]"
  1350. (608): here
  1351. instantiation of class "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL> [with T=__half,
  1352. N_POS_DIMS=3U, N_FEATURES_PER_LEVEL=2U]"
  1353. (608): here
  1354. instantiation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::GridEncodingTemplated
  1355. (uint32_t, uint32_t, uint32_t, float, __nv_bool, tcnn::InterpolationType, tcnn::GridType) [with T=__half, N_POS_DIMS=
  1356. 3U, N_FEATURES_PER_LEVEL=2U]"
  1357. (933): here
  1358. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding_templated<T,N_FEATURES_PER_LEVEL>(uin
  1359. t32_t, const tcnn::json &) [with T=__half, N_FEATURES_PER_LEVEL=2U]"
  1360. (945): here
  1361. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding<T>(uint32_t, const tcnn::json &) [wit
  1362. h T=__half]"
  1363. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\encoding.cu(126): here
  1364. instantiation of "tcnn::Encoding<T> *tcnn::create_encoding<T>(uint32_t, const tcnn::json &, uint32_t) [wi
  1365. th T=__half]"
  1366. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/composite.h(84): here
  1367.  
  1368. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/grid.h(249): error : no operator "+=" match
  1369. es these operands [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1370. operand types are: __half += __half
  1371. detected during:
  1372. instantiation of "void tcnn::kernel_grid<T,N_POS_DIMS,N_FEATURES_PER_LEVEL>(uint32_t, uint32_t, const uin
  1373. t32_t *, uint32_t, float, float, float, const float *, tcnn::InterpolationType, tcnn::GridType, const T *, const floa
  1374. t *, T *, float *) [with T=__half, N_POS_DIMS=2U, N_FEATURES_PER_LEVEL=4U]"
  1375. (679): here
  1376. instantiation of "void tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::encode(cudaStrea
  1377. m_t, uint32_t, tcnn::PitchedPtr<const float>, tcnn::PitchedPtr<T>, float *, __nv_bool) const [with T=__half, N_POS_DI
  1378. MS=2U, N_FEATURES_PER_LEVEL=4U]"
  1379. (608): here
  1380. implicit generation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::~GridEncodingTe
  1381. mplated() [with T=__half, N_POS_DIMS=2U, N_FEATURES_PER_LEVEL=4U]"
  1382. (608): here
  1383. instantiation of class "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL> [with T=__half,
  1384. N_POS_DIMS=2U, N_FEATURES_PER_LEVEL=4U]"
  1385. (608): here
  1386. instantiation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::GridEncodingTemplated
  1387. (uint32_t, uint32_t, uint32_t, float, __nv_bool, tcnn::InterpolationType, tcnn::GridType) [with T=__half, N_POS_DIMS=
  1388. 2U, N_FEATURES_PER_LEVEL=4U]"
  1389. (932): here
  1390. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding_templated<T,N_FEATURES_PER_LEVEL>(uin
  1391. t32_t, const tcnn::json &) [with T=__half, N_FEATURES_PER_LEVEL=4U]"
  1392. (946): here
  1393. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding<T>(uint32_t, const tcnn::json &) [wit
  1394. h T=__half]"
  1395. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\encoding.cu(126): here
  1396. instantiation of "tcnn::Encoding<T> *tcnn::create_encoding<T>(uint32_t, const tcnn::json &, uint32_t) [wi
  1397. th T=__half]"
  1398. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/composite.h(84): here
  1399.  
  1400. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/grid.h(249): error : no operator "+=" match
  1401. es these operands [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1402. operand types are: __half += __half
  1403. detected during:
  1404. instantiation of "void tcnn::kernel_grid<T,N_POS_DIMS,N_FEATURES_PER_LEVEL>(uint32_t, uint32_t, const uin
  1405. t32_t *, uint32_t, float, float, float, const float *, tcnn::InterpolationType, tcnn::GridType, const T *, const floa
  1406. t *, T *, float *) [with T=__half, N_POS_DIMS=3U, N_FEATURES_PER_LEVEL=4U]"
  1407. (679): here
  1408. instantiation of "void tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::encode(cudaStrea
  1409. m_t, uint32_t, tcnn::PitchedPtr<const float>, tcnn::PitchedPtr<T>, float *, __nv_bool) const [with T=__half, N_POS_DI
  1410. MS=3U, N_FEATURES_PER_LEVEL=4U]"
  1411. (608): here
  1412. implicit generation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::~GridEncodingTe
  1413. mplated() [with T=__half, N_POS_DIMS=3U, N_FEATURES_PER_LEVEL=4U]"
  1414. (608): here
  1415. instantiation of class "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL> [with T=__half,
  1416. N_POS_DIMS=3U, N_FEATURES_PER_LEVEL=4U]"
  1417. (608): here
  1418. instantiation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::GridEncodingTemplated
  1419. (uint32_t, uint32_t, uint32_t, float, __nv_bool, tcnn::InterpolationType, tcnn::GridType) [with T=__half, N_POS_DIMS=
  1420. 3U, N_FEATURES_PER_LEVEL=4U]"
  1421. (933): here
  1422. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding_templated<T,N_FEATURES_PER_LEVEL>(uin
  1423. t32_t, const tcnn::json &) [with T=__half, N_FEATURES_PER_LEVEL=4U]"
  1424. (946): here
  1425. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding<T>(uint32_t, const tcnn::json &) [wit
  1426. h T=__half]"
  1427. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\encoding.cu(126): here
  1428. instantiation of "tcnn::Encoding<T> *tcnn::create_encoding<T>(uint32_t, const tcnn::json &, uint32_t) [wi
  1429. th T=__half]"
  1430. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/composite.h(84): here
  1431.  
  1432. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/grid.h(249): error : no operator "+=" match
  1433. es these operands [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1434. operand types are: __half += __half
  1435. detected during:
  1436. instantiation of "void tcnn::kernel_grid<T,N_POS_DIMS,N_FEATURES_PER_LEVEL>(uint32_t, uint32_t, const uin
  1437. t32_t *, uint32_t, float, float, float, const float *, tcnn::InterpolationType, tcnn::GridType, const T *, const floa
  1438. t *, T *, float *) [with T=__half, N_POS_DIMS=2U, N_FEATURES_PER_LEVEL=8U]"
  1439. (679): here
  1440. instantiation of "void tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::encode(cudaStrea
  1441. m_t, uint32_t, tcnn::PitchedPtr<const float>, tcnn::PitchedPtr<T>, float *, __nv_bool) const [with T=__half, N_POS_DI
  1442. MS=2U, N_FEATURES_PER_LEVEL=8U]"
  1443. (608): here
  1444. implicit generation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::~GridEncodingTe
  1445. mplated() [with T=__half, N_POS_DIMS=2U, N_FEATURES_PER_LEVEL=8U]"
  1446. (608): here
  1447. instantiation of class "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL> [with T=__half,
  1448. N_POS_DIMS=2U, N_FEATURES_PER_LEVEL=8U]"
  1449. (608): here
  1450. instantiation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::GridEncodingTemplated
  1451. (uint32_t, uint32_t, uint32_t, float, __nv_bool, tcnn::InterpolationType, tcnn::GridType) [with T=__half, N_POS_DIMS=
  1452. 2U, N_FEATURES_PER_LEVEL=8U]"
  1453. (932): here
  1454. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding_templated<T,N_FEATURES_PER_LEVEL>(uin
  1455. t32_t, const tcnn::json &) [with T=__half, N_FEATURES_PER_LEVEL=8U]"
  1456. (947): here
  1457. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding<T>(uint32_t, const tcnn::json &) [wit
  1458. h T=__half]"
  1459. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\encoding.cu(126): here
  1460. instantiation of "tcnn::Encoding<T> *tcnn::create_encoding<T>(uint32_t, const tcnn::json &, uint32_t) [wi
  1461. th T=__half]"
  1462. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/composite.h(84): here
  1463.  
