NASP/cuda-samples
84
0
Fork 0
mirror of https://github.com/NVIDIA/cuda-samples.git synced 2026-10-13 00:08:26 +08:00
Code Issues Packages Projects Releases Wiki Activity

CUDA 13.4 samples update - v13.4-public

Browse Source 
Release 13.4 of the CUDA Samples supported by CUDA Toolkit 13.4.
See Changelog for more information.
This commit is contained in:
Dheemanth
2026-09-09 15:07:08 -07:00
committed by GitHub
parent b7c5481c55
commit 5443602d89
871 changed files with 6843 additions and 321849 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
cpp/2_Concepts_and_Techniques/CMakeLists.txt
View File

@@ -14,7 +14,6 @@ add_subdirectory(eigenvalues)
add_subdirectory(histogram)
add_subdirectory(imageDenoising)
add_subdirectory(inlinePTX)
add_subdirectory(inlinePTX_nvrtc)
add_subdirectory(interval)
add_subdirectory(particles)
add_subdirectory(radixSortThrust)

18
cpp/2_Concepts_and_Techniques/EGLStream_CUDA_CrossGPU/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/EGLStream_CUDA_CrossGPU/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

56
cpp/2_Concepts_and_Techniques/EGLStream_CUDA_CrossGPU/CMakeLists.txt
View File

@@ -1,54 +1,40 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(EGLStream_CUDA_CrossGPU LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
find_package(EGL)
if(CMAKE_SYSTEM_NAME STREQUAL "Linux")
# Source file
if(${EGL_FOUND})
# Add target for EGLStream_CUDA_CrossGPU
add_executable(EGLStream_CUDA_CrossGPU cuda_consumer.cpp cuda_producer.cpp eglstrm_common.cpp kernel.cu main.cpp)
target_compile_options(EGLStream_CUDA_CrossGPU PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(EGLStream_CUDA_CrossGPU PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(EGLStream_CUDA_CrossGPU PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(EGLStream_CUDA_CrossGPU PUBLIC
${EGL_INCLUDE_DIR}
${CUDAToolkit_INCLUDE_DIRS}
)
target_link_libraries(EGLStream_CUDA_CrossGPU
${EGL_LIBRARY}
CUDA::cuda_driver
)
add_executable(EGLStream_CUDA_CrossGPU cuda_consumer.cpp cuda_producer.cpp eglstrm_common.cpp kernel.cu main.cpp)
target_compile_options(EGLStream_CUDA_CrossGPU PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(EGLStream_CUDA_CrossGPU PRIVATE cxx_std_17 cuda_std_17)
target_include_directories(EGLStream_CUDA_CrossGPU PUBLIC
${EGL_INCLUDE_DIR}
${CUDAToolkit_INCLUDE_DIRS}
)
target_link_libraries(EGLStream_CUDA_CrossGPU
${EGL_LIBRARY}
CUDA::cuda_driver
)
include(InstallSamples)
setup_samples_install()
else()
message(STATUS "EGL not found - will not build sample 'EGLStream_CUDA_CrossGPU'")
endif()
else()
message(STATUS "Will not build sample EGLStream_CUDA_CrossGPU - requires Linux OS")
endif()
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
setup_samples_install()

24
cpp/2_Concepts_and_Techniques/EGLStream_CUDA_CrossGPU/eglstrm_common.cpp
View File

@@ -114,7 +114,7 @@ int EGLStreamInit(bool isCrossDevice, int isConsumer, EGLNativeFileDescriptorKHR
if (egl_device_id >= numDevices) {
printf("No CUDA Capable EGL Device found.. Waiving execution\n");
goto Done;
exit(2);
}
g_consumerEglDisplay = eglGetPlatformDisplayEXT(EGL_PLATFORM_DEVICE_EXT, (void *)devices[egl_device_id], NULL);
@@ -126,6 +126,16 @@ int EGLStreamInit(bool isCrossDevice, int isConsumer, EGLNativeFileDescriptorKHR
eglStatus = eglInitialize(g_consumerEglDisplay, 0, 0);
if (!eglStatus) {
// A GPU built without graphics support is CUDA capable and is offered
// as an EGL device, but has no DRM device for EGL to reach it through.
// Any other reason is a real failure.
if (eglQueryDeviceStringEXT(devices[egl_device_id], EGL_DRM_DEVICE_FILE_EXT) == NULL
&& eglQueryDeviceStringEXT(devices[egl_device_id], EGL_DRM_RENDER_NODE_FILE_EXT) == NULL) {
printf("This GPU has no graphics support, EGL is unavailable.. Waiving execution\n");
// 2 reports an unmet hardware requirement, which run_tests.py does
// not count as a failure.
exit(2);
}
printf("EGL failed to initialize. \n");
eglStatus = EGL_FALSE;
goto Done;
@@ -191,7 +201,7 @@ int EGLStreamInit(bool isCrossDevice, int isConsumer, EGLNativeFileDescriptorKHR
if (egl_device_id >= numDevices) {
printf("No CUDA Capable EGL Device found.. Waiving execution\n");
goto Done;
exit(2);
}
g_producerEglDisplay = eglGetPlatformDisplayEXT(EGL_PLATFORM_DEVICE_EXT, (void *)devices[egl_device_id], NULL);
@@ -203,6 +213,16 @@ int EGLStreamInit(bool isCrossDevice, int isConsumer, EGLNativeFileDescriptorKHR
eglStatus = eglInitialize(g_producerEglDisplay, 0, 0);
if (!eglStatus) {
// A GPU built without graphics support is CUDA capable and is offered
// as an EGL device, but has no DRM device for EGL to reach it through.
// Any other reason is a real failure.
if (eglQueryDeviceStringEXT(devices[egl_device_id], EGL_DRM_DEVICE_FILE_EXT) == NULL
&& eglQueryDeviceStringEXT(devices[egl_device_id], EGL_DRM_RENDER_NODE_FILE_EXT) == NULL) {
printf("This GPU has no graphics support, EGL is unavailable.. Waiving execution\n");
// 2 reports an unmet hardware requirement, which run_tests.py does
// not count as a failure.
exit(2);
}
printf("EGL failed to initialize. \n");
eglStatus = EGL_FALSE;
goto Done;

