mirror of
https://github.com/NVIDIA/cuda-samples.git
synced 2026-11-18 01:48:16 +08:00
Compare commits
11 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
ba04faaf73 | ||
|
|
8789eb6266 | ||
|
|
e950012e72 | ||
|
|
95b7cea7bc | ||
|
|
2aeaf51b11 | ||
|
|
ba5a483c6e | ||
|
|
7a5b3e6c8c | ||
|
|
0787bc0489 | ||
|
|
5c3ec60fae | ||
|
|
568b39bd5b | ||
|
|
067cb65523 |
74
README.md
74
README.md
@@ -1,11 +1,24 @@
|
|||||||
# CUDA Samples
|
# CUDA Samples
|
||||||
|
|
||||||
Samples for CUDA Developers which demonstrates features in CUDA Toolkit. This version supports [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads).
|
Samples for CUDA Developers which demonstrates features in CUDA Toolkit. This version supports [CUDA Toolkit 11.4 update 1](https://developer.nvidia.com/cuda-downloads).
|
||||||
|
|
||||||
## Release Notes
|
## Release Notes
|
||||||
|
|
||||||
This section describes the release notes for the CUDA Samples on GitHub only.
|
This section describes the release notes for the CUDA Samples on GitHub only.
|
||||||
|
|
||||||
|
### CUDA 11.4 update 1
|
||||||
|
* Added support for VS Code on linux platform.
|
||||||
|
|
||||||
|
### CUDA 11.4
|
||||||
|
* Added `cdpQuadtree`. Demonstrates Quad Trees implementation using CUDA Dynamic Parallelism.
|
||||||
|
* Updated `simpleVulkan`, `simpleVulkanMMAP` and `vulkanImageCUDA`. Demonstrates use of SPIR-V shaders.
|
||||||
|
|
||||||
|
### CUDA 11.3
|
||||||
|
* Added `streamOrderedAllocationIPC`. Demonstrates Inter Process Communication using one process per GPU for computation.
|
||||||
|
* Added `simpleCUBLAS_LU`. Demonstrates batched matrix LU decomposition using cuBLAS API `cublas<t>getrfBatched()`
|
||||||
|
* Updated `simpleVulkan`. Demonstrates use of timeline semaphore.
|
||||||
|
* Updated multiple samples to use pinned memory using `cudaMallocHost()`.
|
||||||
|
|
||||||
### CUDA 11.2
|
### CUDA 11.2
|
||||||
* Added `streamOrderedAllocation`. Demonstrates stream ordered memory allocation on a GPU using cudaMallocAsync and cudaMemPool family of APIs.
|
* Added `streamOrderedAllocation`. Demonstrates stream ordered memory allocation on a GPU using cudaMallocAsync and cudaMemPool family of APIs.
|
||||||
* Added `streamOrderedAllocationP2P`. Demonstrates peer-to-peer access of stream ordered memory allocated using cudaMallocAsync and cudaMemPool family of APIs.
|
* Added `streamOrderedAllocationP2P`. Demonstrates peer-to-peer access of stream ordered memory allocated using cudaMallocAsync and cudaMemPool family of APIs.
|
||||||
@@ -103,7 +116,7 @@ This is the first release of CUDA Samples on GitHub:
|
|||||||
|
|
||||||
### Prerequisites
|
### Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
For system requirements and installation instructions of cuda toolkit, please refer to the [Linux Installation Guide](http://docs.nvidia.com/cuda/cuda-installation-guide-linux/), and the [Windows Installation Guide](http://docs.nvidia.com/cuda/cuda-installation-guide-microsoft-windows/index.html).
|
For system requirements and installation instructions of cuda toolkit, please refer to the [Linux Installation Guide](http://docs.nvidia.com/cuda/cuda-installation-guide-linux/), and the [Windows Installation Guide](http://docs.nvidia.com/cuda/cuda-installation-guide-microsoft-windows/index.html).