  1464. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/grid.h(249): error : no operator "+=" match
  1465. es these operands [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1466. operand types are: __half += __half
  1467. detected during:
  1468. instantiation of "void tcnn::kernel_grid<T,N_POS_DIMS,N_FEATURES_PER_LEVEL>(uint32_t, uint32_t, const uin
  1469. t32_t *, uint32_t, float, float, float, const float *, tcnn::InterpolationType, tcnn::GridType, const T *, const floa
  1470. t *, T *, float *) [with T=__half, N_POS_DIMS=3U, N_FEATURES_PER_LEVEL=8U]"
  1471. (679): here
  1472. instantiation of "void tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::encode(cudaStrea
  1473. m_t, uint32_t, tcnn::PitchedPtr<const float>, tcnn::PitchedPtr<T>, float *, __nv_bool) const [with T=__half, N_POS_DI
  1474. MS=3U, N_FEATURES_PER_LEVEL=8U]"
  1475. (608): here
  1476. implicit generation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::~GridEncodingTe
  1477. mplated() [with T=__half, N_POS_DIMS=3U, N_FEATURES_PER_LEVEL=8U]"
  1478. (608): here
  1479. instantiation of class "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL> [with T=__half,
  1480. N_POS_DIMS=3U, N_FEATURES_PER_LEVEL=8U]"
  1481. (608): here
  1482. instantiation of "tcnn::GridEncodingTemplated<T, N_POS_DIMS, N_FEATURES_PER_LEVEL>::GridEncodingTemplated
  1483. (uint32_t, uint32_t, uint32_t, float, __nv_bool, tcnn::InterpolationType, tcnn::GridType) [with T=__half, N_POS_DIMS=
  1484. 3U, N_FEATURES_PER_LEVEL=8U]"
  1485. (933): here
  1486. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding_templated<T,N_FEATURES_PER_LEVEL>(uin
  1487. t32_t, const tcnn::json &) [with T=__half, N_FEATURES_PER_LEVEL=8U]"
  1488. (947): here
  1489. instantiation of "tcnn::GridEncoding<T> *tcnn::create_grid_encoding<T>(uint32_t, const tcnn::json &) [wit
  1490. h T=__half]"
  1491. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\encoding.cu(126): here
  1492. instantiation of "tcnn::Encoding<T> *tcnn::create_encoding<T>(uint32_t, const tcnn::json &, uint32_t) [wi
  1493. th T=__half]"
  1494. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/encodings/composite.h(84): here
  1495.  
  1496. cpp_api.cu
  1497. common_device.cu
  1498. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(519): error : more than one conversio
  1499. n function from "const tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\ti
  1500. ny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1501. function "__half::operator float() const"
  1502. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  1503. function "__half::operator short() const"
  1504. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  1505. function "__half::operator unsigned short() const"
  1506. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  1507. function "__half::operator int() const"
  1508. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  1509. function "__half::operator unsigned int() const"
  1510. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  1511. function "__half::operator long long() const"
  1512. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  1513. function "__half::operator unsigned long long() const"
  1514. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  1515. function "__half::operator __nv_bool() const"
  1516. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  1517. detected during:
  1518. instantiation of "void tcnn::add(uint32_t, const T *, T *) [with T=tcnn::network_precision_t]"
  1519. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(156): here
  1520. instantiation of "void tcnn::CutlassResNet<T, input_activation>::inference_mixed_precision(cudaStream_t,
  1521. const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, input_a
  1522. ctivation=tcnn::Activation::None]"
  1523. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(106): here
  1524.  
  1525. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(519): error : more than one conversio
  1526. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  1527. a-nn\src\tiny-cuda-nn.vcxproj]
  1528. function "__half::operator float() const"
  1529. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  1530. function "__half::operator short() const"
  1531. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  1532. function "__half::operator unsigned short() const"
  1533. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  1534. function "__half::operator int() const"
  1535. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  1536. function "__half::operator unsigned int() const"
  1537. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  1538. function "__half::operator long long() const"
  1539. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  1540. function "__half::operator unsigned long long() const"
  1541. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  1542. function "__half::operator __nv_bool() const"
  1543. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  1544. detected during:
  1545. instantiation of "void tcnn::add(uint32_t, const T *, T *) [with T=tcnn::network_precision_t]"
  1546. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(156): here
  1547. instantiation of "void tcnn::CutlassResNet<T, input_activation>::inference_mixed_precision(cudaStream_t,
  1548. const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, input_a
  1549. ctivation=tcnn::Activation::None]"
  1550. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(106): here
  1551.  
  1552. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  1553. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1554. detected during:
  1555. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1556. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1557. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1558. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1559. (760): here
  1560. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1561. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1562. (714): here
  1563.  
  1564. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  1565. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1566. detected during:
  1567. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1568. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1569. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1570. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1571. (760): here
  1572. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1573. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1574. (714): here
  1575.  
  1576. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  1577. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1578. detected during:
  1579. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1580. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1581. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1582. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1583. (760): here
  1584. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1585. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1586. (714): here
  1587.  
  1588. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  1589. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1590. detected during:
  1591. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1592. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1593. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1594. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1595. (760): here
  1596. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1597. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1598. (714): here
  1599.  
  1600. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(515): error : name followed by "::" must be a class
  1601. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1602. detected during:
  1603. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1604. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1605. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1606. (636): here
  1607. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1608. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1609. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1610. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1611. (760): here
  1612. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1613. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1614. (714): here
  1615.  
  1616. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(516): error : name followed by "::" must be a class
  1617. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1618. detected during:
  1619. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1620. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1621. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1622. (636): here
  1623. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1624. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1625. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1626. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1627. (760): here
  1628. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1629. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1630. (714): here
  1631.  
  1632. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(517): error : name followed by "::" must be a class
  1633. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1634. detected during:
  1635. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1636. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1637. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1638. (636): here
  1639. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1640. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1641. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1642. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1643. (760): here
  1644. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1645. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1646. (714): here
  1647.  
  1648. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(519): error : name followed by "::" must be a class
  1649. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1650. detected during:
  1651. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1652. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1653. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1654. (636): here
  1655. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1656. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1657. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1658. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1659. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(74): error : more than one conversion
  1660. function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cuda
  1661. -nn\src\tiny-cuda-nn.vcxproj]
  1662. function "__half::operator float() const"
  1663. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  1664. function "__half::operator short() const"
  1665. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  1666. function "__half::operator unsigned short() const"
  1667. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  1668. function "__half::operator int() const"
  1669. (760): here
  1670. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  1671. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1672. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1673. (714): here
  1674.  
  1675. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(59): error : name must be a namespace name [C:\ngp\
  1676. instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1677. detected during:
  1678. function "__half::operator unsigned int() const"
  1679. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  1680. function "__half::operator long long() const"
  1681. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  1682. function "__half::operator unsigned long long() const"
  1683. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  1684. function "__half::operator __nv_bool() const"
  1685. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1686. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1687. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  1688. (525): here
  1689. detected during:
  1690. instantiation of "void tcnn::warp_activation<T,fragment_t>(tcnn::Activation, const fragment_t &, fragment
  1691. _t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U
  1692. >>]"
  1693. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1694. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1695. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1696. (244): here
  1697. (636): here
  1698. instantiation of "void tcnn::kernel_activation(uint32_t, tcnn::Activation, const T *, T *) [with T=tcnn::
  1699. network_precision_t, N=8U]"
  1700. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1701. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1702. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1703. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1704. (760): here
  1705. (286): here
  1706. instantiation of "void tcnn::activation_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *, T *) [wi
  1707. th T=tcnn::network_precision_t]"
  1708. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1709. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1710. (714): here
  1711. (295): here
  1712. instantiation of "void tcnn::activation_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrixDynamic<
  1713. T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  1714.  
  1715. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(202): here
  1716. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::CutlassResNet<
  1717. T, input_activation>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool
  1718. , __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::None]"
  1719. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(63): error : name followed by "::" must be a class
  1720. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1721. detected during:
  1722. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1723. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1724. (525): here
  1725. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1726. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1727. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1728. (636): here
  1729. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1730. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1731. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1732. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1733. (760): here
  1734. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1735. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1736. (714): here
  1737. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  1738.  
  1739.  
  1740. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(74): error : more than one conversion
  1741. function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cuda
  1742. -nn\src\tiny-cuda-nn.vcxproj]
  1743. function "__half::operator float() const"
  1744. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  1745. function "__half::operator short() const"
  1746. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  1747. function "__half::operator unsigned short() const"
  1748. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  1749. function "__half::operator int() const"
  1750. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  1751. function "__half::operator unsigned int() const"
  1752. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  1753. function "__half::operator long long() const"
  1754. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  1755. function "__half::operator unsigned long long() const"
  1756. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  1757. function "__half::operator __nv_bool() const"
  1758. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  1759. detected during:
  1760. instantiation of "void tcnn::warp_activation<T,fragment_t>(tcnn::Activation, const fragment_t &, fragment
  1761. _t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U
  1762. >>]"
  1763. (244): here
  1764. instantiation of "void tcnn::kernel_activation(uint32_t, tcnn::Activation, const T *, T *) [with T=tcnn::
  1765. network_precision_t, N=8U]"
  1766. (286): here
  1767. instantiation of "void tcnn::activation_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *, T *) [wi
  1768. th T=tcnn::network_precision_t]"
  1769. (295): here
  1770. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(63): error : name followed by "::" must be a class
  1771. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1772. instantiation of "void tcnn::activation_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrixDynamic<
  1773. T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  1774. detected during:
  1775. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(202): here
  1776. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::CutlassResNet<
  1777. T, input_activation>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool
  1778. , __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::None]"
  1779. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1780. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1781. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  1782. (525): here
  1783.  