1
cpp/2_Concepts_and_Techniques/EGLStream_CUDA_CrossGPU/eglstrm_common.h
View File

@@ -71,6 +71,7 @@ extern bool verbose;
T(PFNEGLGETPLATFORMDISPLAYEXTPROC, eglGetPlatformDisplayEXT) \
T(PFNEGLGETSTREAMFILEDESCRIPTORKHRPROC, eglGetStreamFileDescriptorKHR) \
T(PFNEGLQUERYDEVICEATTRIBEXTPROC, eglQueryDeviceAttribEXT) \
T(PFNEGLQUERYDEVICESTRINGEXTPROC, eglQueryDeviceStringEXT) \
T(PFNEGLCREATESTREAMFROMFILEDESCRIPTORKHRPROC, eglCreateStreamFromFileDescriptorKHR)
#define EXTLST_DECL(tx, x) tx x = NULL;

18
cpp/2_Concepts_and_Techniques/EGLStream_CUDA_Interop/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/EGLStream_CUDA_Interop/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

35
cpp/2_Concepts_and_Techniques/EGLStream_CUDA_Interop/CMakeLists.txt
View File

@@ -1,55 +1,40 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(EGLStream_CUDA_Interop LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
find_package(EGL)
if(CMAKE_SYSTEM_NAME STREQUAL "Linux")
# Source file
if(${EGL_FOUND})
# Add target for EGLStream_CUDA_Interop
add_executable(EGLStream_CUDA_Interop cuda_consumer.cpp cuda_producer.cpp eglstrm_common.cpp main.cpp)
target_compile_options(EGLStream_CUDA_Interop PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(EGLStream_CUDA_Interop PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(EGLStream_CUDA_Interop PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(EGLStream_CUDA_Interop PUBLIC
${EGL_INCLUDE_DIR}
${CUDAToolkit_INCLUDE_DIRS}
)
target_link_libraries(EGLStream_CUDA_Interop
${EGL_LIBRARY}
CUDA::cuda_driver
)
include(InstallSamples)
setup_samples_install()
else()
message(STATUS "EGL not found - will not build sample 'EGLStream_CUDA_Interop'")
endif()
else()
message(STATUS "Will not build sample EGLStream_CUDA_Interop - requires Linux OS")
endif()
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
setup_samples_install()

8
cpp/2_Concepts_and_Techniques/EGLStream_CUDA_Interop/eglstrm_common.cpp
View File

@@ -102,6 +102,14 @@ int EGLStreamInit(int *cuda_device)
eglStatus = eglInitialize(g_display, 0, 0);
if (!eglStatus) {
// A GPU built without graphics support is CUDA capable and is offered as an
// EGL device, but has no DRM device for EGL to reach it through. Any other
// reason for EGL not to initialize is a real failure.
if (eglQueryDeviceStringEXT(devices[egl_device_id], EGL_DRM_DEVICE_FILE_EXT) == NULL
&& eglQueryDeviceStringEXT(devices[egl_device_id], EGL_DRM_RENDER_NODE_FILE_EXT) == NULL) {
printf("This GPU has no graphics support, EGL is unavailable.. Waiving execution\n");
exit(EXIT_WAIVED);
}
printf("EGL failed to initialize. \n");
eglStatus = EGL_FALSE;
exit(EXIT_FAILURE);

2
cpp/2_Concepts_and_Techniques/EGLStream_CUDA_Interop/eglstrm_common.h
View File

@@ -59,6 +59,7 @@
T(PFNEGLQUERYDEVICESEXTPROC, eglQueryDevicesEXT) \
T(PFNEGLGETPLATFORMDISPLAYEXTPROC, eglGetPlatformDisplayEXT) \
T(PFNEGLQUERYDEVICEATTRIBEXTPROC, eglQueryDeviceAttribEXT) \
T(PFNEGLQUERYDEVICESTRINGEXTPROC, eglQueryDeviceStringEXT) \
T(PFNEGLCREATESTREAMFROMFILEDESCRIPTORKHRPROC, eglCreateStreamFromFileDescriptorKHR)
#define eglCreateStreamKHR my_eglCreateStreamKHR
@@ -75,6 +76,7 @@
#define eglQueryDevicesEXT my_eglQueryDevicesEXT
#define eglGetPlatformDisplayEXT my_eglGetPlatformDisplayEXT
#define eglQueryDeviceAttribEXT my_eglQueryDeviceAttribEXT
#define eglQueryDeviceStringEXT my_eglQueryDeviceStringEXT
#define EXTLST_DECL(tx, x) tx my_##x = NULL;
#define EXTLST_EXTERN(tx, x) extern tx my_##x;