  1784. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1785. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1786. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1787. (636): here
  1788. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1789. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1790. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1791. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1792. (760): here
  1793. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1794. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1795. (714): here
  1796.  
  1797. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(187): error : more than one conversio
  1798. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  1799. a-nn\src\tiny-cuda-nn.vcxproj]
  1800. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(66): error : name followed by "::" must be a class
  1801. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1802. detected during:
  1803. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1804. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1805. function "__half::operator float() const"
  1806. (525): here
  1807. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1808. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1809. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1810. (636): here
  1811. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1812. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1813. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1814. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1815. (760): here
  1816. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1817. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1818. (714): here
  1819.  
  1820. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(66): error : type name is not allowed [C:\ngp\insta
  1821. nt-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1822. detected during:
  1823. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1824. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1825. (525): here
  1826. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1827. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1828. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1829. (636): here
  1830. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1831. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1832. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1833. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1834. (760): here
  1835. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1836. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1837. (714): here
  1838.  
  1839. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(66): error : name followed by "::" must be a class
  1840. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1841. detected during:
  1842. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1843. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1844. (525): here
  1845. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1846. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1847. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1848. (636): here
  1849. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1850. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1851. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1852. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1853. (760): here
  1854. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1855. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1856. (714): here
  1857.  
  1858. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(66): error : identifier "act_frag" is undefined [C:
  1859. \ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1860. detected during:
  1861. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1862. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1863. (525): here
  1864. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1865. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1866. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1867. (636): here
  1868. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  1869. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1870. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1871. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1872. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1873. function "__half::operator short() const"
  1874. (760): here
  1875. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1876. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1877. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  1878. function "__half::operator unsigned short() const"
  1879. (714): here
  1880. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  1881.  
  1882. function "__half::operator int() const"
  1883. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(67): error : name followed by "::" must be a class
  1884. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1885. detected during:
  1886. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  1887. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1888. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1889. function "__half::operator unsigned int() const"
  1890. (525): here
  1891. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  1892. function "__half::operator long long() const"
  1893. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1894. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1895. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1896. (636): here
  1897. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1898. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1899. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1900. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1901. (760): here
  1902. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1903. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1904. (714): here
  1905.  
  1906. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(67): error : type name is not allowed [C:\ngp\insta
  1907. nt-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1908. detected during:
  1909. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1910. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1911. (525): here
  1912. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1913. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1914. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1915. (636): here
  1916. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1917. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1918. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1919. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1920. (760): here
  1921. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1922. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1923. (714): here
  1924.  
  1925. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(67): error : type name is not allowed [C:\ngp\insta
  1926. nt-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1927. detected during:
  1928. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1929. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1930. (525): here
  1931. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1932. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1933. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1934. (636): here
  1935. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1936. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1937. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1938. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1939. (760): here
  1940. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1941. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1942. (714): here
  1943.  
  1944. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(67): error : identifier "weights_frag" is undefined
  1945. [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1946. detected during:
  1947. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1948. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1949. (525): here
  1950. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1951. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1952. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1953. (636): here
  1954. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1955. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1956. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1957. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1958. (760): here
  1959. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1960. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1961. (714): here
  1962.  
  1963. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(68): error : name followed by "::" must be a class
  1964. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1965. detected during:
  1966. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1967. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1968. (525): here
  1969. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1970. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1971. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1972. (636): here
  1973. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  1974. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1975. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  1976. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  1977. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1978. function "__half::operator unsigned long long() const"
  1979. (760): here
  1980. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  1981. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  1982. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  1983. (714): here
  1984. function "__half::operator __nv_bool() const"
  1985.  
  1986. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  1987. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(68): error : type name is not allowed [C:\ngp\insta
  1988. nt-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  1989. detected during:
  1990. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  1991. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  1992. (525): here
  1993. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  1994. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  1995. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  1996. (636): here
  1997. detected during:
  1998. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  1999. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2000. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2001. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2002. (760): here
  2003. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  2004. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  2005. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  2006. nn::network_precision_t, 8U>>]"
  2007. (269): here
  2008. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2009. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2010. (714): here
  2011.  
  2012. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(68): error : identifier "result_frag" is undefined
  2013. [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2014. detected during:
  2015. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  2016. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  2017. (525): here
  2018. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2019. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2020. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2021. (636): here
  2022. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2023. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2024. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2025. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2026. (760): here
  2027. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2028. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2029. (714): here
  2030.  
  2031. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(87): error : name followed by "::" must be a class
  2032. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2033. detected during:
  2034. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  2035. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  2036. (525): here
  2037. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2038. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2039. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2040. (636): here
  2041. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2042. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2043. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2044. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2045. (760): here
  2046. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2047. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2048. (714): here
  2049.  
  2050. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(89): error : name followed by "::" must be a class
  2051. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2052. detected during:
  2053. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  2054. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  2055. (525): here
  2056. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2057. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2058. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2059. (636): here
  2060. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2061. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2062. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2063. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2064. (760): here
  2065. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2066. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2067. (714): here
  2068.  
  2069. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(95): error : name followed by "::" must be a class
  2070. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2071. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  2072. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  2073. detected during:
  2074. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  2075. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  2076. (525): here
  2077. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2078. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2079. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2080. (636): here
  2081. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2082. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2083. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2084. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2085. (334): here
  2086. (760): here
  2087. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2088. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2089. (714): here
  2090.  
  2091. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(100): error : name followed by "::" must be a class
  2092. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2093. detected during:
  2094. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  2095. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  2096. (525): here
  2097. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2098. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2099. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2100. (636): here
  2101. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2102. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2103. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2104. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2105. (760): here
  2106. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2107. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2108. (714): here
  2109.  
  2110. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(101): error : name followed by "::" must be a class
  2111. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2112. detected during:
  2113. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  2114. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  2115. (525): here
  2116. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2117. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2118. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2119. (636): here
  2120. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2121. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2122. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2123. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2124. (760): here
  2125. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2126. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2127. (714): here
  2128.  
  2129. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(107): error : name followed by "::" must be a class
  2130. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2131. detected during:
  2132. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  2133. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  2134. (525): here
  2135. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2136. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2137. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2138. (636): here
  2139. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2140. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2141. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2142. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2143. (760): here
  2144. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2145. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2146. (714): here
  2147.  
  2148. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(118): error : name followed by "::" must be a class
  2149. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2150. detected during:
  2151. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  2152. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  2153. (525): here
  2154. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2155. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2156. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2157. (636): here
  2158. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2159. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2160. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2161. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2162. (760): here
  2163. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2164. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2165. (714): here
  2166.  
  2167. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(118): error : name followed by "::" must be a class
  2168. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2169. detected during:
  2170. instantiation of "void tcnn::threadblock_layer<WIDTH,N_ITERS,OUT_T,BACKWARD>(tcnn::Activation, __half *,
  2171. const __half *, OUT_T *, const OUT_T *) [with WIDTH=128, N_ITERS=8, OUT_T=__half, BACKWARD=false]"
  2172. (525): here
  2173. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2174. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2175. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2176. (636): here
  2177. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2178. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2179. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2180. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2181. (760): here
  2182. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2183. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2184. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  2185. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  2186. (714): here
  2187. 8 errors detected in the compilation of "C:/ngp/instant-ngp/dependencies/tiny-cuda-nn/src/encoding.cu".
  2188.  
  2189. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(544): error : name followed by "::" must be a class
  2190. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2191. detected during:
  2192. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2193. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2194. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2195. (636): here
  2196. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2197. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2198. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2199. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=true]"
  2200. (760): here
  2201. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2202. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2203. (714): here
  2204.  
  2205. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  2206. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2207. detected during:
  2208. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2209. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2210. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2211. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2212. (761): here
  2213. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2214. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2215. (714): here
  2216.  
  2217. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  2218. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2219. detected during:
  2220. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2221. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2222. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2223. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2224. (761): here
  2225. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2226. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2227. (714): here
  2228.  
  2229. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  2230. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2231. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(274): here
  2232. detected during:
  2233. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2234. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2235. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2236. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2237. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  2238. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  2239. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  2240. ne]"
  2241. (761): here
  2242. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2243. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2244. (714): here
  2245. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  2246.  
  2247.  
  2248. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  2249. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2250. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(187): error : more than one conversio
  2251. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  2252. a-nn\src\tiny-cuda-nn.vcxproj]
  2253. function "__half::operator float() const"
  2254. detected during:
  2255. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  2256. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2257. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2258. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2259. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2260. (761): here
  2261. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2262. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2263. function "__half::operator short() const"
  2264. (714): here
  2265. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  2266.  
  2267. function "__half::operator unsigned short() const"
  2268. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(515): error : name followed by "::" must be a class
  2269. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2270. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  2271. function "__half::operator int() const"
  2272. detected during:
  2273. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2274. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2275. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2276. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  2277. (636): here
  2278. function "__half::operator unsigned int() const"
  2279. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2280. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2281. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2282. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2283. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  2284. (761): here
  2285. function "__half::operator long long() const"
  2286. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2287. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2288. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  2289. (714): here
  2290. function "__half::operator unsigned long long() const"
  2291.  
  2292. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  2293. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(516): error : name followed by "::" must be a class
  2294. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2295. function "__half::operator __nv_bool() const"
  2296. detected during:
  2297. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  2298. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2299. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2300. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2301. detected during:
  2302. (636): here
  2303. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  2304. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  2305. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  2306. nn::network_precision_t, 8U>>]"
  2307. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2308. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2309. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2310. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2311. (269): here
  2312. (761): here
  2313. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  2314. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  2315. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2316. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2317. (714): here
  2318. (334): here
  2319. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  2320. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  2321.  
  2322. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(274): here
  2323. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(517): error : name followed by "::" must be a class
  2324. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2325. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  2326. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  2327. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  2328. ne]"
  2329. detected during:
  2330. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2331. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2332. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2333. (636): here
  2334. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2335. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2336. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2337. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2338. (761): here
  2339. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2340. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2341. (714): here
  2342.  
  2343. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(519): error : name followed by "::" must be a class
  2344. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2345. detected during:
  2346. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2347. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2348. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2349. (636): here
  2350. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2351. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2352. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2353. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2354. (761): here
  2355. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2356. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2357. (714): here
  2358. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  2359.  
  2360. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(544): error : name followed by "::" must be a class
  2361. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2362. detected during:
  2363. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2364. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2365. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2366. (636): here
  2367. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2368. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2369. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2370. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=true]"
  2371. (761): here
  2372. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2373. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2374. (714): here
  2375.  
  2376. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  2377. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2378. detected during:
  2379. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2380. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2381. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2382. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2383. (762): here
  2384. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2385. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2386. (714): here
  2387.  
  2388. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  2389. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2390. detected during:
  2391. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2392. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2393. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2394. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2395. (762): here
  2396. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2397. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2398. (714): here
  2399.  
  2400. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  2401. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2402. detected during:
  2403.  
  2404. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2405. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2406. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2407. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2408. encoding.cu
  2409. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(193): error : more than one conversio
  2410. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  2411. a-nn\src\tiny-cuda-nn.vcxproj]
  2412. (762): here
  2413. function "__half::operator float() const"
  2414. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2415. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2416. (714): here
  2417. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  2418. function "__half::operator short() const"
  2419.  
  2420. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  2421. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  2422. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2423. function "__half::operator unsigned short() const"
  2424. detected during:
  2425. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  2426. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2427. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2428. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2429. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2430. function "__half::operator int() const"
  2431. (762): here
  2432. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  2433. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2434. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2435. function "__half::operator unsigned int() const"
  2436. (714): here
  2437. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  2438.  
  2439. function "__half::operator long long() const"
  2440. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(515): error : name followed by "::" must be a class
  2441. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2442. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  2443. detected during:
  2444. function "__half::operator unsigned long long() const"
  2445. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2446. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2447. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2448. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  2449. (636): here
  2450. function "__half::operator __nv_bool() const"
  2451. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2452. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2453. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2454. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2455. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  2456. (762): here
  2457. detected during:
  2458. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2459. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2460. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  2461. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  2462. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  2463. nn::network_precision_t, 8U>>]"
  2464. (714): here
  2465. (269): here
  2466.  
  2467. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  2468. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  2469. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(516): error : name followed by "::" must be a class
  2470. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2471. (334): here
  2472. detected during:
  2473. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  2474. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  2475. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2476. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2477. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2478. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(274): here
  2479. (636): here
  2480. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2481. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2482. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2483. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2484. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  2485. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  2486. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  2487. ne]"
  2488. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  2489. (762): here
  2490.  
  2491. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2492. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2493. (714): here
  2494. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(193): error : more than one conversio
  2495. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  2496. a-nn\src\tiny-cuda-nn.vcxproj]
  2497. function "__half::operator float() const"
  2498.  
  2499. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(517): error : name followed by "::" must be a class
  2500. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2501. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  2502. function "__half::operator short() const"
  2503. detected during:
  2504. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  2505. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2506. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2507. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2508. function "__half::operator unsigned short() const"
  2509. (636): here
  2510. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  2511. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2512. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2513. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2514. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2515. function "__half::operator int() const"
  2516. (762): here
  2517. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  2518. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2519. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2520. function "__half::operator unsigned int() const"
  2521. (714): here
  2522. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  2523.  
  2524. function "__half::operator long long() const"
  2525. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(519): error : name followed by "::" must be a class
  2526. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2527. detected during:
  2528. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  2529. function "__half::operator unsigned long long() const"
  2530. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2531. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2532. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2533. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  2534. (636): here
  2535. function "__half::operator __nv_bool() const"
  2536. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2537. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2538. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2539. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2540. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  2541. (762): here
  2542. detected during:
  2543. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2544. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2545. (714): here
  2546. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  2547. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  2548. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  2549. nn::network_precision_t, 8U>>]"
  2550. (269): here
  2551.  
  2552. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  2553. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  2554. (334): here
  2555. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(544): error : name followed by "::" must be a class
  2556. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2557. detected during:
  2558. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2559. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2560. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2561. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  2562. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  2563. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(274): here
  2564. (636): here
  2565. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  2566. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  2567. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  2568. ne]"
  2569. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2570. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2571. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2572. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=true]"
  2573. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  2574. (762): here
  2575.  
  2576. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2577. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2578. (714): here
  2579. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(204): error : more than one conversio
  2580. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  2581. a-nn\src\tiny-cuda-nn.vcxproj]
  2582.  
  2583. function "__half::operator float() const"
  2584. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  2585. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  2586. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2587. detected during:
  2588. function "__half::operator short() const"
  2589. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  2590. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2591. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2592. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2593. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2594. function "__half::operator unsigned short() const"
  2595. (763): here
  2596. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2597. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2598. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  2599. (714): here
  2600. function "__half::operator int() const"
  2601.  
  2602. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  2603. function "__half::operator unsigned int() const"
  2604. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  2605. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2606. detected during:
  2607. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  2608. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2609. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2610. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2611. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2612. function "__half::operator long long() const"
  2613. (763): here
  2614. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  2615. function "__half::operator unsigned long long() const"
  2616. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2617. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2618. (714): here
  2619. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  2620. function "__half::operator __nv_bool() const"
  2621.  
  2622. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  2623. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2624. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  2625. detected during:
  2626. detected during:
  2627. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2628. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2629. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2630. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2631. (763): here
  2632. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  2633. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  2634. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  2635. nn::network_precision_t, 8U>>]"
  2636. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2637. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2638. (269): here
  2639. (714): here
  2640. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  2641. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  2642. (334): here
  2643.  
  2644. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  2645. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  2646. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  2647. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2648. detected during:
  2649. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(274): here
  2650. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  2651. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  2652. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  2653. ne]"
  2654. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2655. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2656. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2657. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2658. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  2659. (763): here
  2660.  
  2661. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2662. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2663. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(204): error : more than one conversio
  2664. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  2665. a-nn\src\tiny-cuda-nn.vcxproj]
  2666. (714): here
  2667. function "__half::operator float() const"
  2668.  
  2669. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  2670. function "__half::operator short() const"
  2671. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(515): error : name followed by "::" must be a class
  2672. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2673. detected during:
  2674. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  2675. function "__half::operator unsigned short() const"
  2676. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2677. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2678. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2679. (636): here
  2680. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  2681. function "__half::operator int() const"
  2682. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2683. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2684. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2685. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2686. (763): here
  2687. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  2688. function "__half::operator unsigned int() const"
  2689. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2690. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2691. (714): here
  2692. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  2693.  
  2694. function "__half::operator long long() const"
  2695. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  2696. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(516): error : name followed by "::" must be a class
  2697. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2698. detected during:
  2699. function "__half::operator unsigned long long() const"
  2700. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2701. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2702. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2703. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  2704. (636): here
  2705. function "__half::operator __nv_bool() const"
  2706. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2707. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2708. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2709. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2710. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  2711. (763): here
  2712. detected during:
  2713. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  2714. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  2715. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  2716. nn::network_precision_t, 8U>>]"
  2717. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2718. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2719. (714): here
  2720. (269): here
  2721. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  2722. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  2723.  