18
cpp/2_Concepts_and_Techniques/FunctionPointers/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/FunctionPointers/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

39
cpp/2_Concepts_and_Techniques/FunctionPointers/CMakeLists.txt
View File

@@ -1,22 +1,18 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(FunctionPointers LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
if(WIN32)
@@ -33,49 +29,37 @@ endif()
find_package(OpenGL)
find_package(GLUT)
# Source file
if(${OpenGL_FOUND})
if (${GLUT_FOUND})
# Add target for FunctionPointers
add_executable(FunctionPointers FunctionPointers.cpp FunctionPointers_kernels.cu)
target_compile_options(FunctionPointers PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(FunctionPointers PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(FunctionPointers PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(FunctionPointers PUBLIC
${OPENGL_INCLUDE_DIR}
${CUDAToolkit_INCLUDE_DIRS}
${GLUT_INCLUDE_DIRS}
)
target_link_libraries(FunctionPointers
${OPENGL_LIBRARIES}
${GLUT_LIBRARIES}
)
# Copy data files to output directory
add_custom_command(TARGET FunctionPointers POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy_directory
${CMAKE_CURRENT_SOURCE_DIR}/data
${CMAKE_CURRENT_BINARY_DIR}/data
)
if(WIN32)
target_link_libraries(FunctionPointers
${PC_GLUT_LIBRARY_DIRS}/freeglut.lib
${PC_GLUT_LIBRARY_DIRS}/${GLEW_LIB_NAME}.lib
)
add_custom_command(TARGET FunctionPointers
POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy
${CMAKE_CURRENT_SOURCE_DIR}/../../../bin/win64/$<CONFIGURATION>/freeglut.dll
${CMAKE_CURRENT_BINARY_DIR}/$<CONFIGURATION>
)
add_custom_command(TARGET FunctionPointers
POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy
@@ -83,14 +67,11 @@ if(${OpenGL_FOUND})
${CMAKE_CURRENT_BINARY_DIR}/$<CONFIGURATION>
)
endif()
include(InstallSamples)
setup_samples_install()
else()
message(STATUS "GLUT not found - will not build sample 'FunctionPointers'")
endif()
else()
message(STATUS "OpenGL not found - will not build sample 'FunctionPointers'")
endif()
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/MC_EstimatePiInlineP/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/MC_EstimatePiInlineP/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/MC_EstimatePiInlineP/CMakeLists.txt
View File

@@ -1,39 +1,30 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(MC_EstimatePiInlineP LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for MC_EstimatePiInlineP
add_executable(MC_EstimatePiInlineP src/main.cpp src/piestimator.cu src/test.cpp)
target_compile_options(MC_EstimatePiInlineP PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(MC_EstimatePiInlineP PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(MC_EstimatePiInlineP PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(MC_EstimatePiInlineP PUBLIC
${CMAKE_CURRENT_SOURCE_DIR}/inc
${CUDAToolkit_INCLUDE_DIRS}
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/MC_EstimatePiInlineQ/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/MC_EstimatePiInlineQ/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/MC_EstimatePiInlineQ/CMakeLists.txt
View File

@@ -1,34 +1,26 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(MC_EstimatePiInlineQ LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for MC_EstimatePiInlineQ
add_executable(MC_EstimatePiInlineQ src/main.cpp src/piestimator.cu src/test.cpp)
target_compile_options(MC_EstimatePiInlineQ PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(MC_EstimatePiInlineQ PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(MC_EstimatePiInlineQ PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(MC_EstimatePiInlineQ PUBLIC
${CMAKE_CURRENT_SOURCE_DIR}/inc
${CUDAToolkit_INCLUDE_DIRS}
@@ -38,6 +30,5 @@ target_link_libraries(MC_EstimatePiInlineQ PUBLIC
CUDA::curand
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/MC_EstimatePiP/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/MC_EstimatePiP/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/MC_EstimatePiP/CMakeLists.txt
View File

@@ -1,34 +1,26 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(MC_EstimatePiP LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for MC_EstimatePiP
add_executable(MC_EstimatePiP src/main.cpp src/piestimator.cu src/test.cpp)
target_compile_options(MC_EstimatePiP PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(MC_EstimatePiP PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(MC_EstimatePiP PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(MC_EstimatePiP PUBLIC
${CMAKE_CURRENT_SOURCE_DIR}/inc
${CUDAToolkit_INCLUDE_DIRS}
@@ -38,6 +30,5 @@ target_link_libraries(MC_EstimatePiP PUBLIC
CUDA::curand
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/MC_EstimatePiQ/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/MC_EstimatePiQ/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/MC_EstimatePiQ/CMakeLists.txt
View File