  2724. (334): here
  2725. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(517): error : name followed by "::" must be a class
  2726. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2727. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  2728. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  2729. detected during:
  2730. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(274): here
  2731. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  2732. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  2733. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  2734. ne]"
  2735. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2736. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2737. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2738. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  2739. (636): here
  2740.  
  2741. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2742. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2743. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2744. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2745. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(211): error : more than one conversio
  2746. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  2747. a-nn\src\tiny-cuda-nn.vcxproj]
  2748. function "__half::operator float() const"
  2749. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  2750. function "__half::operator short() const"
  2751. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  2752. function "__half::operator unsigned short() const"
  2753. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  2754. function "__half::operator int() const"
  2755. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  2756. function "__half::operator unsigned int() const"
  2757. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  2758. function "__half::operator long long() const"
  2759. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  2760. function "__half::operator unsigned long long() const"
  2761. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  2762. (763): here
  2763. function "__half::operator __nv_bool() const"
  2764. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  2765. detected during:
  2766. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2767. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2768. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  2769. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  2770. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  2771. nn::network_precision_t, 8U>>]"
  2772. (269): here
  2773. (714): here
  2774. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  2775. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  2776. (334): here
  2777. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  2778. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  2779.  
  2780. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(274): here
  2781. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  2782. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  2783. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  2784. ne]"
  2785. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(519): error : name followed by "::" must be a class
  2786. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2787. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  2788.  
  2789. detected during:
  2790. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2791. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2792. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2793. (636): here
  2794. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2795. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2796. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2797. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2798. (763): here
  2799. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2800. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2801. (714): here
  2802.  
  2803. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(544): error : name followed by "::" must be a class
  2804. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2805. detected during:
  2806. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2807. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2808. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2809. (636): here
  2810. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2811. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2812. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2813. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::ReLU, INFERENCE=true]"
  2814. (763): here
  2815. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2816. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2817. (714): here
  2818.  
  2819. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  2820. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2821. detected during:
  2822. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2823. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2824. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2825. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2826. (764): here
  2827. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2828. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2829. (714): here
  2830.  
  2831. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  2832. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2833. detected during:
  2834. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2835. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2836. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2837. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2838. (764): here
  2839. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2840. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2841. (714): here
  2842.  
  2843. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  2844. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2845. detected during:
  2846. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2847. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2848. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2849. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2850. (764): here
  2851. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2852. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2853. (714): here
  2854.  
  2855. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  2856. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2857. detected during:
  2858. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2859. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2860. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2861. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2862. (764): here
  2863. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2864. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2865. (714): here
  2866.  
  2867. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(515): error : name followed by "::" must be a class
  2868. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2869. detected during:
  2870. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2871. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2872. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2873. (636): here
  2874. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2875. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2876. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2877. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2878. (764): here
  2879. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2880. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2881. (714): here
  2882.  
  2883. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(516): error : name followed by "::" must be a class
  2884. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2885. detected during:
  2886. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2887. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2888. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2889. (636): here
  2890. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2891. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2892. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2893. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2894. (764): here
  2895. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2896. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2897. (714): here
  2898.  
  2899. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(517): error : name followed by "::" must be a class
  2900. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2901. detected during:
  2902. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2903. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2904. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2905. (636): here
  2906. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2907. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2908. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2909. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2910. (764): here
  2911. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2912. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2913. (714): here
  2914.  
  2915. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(519): error : name followed by "::" must be a class
  2916. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2917. detected during:
  2918. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2919. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2920. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2921. (636): here
  2922. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2923. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2924. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2925. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2926. (764): here
  2927. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2928. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2929. (714): here
  2930.  
  2931. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(544): error : name followed by "::" must be a class
  2932. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2933. detected during:
  2934. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  2935. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  2936. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2937. (636): here
  2938. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2939. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2940. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2941. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Squareplus, INFERENCE=true]"
  2942. (764): here
  2943. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2944. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2945. (714): here
  2946.  
  2947. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  2948. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2949. detected during:
  2950. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2951. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2952. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2953. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  2954. (765): here
  2955. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2956. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2957. (714): here
  2958.  
  2959. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  2960. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2961. detected during:
  2962. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2963. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2964. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2965. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  2966. (765): here
  2967. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2968. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2969. (714): here
  2970.  
  2971. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  2972. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2973. detected during:
  2974. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2975. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2976. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2977. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  2978. (765): here
  2979. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  2980. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  2981. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(211): error : more than one conversio
  2982. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  2983. a-nn\src\tiny-cuda-nn.vcxproj]
  2984. function "__half::operator float() const"
  2985. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  2986. function "__half::operator short() const"
  2987. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  2988. function "__half::operator unsigned short() const"
  2989. (714): here
  2990. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  2991.  
  2992. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  2993. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  2994. detected during:
  2995. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  2996. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  2997. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  2998. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  2999. (765): here
  3000. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  3001. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  3002. (714): here
  3003.  
  3004. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(515): error : name followed by "::" must be a class
  3005. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3006. detected during:
  3007. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3008. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3009. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  3010. (636): here
  3011. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3012. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3013. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3014. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  3015. (765): here
  3016. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  3017. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  3018. (714): here
  3019.  
  3020. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(516): error : name followed by "::" must be a class
  3021. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3022. detected during:
  3023. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3024. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3025. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  3026. (636): here
  3027. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3028. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3029. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3030. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  3031. (765): here
  3032. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  3033. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  3034. (714): here
  3035.  
  3036. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(517): error : name followed by "::" must be a class
  3037. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3038. detected during:
  3039. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3040. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3041. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  3042. (636): here
  3043. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3044. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3045. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3046. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  3047. (765): here
  3048. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  3049. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  3050. (714): here
  3051.  
  3052. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(519): error : name followed by "::" must be a class
  3053. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3054. detected during:
  3055. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3056. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3057. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  3058. (636): here
  3059. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3060. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3061. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3062. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  3063. (765): here
  3064. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  3065. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  3066. (714): here
  3067.  
  3068. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(544): error : name followed by "::" must be a class
  3069. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3070. detected during:
  3071. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3072. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3073. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  3074. (636): here
  3075. function "__half::operator int() const"
  3076. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3077. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3078. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3079. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Softplus, INFERENCE=true]"
  3080. C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\BuildCustomizations\CUDA 11.6.t
  3081. argets(790,9): error MSB3721: The command ""C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -gen
  3082. code=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft Visual Studio\2
  3083. 019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies" -I"C:\ngp\instant-
  3084. ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\dependencies\glfw\include"
  3085. -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:\ProgramData\NVIDIA Co
  3086. rporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"C:\ngp\instant-ngp\depen
  3087. dencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" --keep-dir
  3088. x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambda --expt-relaxed-constexpr -
  3089. std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX -DTCNN_MIN_GPU_ARCH=75 -DTCNN
  3090. _SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -Xcompiler "/EHsc /W1 /nologo
  3091. /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cuda-nn.pdb /FS /Zi /MD /GR" -o tiny
  3092. -cuda-nn.dir\RelWithDebInfo\encoding.obj "C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\encoding.cu"" exited with co
  3093. de 1. [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3094. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3095. (765): here
  3096. instantiation of "void tcnn::FullyFusedMLP<T, WIDTH>::inference_mixed_precision(cudaStream_t, const tcnn:
  3097. :GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> &, __nv_bool) [with T=tcnn::network_precision_t, WIDTH=128]"
  3098. function "__half::operator unsigned int() const"
  3099. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3100. (714): here
  3101.  
  3102. function "__half::operator long long() const"
  3103. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3104. function "__half::operator unsigned long long() const"
  3105. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3106. function "__half::operator __nv_bool() const"
  3107. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3108. detected during:
  3109. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  3110. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  3111. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  3112. nn::network_precision_t, 8U>>]"
  3113. (269): here
  3114. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  3115. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  3116. (334): here
  3117. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  3118. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  3119. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(274): here
  3120. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3121. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3122. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3123. ne]"
  3124. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3125.  
  3126. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(217): error : more than one conversio
  3127. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3128. a-nn\src\tiny-cuda-nn.vcxproj]
  3129. function "__half::operator float() const"
  3130. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3131. function "__half::operator short() const"
  3132. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3133. function "__half::operator unsigned short() const"
  3134. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3135. function "__half::operator int() const"
  3136. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3137. function "__half::operator unsigned int() const"
  3138. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3139. function "__half::operator long long() const"
  3140. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3141. function "__half::operator unsigned long long() const"
  3142. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3143. function "__half::operator __nv_bool() const"
  3144. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3145. detected during:
  3146. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  3147. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  3148. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  3149. nn::network_precision_t, 8U>>]"
  3150. (269): here
  3151. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  3152. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  3153. (334): here
  3154. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  3155. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  3156. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(274): here
  3157. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3158. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3159. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3160. ne]"
  3161. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3162.  
  3163. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(217): error : more than one conversio
  3164. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3165. a-nn\src\tiny-cuda-nn.vcxproj]
  3166. function "__half::operator float() const"
  3167. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3168. function "__half::operator short() const"
  3169. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3170. function "__half::operator unsigned short() const"
  3171. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3172. function "__half::operator int() const"
  3173. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3174. function "__half::operator unsigned int() const"
  3175. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3176. function "__half::operator long long() const"
  3177. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3178. function "__half::operator unsigned long long() const"
  3179. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3180. function "__half::operator __nv_bool() const"
  3181. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3182. detected during:
  3183. instantiation of "void tcnn::warp_activation_backward<T,fragment_t,forward_fragment_t>(tcnn::Activation,
  3184. const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::Vec
  3185. torFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t<tc
  3186. nn::network_precision_t, 8U>>]"
  3187. (269): here
  3188. instantiation of "void tcnn::kernel_activation_backward_output(uint32_t, tcnn::Activation, const T *, con
  3189. st T *, T *) [with T=tcnn::network_precision_t, N=8U]"
  3190. (334): here
  3191. instantiation of "void tcnn::activation_backward_output_gpu(cudaStream_t, uint32_t, tcnn::Activation, con
  3192. st T *, const T *, T *) [with T=tcnn::network_precision_t]"
  3193. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(274): here
  3194. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3195. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3196. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3197. ne]"
  3198. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3199.  
  3200. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  3201. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3202. detected during:
  3203. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3204. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3205. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3206. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3207. (799): here
  3208. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3209. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3210. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3211. (998): here
  3212.  
  3213. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  3214. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3215. detected during:
  3216. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3217. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3218. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3219. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3220. (799): here
  3221. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3222. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3223. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3224. (998): here
  3225.  
  3226. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  3227. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3228. detected during:
  3229. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3230. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3231. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3232. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3233. (799): here
  3234. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3235. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3236. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3237. (998): here
  3238.  
  3239. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  3240. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3241. detected during:
  3242. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3243. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3244. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3245. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3246. (799): here
  3247. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3248. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3249. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3250. (998): here
  3251.  
  3252. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(515): error : name followed by "::" must be a class
  3253. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3254. detected during:
  3255. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3256. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3257. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3258. (636): here
  3259. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3260. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3261. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3262. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3263. (799): here
  3264. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3265. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3266. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3267. (998): here
  3268.  
  3269. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(516): error : name followed by "::" must be a class
  3270. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3271. detected during:
  3272. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3273. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3274. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3275. (636): here
  3276. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3277. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3278. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3279. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3280. (799): here
  3281. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3282. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3283. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3284. (998): here
  3285.  
  3286. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(517): error : name followed by "::" must be a class
  3287. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3288. detected during:
  3289. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3290. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3291. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3292. (636): here
  3293. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3294. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3295. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3296. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3297. (799): here
  3298. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3299. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3300. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3301. (998): here
  3302.  
  3303. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(519): error : name followed by "::" must be a class
  3304. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3305. detected during:
  3306. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3307. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3308. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3309. (636): here
  3310. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3311. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3312. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3313. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3314. (799): here
  3315. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3316. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3317. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3318. (998): here
  3319.  
  3320. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(544): error : name followed by "::" must be a class
  3321. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3322. detected during:
  3323. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3324. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3325. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3326. (636): here
  3327. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3328. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3329. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3330. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::None, INFERENCE=false]"
  3331. (799): here
  3332. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3333. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3334. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3335. (998): here
  3336.  
  3337. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  3338. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3339. detected during:
  3340. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3341. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3342. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3343. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3344. (800): here
  3345. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3346. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3347. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3348. (998): here
  3349.  
  3350. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  3351. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3352. detected during:
  3353. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3354. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3355. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3356. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3357. (800): here
  3358. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3359. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3360. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3361. (998): here
  3362.  
  3363. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  3364. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3365. detected during:
  3366. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3367. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3368. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3369. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3370. (800): here
  3371. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3372. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3373. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3374. (998): here
  3375.  
  3376. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  3377. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3378. detected during:
  3379. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3380. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3381. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3382. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3383. (800): here
  3384. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3385. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3386. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3387. (998): here
  3388.  
  3389. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(515): error : name followed by "::" must be a class
  3390. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3391. detected during:
  3392. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3393. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3394. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3395. (636): here
  3396. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3397. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3398. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3399. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3400. (800): here
  3401. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3402. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3403. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3404. (998): here
  3405.  
  3406. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(516): error : name followed by "::" must be a class
  3407. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3408. detected during:
  3409. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3410. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3411. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3412. (636): here
  3413. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3414. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3415. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3416. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3417. (800): here
  3418. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3419. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3420. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3421. (998): here
  3422.  
  3423. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(517): error : name followed by "::" must be a class
  3424. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3425. detected during:
  3426. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3427. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3428. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3429. (636): here
  3430. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3431. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3432. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3433. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3434. (800): here
  3435. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3436. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3437. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3438. (998): here
  3439.  
  3440. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(519): error : name followed by "::" must be a class
  3441. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3442. detected during:
  3443. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3444. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3445. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3446. (636): here
  3447. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3448. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3449. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3450. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3451. (800): here
  3452. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3453. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3454. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3455. (998): here
  3456.  
  3457. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(544): error : name followed by "::" must be a class
  3458. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3459. detected during:
  3460. instantiation of "void tcnn::kernel_mlp_fused<WIDTH,N_ITERS,OUT_T,ACTIVATION,INFERENCE>(tcnn::Activation,
  3461. const __half *, const __half *, OUT_T *, OUT_T *, uint32_t, uint32_t, uint32_t, uint32_t, <error-type>, <error-type>
  3462. ) [with WIDTH=128, N_ITERS=8, OUT_T=__half, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3463. (636): here
  3464. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3465. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3466. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3467. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Exponential, INFERENCE=false]"
  3468. (800): here
  3469. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3470. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3471. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3472. (998): here
  3473.  
  3474. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  3475. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3476. detected during:
  3477. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3478. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3479. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3480. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=false]"
  3481. (801): here
  3482. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3483. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3484. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3485. (998): here
  3486.  
  3487. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(634): error : name followed by "::" must be a class
  3488. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3489. detected during:
  3490. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3491. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3492. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3493. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=false]"
  3494. (801): here
  3495. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3496. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3497. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3498. (998): here
  3499.  
  3500. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu(635): error : name followed by "::" must be a class
  3501. or namespace name [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3502. detected during:
  3503. instantiation of "std::enable_if_t<std::is_same<__half, T>::value, void> tcnn::mlp_fused_forward<WIDTH,T,
  3504. ACTIVATION,INFERENCE>(cudaStream_t, tcnn::Activation, const tcnn::GPUMatrix<T, tcnn::MatrixLayout::RowMajor> &, const
  3505. tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrix<T, tcnn::MatrixLayout::ColumnMajor> &, tcnn::GPUMatrixDynamic<T> *, uin
  3506. t32_t) [with WIDTH=128, T=tcnn::network_precision_t, ACTIVATION=tcnn::Activation::Sigmoid, INFERENCE=false]"
  3507. (801): here
  3508. instantiation of "std::unique_ptr<tcnn::Context, std::default_delete<tcnn::Context>> tcnn::FullyFusedMLP<
  3509. T, WIDTH>::forward(cudaStream_t, const tcnn::GPUMatrixDynamic<T> &, tcnn::GPUMatrixDynamic<T> *, __nv_bool, __nv_bool
  3510. ) [with T=tcnn::network_precision_t, WIDTH=128]"
  3511. (998): here
  3512.  
  3513. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(129): error : more than one conversio
  3514. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3515. a-nn\src\tiny-cuda-nn.vcxproj]
  3516. function "__half::operator float() const"
  3517. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3518. function "__half::operator short() const"
  3519. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3520. function "__half::operator unsigned short() const"
  3521. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3522. function "__half::operator int() const"
  3523. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3524. function "__half::operator unsigned int() const"
  3525. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3526. function "__half::operator long long() const"
  3527. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3528. function "__half::operator unsigned long long() const"
  3529. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3530. function "__half::operator __nv_bool() const"
  3531. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3532. detected during:
  3533. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  3534. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  3535. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  3536. <tcnn::network_precision_t, 8U>>]"
  3537. (256): here
  3538. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  3539. T *) [with T=tcnn::network_precision_t, N=8U]"
  3540. (310): here
  3541. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  3542. const T *, T *) [with T=tcnn::network_precision_t]"
  3543. (319): here
  3544. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  3545. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  3546. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(323): here
  3547. Error limit reached.
  3548. 100 errors detected in the compilation of "C:/ngp/instant-ngp/dependencies/tiny-cuda-nn/src/fully_fused_mlp.cu".