@@ -1,34 +1,26 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(MC_EstimatePiQ LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for MC_EstimatePiQ
add_executable(MC_EstimatePiQ src/main.cpp src/piestimator.cu src/test.cpp)
target_compile_options(MC_EstimatePiQ PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(MC_EstimatePiQ PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(MC_EstimatePiQ PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(MC_EstimatePiQ PUBLIC
${CMAKE_CURRENT_SOURCE_DIR}/inc
${CUDAToolkit_INCLUDE_DIRS}
@@ -38,6 +30,5 @@ target_link_libraries(MC_EstimatePiQ PUBLIC
CUDA::curand
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/MC_SingleAsianOptionP/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/MC_SingleAsianOptionP/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/MC_SingleAsianOptionP/CMakeLists.txt
View File

@@ -1,39 +1,30 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(MC_SingleAsianOptionP LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for MC_SingleAsianOptionP
add_executable(MC_SingleAsianOptionP src/main.cpp src/pricingengine.cu src/test.cpp)
target_compile_options(MC_SingleAsianOptionP PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(MC_SingleAsianOptionP PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(MC_SingleAsianOptionP PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(MC_SingleAsianOptionP PUBLIC
${CMAKE_CURRENT_SOURCE_DIR}/inc
${CUDAToolkit_INCLUDE_DIRS}
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

3
cpp/2_Concepts_and_Techniques/README.md
View File

@@ -34,9 +34,6 @@ This sample demonstrates two adaptive image denoising techniques: KNN and NLM, b
### [inlinePTX](./inlinePTX)
A simple test application that demonstrates a new CUDA 4.0 ability to embed PTX in a CUDA kernel.
### [inlinePTX_nvrtc](./inlinePTX_nvrtc)
A simple test application that demonstrates a new CUDA 4.0 ability to embed PTX in a CUDA kernel.
### [interval](./interval)
Interval arithmetic operators example. Uses various C++ features (templates and recursion). The recursive mode requires Compute SM 2.0 capabilities.

18
cpp/2_Concepts_and_Techniques/boxFilter/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/boxFilter/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

39
cpp/2_Concepts_and_Techniques/boxFilter/CMakeLists.txt
View File

@@ -1,22 +1,18 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(boxFilter LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
if(WIN32)
@@ -33,49 +29,37 @@ endif()
find_package(OpenGL)
find_package(GLUT)
# Source file
if(${OpenGL_FOUND})
if (${GLUT_FOUND})
# Add target for boxFilter
add_executable(boxFilter boxFilter.cpp boxFilter_cpu.cpp boxFilter_kernel.cu)
target_compile_options(boxFilter PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(boxFilter PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(boxFilter PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(boxFilter PUBLIC
${OPENGL_INCLUDE_DIR}
${CUDAToolkit_INCLUDE_DIRS}
${GLUT_INCLUDE_DIRS}
)
target_link_libraries(boxFilter
${OPENGL_LIBRARIES}
${GLUT_LIBRARIES}
)
# Copy data files to output directory
add_custom_command(TARGET boxFilter POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy_directory
${CMAKE_CURRENT_SOURCE_DIR}/data
${CMAKE_CURRENT_BINARY_DIR}/data
)
if(WIN32)
target_link_libraries(boxFilter
${PC_GLUT_LIBRARY_DIRS}/freeglut.lib
${PC_GLUT_LIBRARY_DIRS}/${GLEW_LIB_NAME}.lib
)
add_custom_command(TARGET boxFilter
POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy
${CMAKE_CURRENT_SOURCE_DIR}/../../../bin/win64/$<CONFIGURATION>/freeglut.dll
${CMAKE_CURRENT_BINARY_DIR}/$<CONFIGURATION>
)
add_custom_command(TARGET boxFilter
POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy
@@ -83,14 +67,11 @@ if(${OpenGL_FOUND})
${CMAKE_CURRENT_BINARY_DIR}/$<CONFIGURATION>
)
endif()
include(InstallSamples)
setup_samples_install()
else()
message(STATUS "GLUT not found - will not build sample 'boxFilter'")
endif()
else()
message(STATUS "OpenGL not found - will not build sample 'boxFilter'")
endif()
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/convolutionSeparable/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/convolutionSeparable/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/convolutionSeparable/CMakeLists.txt
View File

@@ -1,38 +1,29 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(convolutionSeparable LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for convolutionSeparable
add_executable(convolutionSeparable convolutionSeparable.cu convolutionSeparable_gold.cpp main.cpp)
target_compile_options(convolutionSeparable PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(convolutionSeparable PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(convolutionSeparable PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(convolutionSeparable PUBLIC
${CUDAToolkit_INCLUDE_DIRS}
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/convolutionTexture/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/convolutionTexture/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/convolutionTexture/CMakeLists.txt
View File