  3549. Compilation terminated.
  3550. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3551. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3552. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3553. ne]"
  3554. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3555.  
  3556. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(129): error : more than one conversio
  3557. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3558. a-nn\src\tiny-cuda-nn.vcxproj]
  3559. function "__half::operator float() const"
  3560. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3561. function "__half::operator short() const"
  3562. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3563. function "__half::operator unsigned short() const"
  3564. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3565. function "__half::operator int() const"
  3566. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3567. function "__half::operator unsigned int() const"
  3568. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3569. function "__half::operator long long() const"
  3570. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3571. function "__half::operator unsigned long long() const"
  3572. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3573. function "__half::operator __nv_bool() const"
  3574. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3575. detected during:
  3576. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  3577. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  3578. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  3579. <tcnn::network_precision_t, 8U>>]"
  3580. (256): here
  3581. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  3582. T *) [with T=tcnn::network_precision_t, N=8U]"
  3583. (310): here
  3584. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  3585. const T *, T *) [with T=tcnn::network_precision_t]"
  3586. (319): here
  3587. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  3588. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  3589. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(323): here
  3590. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3591. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3592. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3593. ne]"
  3594. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3595.  
  3596. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(135): error : more than one conversio
  3597. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3598. a-nn\src\tiny-cuda-nn.vcxproj]
  3599. function "__half::operator float() const"
  3600. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3601. function "__half::operator short() const"
  3602. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3603. function "__half::operator unsigned short() const"
  3604. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3605. function "__half::operator int() const"
  3606. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3607. function "__half::operator unsigned int() const"
  3608. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3609. function "__half::operator long long() const"
  3610. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3611. function "__half::operator unsigned long long() const"
  3612. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3613. function "__half::operator __nv_bool() const"
  3614. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3615. detected during:
  3616. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  3617. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  3618. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  3619. <tcnn::network_precision_t, 8U>>]"
  3620. (256): here
  3621. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  3622. T *) [with T=tcnn::network_precision_t, N=8U]"
  3623. (310): here
  3624. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  3625. const T *, T *) [with T=tcnn::network_precision_t]"
  3626. (319): here
  3627. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  3628. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  3629. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(323): here
  3630. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3631. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3632. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3633. ne]"
  3634. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3635.  
  3636. fully_fused_mlp.cu
  3637. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(135): error : more than one conversio
  3638. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3639. a-nn\src\tiny-cuda-nn.vcxproj]
  3640. C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\BuildCustomizations\CUDA 11.6.t
  3641. argets(790,9): error MSB3721: The command ""C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -gen
  3642. code=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft Visual Studio\2
  3643. 019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies" -I"C:\ngp\instant-
  3644. ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\dependencies\glfw\include"
  3645. -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:\ProgramData\NVIDIA Co
  3646. rporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"C:\ngp\instant-ngp\depen
  3647. dencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" --keep-dir
  3648. x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambda --expt-relaxed-constexpr -
  3649. std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX -DTCNN_MIN_GPU_ARCH=75 -DTCNN
  3650. _SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -Xcompiler "/EHsc /W1 /nologo
  3651. /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cuda-nn.pdb /FS /Zi /MD /GR" -o tiny
  3652. -cuda-nn.dir\RelWithDebInfo\fully_fused_mlp.obj "C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\fully_fused_mlp.cu""
  3653. exited with code 1. [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  3654. function "__half::operator float() const"
  3655. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3656. function "__half::operator short() const"
  3657. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3658. function "__half::operator unsigned short() const"
  3659. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3660. function "__half::operator int() const"
  3661. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3662. function "__half::operator unsigned int() const"
  3663. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3664. function "__half::operator long long() const"
  3665. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3666. function "__half::operator unsigned long long() const"
  3667. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3668. function "__half::operator __nv_bool() const"
  3669. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3670. detected during:
  3671. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  3672. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  3673. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  3674. <tcnn::network_precision_t, 8U>>]"
  3675. (256): here
  3676. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  3677. T *) [with T=tcnn::network_precision_t, N=8U]"
  3678. (310): here
  3679. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  3680. const T *, T *) [with T=tcnn::network_precision_t]"
  3681. (319): here
  3682. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  3683. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  3684. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(323): here
  3685. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3686. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3687. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3688. ne]"
  3689. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3690.  
  3691. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(141): error : more than one conversio
  3692. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3693. a-nn\src\tiny-cuda-nn.vcxproj]
  3694. function "__half::operator float() const"
  3695. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3696. function "__half::operator short() const"
  3697. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3698. function "__half::operator unsigned short() const"
  3699. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3700. function "__half::operator int() const"
  3701. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3702. function "__half::operator unsigned int() const"
  3703. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3704. function "__half::operator long long() const"
  3705. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3706. function "__half::operator unsigned long long() const"
  3707. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3708. function "__half::operator __nv_bool() const"
  3709. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3710. detected during:
  3711. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  3712. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  3713. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  3714. <tcnn::network_precision_t, 8U>>]"
  3715. (256): here
  3716. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  3717. T *) [with T=tcnn::network_precision_t, N=8U]"
  3718. (310): here
  3719. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  3720. const T *, T *) [with T=tcnn::network_precision_t]"
  3721. (319): here
  3722. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  3723. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  3724. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(323): here
  3725. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3726. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3727. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3728. ne]"
  3729. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3730.  
  3731. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(141): error : more than one conversio
  3732. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3733. a-nn\src\tiny-cuda-nn.vcxproj]
  3734. function "__half::operator float() const"
  3735. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3736. function "__half::operator short() const"
  3737. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3738. function "__half::operator unsigned short() const"
  3739. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3740. function "__half::operator int() const"
  3741. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3742. function "__half::operator unsigned int() const"
  3743. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3744. function "__half::operator long long() const"
  3745. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3746. function "__half::operator unsigned long long() const"
  3747. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3748. function "__half::operator __nv_bool() const"
  3749. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3750. detected during:
  3751. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  3752. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  3753. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  3754. <tcnn::network_precision_t, 8U>>]"
  3755. (256): here
  3756. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  3757. T *) [with T=tcnn::network_precision_t, N=8U]"
  3758. (310): here
  3759. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  3760. const T *, T *) [with T=tcnn::network_precision_t]"
  3761. (319): here
  3762. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  3763. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  3764. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(323): here
  3765. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3766. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3767. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3768. ne]"
  3769. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3770.  
  3771. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(148): error : more than one conversio
  3772. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3773. a-nn\src\tiny-cuda-nn.vcxproj]
  3774. function "__half::operator float() const"
  3775. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3776. function "__half::operator short() const"
  3777. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3778. function "__half::operator unsigned short() const"
  3779. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3780. function "__half::operator int() const"
  3781. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3782. function "__half::operator unsigned int() const"
  3783. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3784. function "__half::operator long long() const"
  3785. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3786. function "__half::operator unsigned long long() const"
  3787. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3788. function "__half::operator __nv_bool() const"
  3789. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3790. detected during:
  3791. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  3792. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  3793. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  3794. <tcnn::network_precision_t, 8U>>]"
  3795. (256): here
  3796. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  3797. T *) [with T=tcnn::network_precision_t, N=8U]"
  3798. (310): here
  3799. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  3800. const T *, T *) [with T=tcnn::network_precision_t]"
  3801. (319): here
  3802. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  3803. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  3804. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(323): here
  3805. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3806. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3807. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3808. ne]"
  3809. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3810.  
  3811. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(148): error : more than one conversio
  3812. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3813. a-nn\src\tiny-cuda-nn.vcxproj]
  3814. function "__half::operator float() const"
  3815. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3816. function "__half::operator short() const"
  3817. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3818. function "__half::operator unsigned short() const"
  3819. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3820. function "__half::operator int() const"
  3821. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3822. function "__half::operator unsigned int() const"
  3823. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3824. function "__half::operator long long() const"
  3825. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3826. function "__half::operator unsigned long long() const"
  3827. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3828. function "__half::operator __nv_bool() const"
  3829. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3830. detected during:
  3831. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  3832. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  3833. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  3834. <tcnn::network_precision_t, 8U>>]"
  3835. (256): here
  3836. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  3837. T *) [with T=tcnn::network_precision_t, N=8U]"
  3838. (310): here
  3839. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  3840. const T *, T *) [with T=tcnn::network_precision_t]"
  3841. (319): here
  3842. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  3843. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  3844. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(323): here
  3845. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3846. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3847. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3848. ne]"
  3849. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3850.  