@@ -1,38 +1,29 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(convolutionTexture LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for convolutionTexture
add_executable(convolutionTexture convolutionTexture.cu convolutionTexture_gold.cpp main.cpp)
target_compile_options(convolutionTexture PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(convolutionTexture PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(convolutionTexture PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(convolutionTexture PUBLIC
${CUDAToolkit_INCLUDE_DIRS}
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/dct8x8/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/dct8x8/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/dct8x8/CMakeLists.txt
View File

@@ -1,34 +1,26 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(dct8x8 LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for dct8x8
add_executable(dct8x8 dct8x8.cu BmpUtil.cpp DCT8x8_Gold.cpp)
target_compile_options(dct8x8 PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(dct8x8 PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(dct8x8 PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(dct8x8 PUBLIC
${CUDAToolkit_INCLUDE_DIRS}
)
@@ -54,6 +46,5 @@ add_custom_command(TARGET dct8x8 POST_BUILD
${CMAKE_CURRENT_BINARY_DIR}/
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/eigenvalues/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/eigenvalues/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/eigenvalues/CMakeLists.txt
View File

@@ -1,34 +1,26 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(eigenvalues LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for eigenvalues
add_executable(eigenvalues bisect_large.cu bisect_small.cu bisect_util.cu gerschgorin.cpp main.cu matlab.cpp)
target_compile_options(eigenvalues PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(eigenvalues PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(eigenvalues PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
# Copy data files to output directory
add_custom_command(TARGET eigenvalues POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy_if_different
@@ -50,6 +42,5 @@ add_custom_command(TARGET eigenvalues POST_BUILD
${CMAKE_CURRENT_BINARY_DIR}/
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/histogram/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/histogram/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/histogram/CMakeLists.txt
View File

@@ -1,38 +1,29 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(histogram LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for histogram
add_executable(histogram histogram256.cu histogram64.cu histogram_gold.cpp main.cpp)
target_compile_options(histogram PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(histogram PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(histogram PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(histogram PUBLIC
${CUDAToolkit_INCLUDE_DIRS}
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/imageDenoising/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/imageDenoising/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

40
cpp/2_Concepts_and_Techniques/imageDenoising/CMakeLists.txt
View File

@@ -1,22 +1,18 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(imageDenoising LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
if(WIN32)
@@ -30,53 +26,40 @@ if(WIN32)
endif()
endif()
find_package(OpenGL)
find_package(GLUT)
# Source file
if(${OpenGL_FOUND})
if (${GLUT_FOUND})
# Add target for imageDenoising
add_executable(imageDenoising bmploader.cpp imageDenoising.cu imageDenoisingGL.cpp)
target_compile_options(imageDenoising PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(imageDenoising PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(imageDenoising PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(imageDenoising PUBLIC
${OPENGL_INCLUDE_DIR}
${CUDAToolkit_INCLUDE_DIRS}
${GLUT_INCLUDE_DIRS}
)
target_link_libraries(imageDenoising
${OPENGL_LIBRARIES}
${GLUT_LIBRARIES}
)
# Copy data files to the output directory
add_custom_command(TARGET imageDenoising POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy_directory
${CMAKE_CURRENT_SOURCE_DIR}/data
${CMAKE_CURRENT_BINARY_DIR}/data
)
if(WIN32)
target_link_libraries(imageDenoising
${PC_GLUT_LIBRARY_DIRS}/freeglut.lib
${PC_GLUT_LIBRARY_DIRS}/${GLEW_LIB_NAME}.lib
)
add_custom_command(TARGET imageDenoising
POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy
${CMAKE_CURRENT_SOURCE_DIR}/../../../bin/win64/$<CONFIGURATION>/freeglut.dll
${CMAKE_CURRENT_BINARY_DIR}/$<CONFIGURATION>
)
add_custom_command(TARGET imageDenoising
POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy
@@ -84,14 +67,11 @@ if(${OpenGL_FOUND})
${CMAKE_CURRENT_BINARY_DIR}/$<CONFIGURATION>
)
endif()
include(InstallSamples)
setup_samples_install()
else()
message(STATUS "GLUT not found - will not build sample 'imageDenoising'")
endif()
else()
message(STATUS "OpenGL not found - will not build sample 'imageDenoising'")
endif()
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/inlinePTX/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/inlinePTX/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/inlinePTX/CMakeLists.txt
View File

@@ -1,34 +1,25 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(inlinePTX LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for inlinePTX
add_executable(inlinePTX inlinePTX.cu)
target_compile_options(inlinePTX PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(inlinePTX PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(inlinePTX PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/inlinePTX_nvrtc/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/inlinePTX_nvrtc/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

45
cpp/2_Concepts_and_Techniques/inlinePTX_nvrtc/CMakeLists.txt
View File

@@ -1,45 +0,0 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
project(inlinePTX_nvrtc LANGUAGES C CXX CUDA)
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for inlinePTX_nvrtc
add_executable(inlinePTX_nvrtc inlinePTX.cpp)
target_compile_options(inlinePTX_nvrtc PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(inlinePTX_nvrtc PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(inlinePTX_nvrtc PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_link_libraries(inlinePTX_nvrtc PRIVATE
CUDA::nvrtc
CUDA::cuda_driver
)
# Copy clock_kernel.cu to the output directory
add_custom_command(TARGET inlinePTX_nvrtc POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy_if_different
${CMAKE_CURRENT_SOURCE_DIR}/inlinePTX_kernel.cu ${CMAKE_CURRENT_BINARY_DIR}
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
setup_samples_install()