  3851. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(156): error : more than one conversio
  3852. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3853. a-nn\src\tiny-cuda-nn.vcxproj]
  3854. function "__half::operator float() const"
  3855. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3856. function "__half::operator short() const"
  3857. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3858. function "__half::operator unsigned short() const"
  3859. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3860. function "__half::operator int() const"
  3861. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3862. function "__half::operator unsigned int() const"
  3863. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3864. function "__half::operator long long() const"
  3865. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3866. function "__half::operator unsigned long long() const"
  3867. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3868. function "__half::operator __nv_bool() const"
  3869. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3870. detected during:
  3871. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  3872. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  3873. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  3874. <tcnn::network_precision_t, 8U>>]"
  3875. (256): here
  3876. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  3877. T *) [with T=tcnn::network_precision_t, N=8U]"
  3878. (310): here
  3879. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  3880. const T *, T *) [with T=tcnn::network_precision_t]"
  3881. (319): here
  3882. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  3883. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  3884. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(323): here
  3885. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3886. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3887. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3888. ne]"
  3889. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3890.  
  3891. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(156): error : more than one conversio
  3892. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3893. a-nn\src\tiny-cuda-nn.vcxproj]
  3894. function "__half::operator float() const"
  3895. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3896. function "__half::operator short() const"
  3897. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3898. function "__half::operator unsigned short() const"
  3899. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3900. function "__half::operator int() const"
  3901. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3902. function "__half::operator unsigned int() const"
  3903. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3904. function "__half::operator long long() const"
  3905. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3906. function "__half::operator unsigned long long() const"
  3907. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3908. function "__half::operator __nv_bool() const"
  3909. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3910. detected during:
  3911. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  3912. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  3913. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  3914. <tcnn::network_precision_t, 8U>>]"
  3915. (256): here
  3916. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  3917. T *) [with T=tcnn::network_precision_t, N=8U]"
  3918. (310): here
  3919. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  3920. const T *, T *) [with T=tcnn::network_precision_t]"
  3921. (319): here
  3922. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  3923. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  3924. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(323): here
  3925. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3926. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3927. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3928. ne]"
  3929. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3930.  
  3931. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(163): error : more than one conversio
  3932. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3933. a-nn\src\tiny-cuda-nn.vcxproj]
  3934. function "__half::operator float() const"
  3935. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3936. function "__half::operator short() const"
  3937. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3938. function "__half::operator unsigned short() const"
  3939. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3940. function "__half::operator int() const"
  3941. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3942. function "__half::operator unsigned int() const"
  3943. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3944. function "__half::operator long long() const"
  3945. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3946. function "__half::operator unsigned long long() const"
  3947. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3948. function "__half::operator __nv_bool() const"
  3949. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3950. detected during:
  3951. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  3952. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  3953. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  3954. <tcnn::network_precision_t, 8U>>]"
  3955. (256): here
  3956. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  3957. T *) [with T=tcnn::network_precision_t, N=8U]"
  3958. (310): here
  3959. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  3960. const T *, T *) [with T=tcnn::network_precision_t]"
  3961. (319): here
  3962. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  3963. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  3964. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(323): here
  3965. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  3966. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  3967. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  3968. ne]"
  3969. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  3970.  
  3971. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include\tiny-cuda-nn/common_device.h(163): error : more than one conversio
  3972. n function from "tcnn::network_precision_t" to a built-in type applies: [C:\ngp\instant-ngp\build\dependencies\tiny-cud
  3973. a-nn\src\tiny-cuda-nn.vcxproj]
  3974. function "__half::operator float() const"
  3975. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(204): here
  3976. function "__half::operator short() const"
  3977. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(222): here
  3978. function "__half::operator unsigned short() const"
  3979. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(225): here
  3980. function "__half::operator int() const"
  3981. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(228): here
  3982. function "__half::operator unsigned int() const"
  3983. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(231): here
  3984. function "__half::operator long long() const"
  3985. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(234): here
  3986. function "__half::operator unsigned long long() const"
  3987. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(237): here
  3988. function "__half::operator __nv_bool() const"
  3989. C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include\cuda_fp16.hpp(241): here
  3990. detected during:
  3991. instantiation of "void tcnn::warp_activation_backward_in<T,fragment_t,forward_fragment_t>(tcnn::Activatio
  3992. n, const fragment_t &, const forward_fragment_t &, fragment_t &) [with T=tcnn::network_precision_t, fragment_t=tcnn::
  3993. VectorFragment<tcnn::vector_t<tcnn::network_precision_t, 8U>>, forward_fragment_t=tcnn::VectorFragment<tcnn::vector_t
  3994. <tcnn::network_precision_t, 8U>>]"
  3995. (256): here
  3996. instantiation of "void tcnn::kernel_activation_backward(uint32_t, tcnn::Activation, const T *, const T *,
  3997. T *) [with T=tcnn::network_precision_t, N=8U]"
  3998. (310): here
  3999. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, uint32_t, tcnn::Activation, const T *,
  4000. const T *, T *) [with T=tcnn::network_precision_t]"
  4001. (319): here
  4002. instantiation of "void tcnn::activation_backward_gpu(cudaStream_t, tcnn::Activation, const tcnn::GPUMatri
  4003. xDynamic<T> &, tcnn::GPUMatrixDynamic<T> &) [with T=tcnn::network_precision_t]"
  4004. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(323): here
  4005. instantiation of "void tcnn::CutlassResNet<T, input_activation>::backward(cudaStream_t, const tcnn::Conte
  4006. xt &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, const tcnn::GPUMatrixDynamic<T> &, tcnn::
  4007. GPUMatrixDynamic<T> *, __nv_bool, __nv_bool) [with T=tcnn::network_precision_t, input_activation=tcnn::Activation::No
  4008. ne]"
  4009. C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu(391): here
  4010.  
  4011. 26 errors detected in the compilation of "C:/ngp/instant-ngp/dependencies/tiny-cuda-nn/src/cutlass_resnet.cu".
  4012. cutlass_resnet.cu
  4013. C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\MSBuild\Microsoft\VC\v160\BuildCustomizations\CUDA 11.6.t
  4014. argets(790,9): error MSB3721: The command ""C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\bin\nvcc.exe" -gen
  4015. code=arch=compute_52,code=\"sm_52,compute_52\" --use-local-env -ccbin "C:\Program Files (x86)\Microsoft Visual Studio\2
  4016. 019\Community\VC\Tools\MSVC\14.29.30133\bin\HostX64\x64" -x cu -I"C:\ngp\instant-ngp\dependencies" -I"C:\ngp\instant-
  4017. ngp\dependencies\eigen" -I"C:\ngp\instant-ngp\dependencies\filesystem" -I"C:\ngp\instant-ngp\dependencies\glfw\include"
  4018. -I"C:\ngp\instant-ngp\dependencies\imgui\gl3w" -I"C:\ngp\instant-ngp\dependencies\nanovdb" -I"C:\ProgramData\NVIDIA Co
  4019. rporation\OptiX SDK 7.4.0\include" -I"C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\include" -I"C:\ngp\instant-ngp\depen
  4020. dencies\tiny-cuda-nn\dependencies" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.6\include" --keep-dir
  4021. x64\RelWithDebInfo -maxrregcount=0 --machine 64 --compile -cudart static --extended-lambda --expt-relaxed-constexpr -
  4022. std=c++14 -Xcompiler="/EHsc -Zi -Ob1 -bigobj" -D_WINDOWS -DNDEBUG -DNGP_GUI -DNGP_OPTIX -DTCNN_MIN_GPU_ARCH=75 -DTCNN
  4023. _SHAMPOO -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -D_MBCS -D"CMAKE_INTDIR=\"RelWithDebInfo\"" -Xcompiler "/EHsc /W1 /nologo
  4024. /O2 /FdC:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\RelWithDebInfo\tiny-cuda-nn.pdb /FS /Zi /MD /GR" -o tiny
  4025. -cuda-nn.dir\RelWithDebInfo\cutlass_resnet.obj "C:\ngp\instant-ngp\dependencies\tiny-cuda-nn\src\cutlass_resnet.cu"" ex
  4026. ited with code 1. [C:\ngp\instant-ngp\build\dependencies\tiny-cuda-nn\src\tiny-cuda-nn.vcxproj]
  4027. object.cu
  4028. reduce_sum.cu
  4029. network.cu
  4030. loss.cu
  4031. optimizer.cu
  4032. PS C:\ngp\instant-ngp>
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!