39
cpp/2_Concepts_and_Techniques/inlinePTX_nvrtc/README.md
View File

@@ -1,39 +0,0 @@
# inlinePTX_nvrtc - Using Inline PTX with libNVRTC
## Description
A simple test application that demonstrates a new CUDA 4.0 ability to embed PTX in a CUDA kernel.
## Key Concepts
Performance Strategies, PTX Assembly, CUDA Driver API, Runtime Compilation
## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes
Linux, Windows, QNX
## Supported CPU Architecture
x86_64, aarch64
## CUDA APIs involved
### [CUDA Driver API](http://docs.nvidia.com/cuda/cuda-driver-api/index.html)
cuMemcpyDtoH, cuLaunchKernel, cuCtxSynchronize, cuMemAlloc, cuMemFree, cuModuleGetFunction
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaBlockSize, cudaGridSize
## Dependencies needed to build/run
[NVRTC](../../../README.md#nvrtc)
## Prerequisites
Download and install the [CUDA Toolkit](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
## References (for more details)

108
cpp/2_Concepts_and_Techniques/inlinePTX_nvrtc/inlinePTX.cpp
View File

@@ -1,108 +0,0 @@
/* Copyright (c) 2022, NVIDIA CORPORATION. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* * Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of NVIDIA CORPORATION nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
* EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
* CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
* EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
* PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
* PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
* OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
/*
* Demonstration of inline PTX (assembly language) usage in CUDA kernels
*/
// System includes
#include <assert.h>
#include <stdio.h>
// CUDA runtime
#include <cuda_runtime.h>
#include <nvrtc_helper.h>
// helper functions and utilities to work with CUDA
#include <helper_functions.h>
void sequence_cpu(int *h_ptr, int length)
{
for (int elemID = 0; elemID < length; elemID++) {
h_ptr[elemID] = elemID % 32;
}
}
int main(int argc, char **argv)
{
printf("CUDA inline PTX assembler sample\n");
char *cubin, *kernel_file;
size_t cubinSize;
kernel_file = sdkFindFilePath("inlinePTX_kernel.cu", argv[0]);
compileFileToCUBIN(kernel_file, argc, argv, &cubin, &cubinSize, 0);
CUmodule module = loadCUBIN(cubin, argc, argv);
CUfunction kernel_addr;
checkCudaErrors(cuModuleGetFunction(&kernel_addr, module, "sequence_gpu"));
const int N = 1000;
int *h_ptr = (int *)malloc(N * sizeof(int));
dim3 cudaBlockSize(256, 1, 1);
dim3 cudaGridSize((N + cudaBlockSize.x - 1) / cudaBlockSize.x, 1, 1);
CUdeviceptr d_ptr;
checkCudaErrors(cuMemAlloc(&d_ptr, N * sizeof(int)));
void *arr[] = {(void *)&d_ptr, (void *)&N};
checkCudaErrors(cuLaunchKernel(kernel_addr,
cudaGridSize.x,
cudaGridSize.y,
cudaGridSize.z, /* grid dim */
cudaBlockSize.x,
cudaBlockSize.y,
cudaBlockSize.z, /* block dim */
0,
0, /* shared mem, stream */
&arr[0], /* arguments */
0));
checkCudaErrors(cuCtxSynchronize());
sequence_cpu(h_ptr, N);
int *h_d_ptr = (int *)malloc(N * sizeof(int));
checkCudaErrors(cuMemcpyDtoH(h_d_ptr, d_ptr, N * sizeof(int)));
bool bValid = true;
for (int i = 0; i < N && bValid; i++) {
if (h_ptr[i] != h_d_ptr[i]) {
bValid = false;
}
}
printf("Test %s.\n", bValid ? "Successful" : "Failed");
checkCudaErrors(cuMemFree(d_ptr));
return bValid ? EXIT_SUCCESS : EXIT_FAILURE;
}

40
cpp/2_Concepts_and_Techniques/inlinePTX_nvrtc/inlinePTX_kernel.cu
View File

@@ -1,40 +0,0 @@
/* Copyright (c) 2022, NVIDIA CORPORATION. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* * Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of NVIDIA CORPORATION nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
* EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
* CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
* EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
* PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
* PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
* OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
extern "C" __global__ void sequence_gpu(int *d_ptr, int length)
{
int elemID = blockIdx.x * blockDim.x + threadIdx.x;
if (elemID < length) {
unsigned int laneid;
// This command gets the lane ID within the current warp
asm("mov.u32 %0, %%laneid;" : "=r"(laneid));
d_ptr[elemID] = laneid;
}
}

18
cpp/2_Concepts_and_Techniques/interval/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/interval/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/interval/CMakeLists.txt
View File

@@ -1,38 +1,29 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(interval LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for interval
add_executable(interval interval.cu)
target_compile_options(interval PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(interval PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(interval PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(interval PUBLIC
${CMAKE_CURRENT_SOURCE_DIR}
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/particles/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/particles/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

45
cpp/2_Concepts_and_Techniques/particles/CMakeLists.txt
View File

@@ -1,22 +1,18 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(particles LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
if(WIN32)
@@ -33,49 +29,37 @@ endif()
find_package(OpenGL)
find_package(GLUT)
# Source file
if(${OpenGL_FOUND})
if (${GLUT_FOUND})
# Add target for particles
add_executable(particles particleSystem.cpp particleSystem_cuda.cu particles.cpp render_particles.cpp shaders.cpp)
target_compile_options(particles PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(particles PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(particles PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_compile_options(particles PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(particles PRIVATE cxx_std_17 cuda_std_17)
target_include_directories(particles PUBLIC
${OPENGL_INCLUDE_DIR}
${CUDAToolkit_INCLUDE_DIRS}
${GLUT_INCLUDE_DIRS}
)
target_link_libraries(particles
${OPENGL_LIBRARIES}
${GLUT_LIBRARIES}
)
# Copy clock_kernel.cu to the output directory
# Copy the data directory to the output directory
add_custom_command(TARGET particles POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy_directory
${CMAKE_CURRENT_SOURCE_DIR}/data
${CMAKE_CURRENT_BINARY_DIR}/data
)
if(WIN32)
target_link_libraries(particles
${PC_GLUT_LIBRARY_DIRS}/freeglut.lib
${PC_GLUT_LIBRARY_DIRS}/${GLEW_LIB_NAME}.lib
)
add_custom_command(TARGET particles
POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy
${CMAKE_CURRENT_SOURCE_DIR}/../../../bin/win64/$<CONFIGURATION>/freeglut.dll
${CMAKE_CURRENT_BINARY_DIR}/$<CONFIGURATION>
)
add_custom_command(TARGET particles
POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy
@@ -83,14 +67,11 @@ target_compile_features(particles PRIVATE cxx_std_17 cuda_std_17)
${CMAKE_CURRENT_BINARY_DIR}/$<CONFIGURATION>
)
endif()
include(InstallSamples)
setup_samples_install()
else()
message(STATUS "GLUT not found - will not build sample 'particles'")
endif()
else()
message(STATUS "OpenGL not found - will not build sample 'particles'")
endif()
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
setup_samples_install()

4
cpp/2_Concepts_and_Techniques/particles/particleSystem.cpp
View File

@@ -91,7 +91,11 @@ uint ParticleSystem::createVBO(uint size)
return vbo;
}
// std::lerp (C++20) is pulled into the global namespace by GCC's <cmath>;
// only define our own lerp when std::lerp is unavailable to avoid a clash.
#if !defined(__cplusplus) || (__cplusplus < 202002L)
inline float lerp(float a, float b, float t) { return a + t * (b - a); }
#endif
// create a color ramp
void colorRamp(float t, float *r)

18
cpp/2_Concepts_and_Techniques/radixSortThrust/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/radixSortThrust/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/radixSortThrust/CMakeLists.txt
View File

@@ -1,34 +1,25 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(radixSortThrust LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for radixSortThrust
add_executable(radixSortThrust radixSortThrust.cu)
target_compile_options(radixSortThrust PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(radixSortThrust PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(radixSortThrust PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/reduction/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/reduction/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/reduction/CMakeLists.txt
View File

@@ -1,38 +1,29 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(reduction LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for reduction
add_executable(reduction reduction.cpp reduction_kernel.cu)
target_compile_options(reduction PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(reduction PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(reduction PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(reduction PUBLIC
${CUDAToolkit_INCLUDE_DIRS}
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/reductionMultiBlockCG/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/reductionMultiBlockCG/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/reductionMultiBlockCG/CMakeLists.txt
View File

@@ -1,34 +1,25 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(reductionMultiBlockCG LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for reductionMultiBlockCG
add_executable(reductionMultiBlockCG reductionMultiBlockCG.cu)
target_compile_options(reductionMultiBlockCG PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(reductionMultiBlockCG PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(reductionMultiBlockCG PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/scalarProd/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/scalarProd/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/scalarProd/CMakeLists.txt
View File

@@ -1,34 +1,25 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(scalarProd LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for scalarProd
add_executable(scalarProd scalarProd_cpu.cpp scalarProd.cu)
target_compile_options(scalarProd PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(scalarProd PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(scalarProd PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/scan/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/scan/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/scan/CMakeLists.txt
View File

@@ -1,38 +1,29 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(scan LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for scan
add_executable(scan main.cpp scan_gold.cpp scan.cu)
target_compile_options(scan PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(scan PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(scan PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(scan PUBLIC
${CUDAToolkit_INCLUDE_DIRS}
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/segmentationTreeThrust/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/segmentationTreeThrust/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/segmentationTreeThrust/CMakeLists.txt
View File

@@ -1,34 +1,26 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(segmentationTreeThrust LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for segmentationTreeThrust
add_executable(segmentationTreeThrust segmentationTree.cu)
target_compile_options(segmentationTreeThrust PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(segmentationTreeThrust PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(segmentationTreeThrust PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
# Copy data files to output directory
add_custom_command(TARGET segmentationTreeThrust POST_BUILD
COMMAND ${CMAKE_COMMAND} -E copy_if_different
@@ -50,6 +42,5 @@ add_custom_command(TARGET segmentationTreeThrust POST_BUILD
${CMAKE_CURRENT_BINARY_DIR}/
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/shfl_scan/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/shfl_scan/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/shfl_scan/CMakeLists.txt
View File

@@ -1,34 +1,25 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(shfl_scan LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for shfl_scan
add_executable(shfl_scan shfl_scan.cu)
target_compile_options(shfl_scan PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(shfl_scan PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(shfl_scan PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/sortingNetworks/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/sortingNetworks/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/sortingNetworks/CMakeLists.txt
View File

@@ -1,38 +1,29 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(sortingNetworks LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for sortingNetworks
add_executable(sortingNetworks main.cpp bitonicSort.cu sortingNetworks_validate.cpp)
target_compile_options(sortingNetworks PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(sortingNetworks PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(sortingNetworks PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_include_directories(sortingNetworks PUBLIC
${CUDAToolkit_INCLUDE_DIRS}
)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/streamOrderedAllocation/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/streamOrderedAllocation/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/streamOrderedAllocation/CMakeLists.txt
View File

@@ -1,34 +1,25 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(streamOrderedAllocation LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for streamOrderedAllocation
add_executable(streamOrderedAllocation streamOrderedAllocation.cu)
target_compile_options(streamOrderedAllocation PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(streamOrderedAllocation PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(streamOrderedAllocation PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/streamOrderedAllocationIPC/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/streamOrderedAllocationIPC/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/streamOrderedAllocationIPC/CMakeLists.txt
View File

@@ -1,27 +1,21 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(streamOrderedAllocationIPC LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
if(CMAKE_SYSTEM_NAME STREQUAL "Linux")
# Source file
# Add target for streamOrderedAllocationIPC
if(CMAKE_SYSTEM_PROCESSOR STREQUAL "aarch64")
message(STATUS "Will not build sample streamOrderedAllocationIPC - not supported on aarch64")
else()
@@ -31,8 +25,6 @@ if(CMAKE_SYSTEM_NAME STREQUAL "Linux")
target_compile_features(streamOrderedAllocationIPC PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(streamOrderedAllocationIPC PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
target_link_libraries(streamOrderedAllocationIPC PUBLIC
CUDA::cuda_driver
)
@@ -41,6 +33,5 @@ else()
message(STATUS "Will not build sample streamOrderedAllocationIPC - requires Linux OS")
endif()
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

18
cpp/2_Concepts_and_Techniques/streamOrderedAllocationP2P/.vscode/c_cpp_properties.json vendored
View File

@@ -1,18 +0,0 @@
{
"configurations": [
{
"name": "Linux",
"includePath": [
"${workspaceFolder}/**",
"${workspaceFolder}/../../../Common"
],
"defines": [],
"compilerPath": "/usr/local/cuda/bin/nvcc",
"cStandard": "gnu17",
"cppStandard": "gnu++14",
"intelliSenseMode": "linux-gcc-x64",
"configurationProvider": "ms-vscode.makefile-tools"
}
],
"version": 4
}

7
cpp/2_Concepts_and_Techniques/streamOrderedAllocationP2P/.vscode/extensions.json vendored
View File

@@ -1,7 +0,0 @@
{
"recommendations": [
"nvidia.nsight-vscode-edition",
"ms-vscode.cpptools",
"ms-vscode.makefile-tools"
]
}

27
cpp/2_Concepts_and_Techniques/streamOrderedAllocationP2P/CMakeLists.txt
View File

@@ -1,34 +1,25 @@
cmake_minimum_required(VERSION 3.20)
list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/Modules")
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/DetectCudaArch.cmake")
if(SAMPLE_SKIP_BUILD)
return()
endif()
project(streamOrderedAllocationP2P LANGUAGES C CXX CUDA)
include("${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/CudaSampleCommon.cmake")
find_package(CUDAToolkit REQUIRED)
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
set(CMAKE_CUDA_ARCHITECTURES 75 80 86 87 89 90 100 110 120)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -Wno-deprecated-gpu-targets")
if(ENABLE_CUDA_DEBUG)
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -G") # enable cuda-gdb (may significantly affect performance on some targets)
else()
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -lineinfo") # add line information to all builds for debug tools (exclusive to -G option)
endif()
# Include directories and libraries
include_directories(../../../Common)
# Source file
# Add target for streamOrderedAllocationP2P
add_executable(streamOrderedAllocationP2P streamOrderedAllocationP2P.cu)
target_compile_options(streamOrderedAllocationP2P PRIVATE $<$<COMPILE_LANGUAGE:CUDA>:--extended-lambda>)
target_compile_features(streamOrderedAllocationP2P PRIVATE cxx_std_17 cuda_std_17)
set_target_properties(streamOrderedAllocationP2P PROPERTIES CUDA_SEPARABLE_COMPILATION ON)
# Include installation configuration
include(${CMAKE_CURRENT_SOURCE_DIR}/../../../cmake/InstallSamples.cmake)
include(InstallSamples)
setup_samples_install()

Some files were not shown because too many files have changed in this diff Show More