|
||||||
|
|
||||||
### Getting the CUDA Samples
|
### Getting the CUDA Samples
|
||||||
@@ -160,38 +173,39 @@ The samples makefiles can take advantage of certain options:
|
|||||||
### Samples by OS
|
### Samples by OS
|
||||||
|
|
||||||
#### Linux
|
#### Linux
|
||||||
**[warpAggregatedAtomicsCG](./Samples/warpAggregatedAtomicsCG)** | **[boxFilterNPP](./Samples/boxFilterNPP)** | **[streamOrderedAllocationP2P](./Samples/streamOrderedAllocationP2P)** | **[binaryPartitionCG](./Samples/binaryPartitionCG)** |
|
**[vectorAdd_nvrtc](./Samples/vectorAdd_nvrtc)** | **[shfl_scan](./Samples/shfl_scan)** | **[binaryPartitionCG](./Samples/binaryPartitionCG)** | **[concurrentKernels](./Samples/concurrentKernels)** |
|
||||||
---|---|---|---|
|
---|---|---|---|
|
||||||
**[dmmaTensorCoreGemm](./Samples/dmmaTensorCoreGemm)** | **[EGLStream_CUDA_Interop](./Samples/EGLStream_CUDA_Interop)** | **[conjugateGradientMultiBlockCG](./Samples/conjugateGradientMultiBlockCG)** | **[simpleIPC](./Samples/simpleIPC)** |
|
**[streamOrderedAllocationIPC](./Samples/streamOrderedAllocationIPC)** | **[UnifiedMemoryPerf](./Samples/UnifiedMemoryPerf)** | **[bf16TensorCoreGemm](./Samples/bf16TensorCoreGemm)** | **[cuSolverSp_LinearSolver](./Samples/cuSolverSp_LinearSolver)** |
|
||||||
**[memMapIPCDrv](./Samples/memMapIPCDrv)** | **[vectorAddMMAP](./Samples/vectorAddMMAP)** | **[shfl_scan](./Samples/shfl_scan)** | **[simpleZeroCopy](./Samples/simpleZeroCopy)** |
|
**[nvJPEG](./Samples/nvJPEG)** | **[watershedSegmentationNPP](./Samples/watershedSegmentationNPP)** | **[vulkanImageCUDA](./Samples/vulkanImageCUDA)** | **[boxFilterNPP](./Samples/boxFilterNPP)** |
|
||||||
**[conjugateGradientCudaGraphs](./Samples/conjugateGradientCudaGraphs)** | **[globalToShmemAsyncCopy](./Samples/globalToShmemAsyncCopy)** | **[cudaNvSciNvMedia](./Samples/cudaNvSciNvMedia)** | **[nvJPEG](./Samples/nvJPEG)** |
|
**[matrixMul](./Samples/matrixMul)** | **[nvJPEG_encoder](./Samples/nvJPEG_encoder)** | **[simpleCudaGraphs](./Samples/simpleCudaGraphs)** | **[simpleCUFFT](./Samples/simpleCUFFT)** |
|
||||||
**[batchedLabelMarkersAndLabelCompressionNPP](./Samples/batchedLabelMarkersAndLabelCompressionNPP)** | **[watershedSegmentationNPP](./Samples/watershedSegmentationNPP)** | **[simpleCudaGraphs](./Samples/simpleCudaGraphs)** | **[streamOrderedAllocation](./Samples/streamOrderedAllocation)** |
|
**[cudaTensorCoreGemm](./Samples/cudaTensorCoreGemm)** | **[cudaOpenMP](./Samples/cudaOpenMP)** | **[NV12toBGRandResize](./Samples/NV12toBGRandResize)** | **[cudaNvSci](./Samples/cudaNvSci)** |
|
||||||
**[deviceQuery](./Samples/deviceQuery)** | **[simpleVoteIntrinsics](./Samples/simpleVoteIntrinsics)** | **[simpleCUBLASXT](./Samples/simpleCUBLASXT)** | **[simpleAttributes](./Samples/simpleAttributes)** |
|
**[immaTensorCoreGemm](./Samples/immaTensorCoreGemm)** | **[simpleVulkan](./Samples/simpleVulkan)** | **[simpleCUBLAS](./Samples/simpleCUBLAS)** | **[jacobiCudaGraphs](./Samples/jacobiCudaGraphs)** |
|
||||||
**[cudaNvSci](./Samples/cudaNvSci)** | **[tf32TensorCoreGemm](./Samples/tf32TensorCoreGemm)** | **[UnifiedMemoryPerf](./Samples/UnifiedMemoryPerf)** | **[cudaCompressibleMemory](./Samples/cudaCompressibleMemory)** |
|
**[streamOrderedAllocationP2P](./Samples/streamOrderedAllocationP2P)** | **[dmmaTensorCoreGemm](./Samples/dmmaTensorCoreGemm)** | **[EGLStream_CUDA_Interop](./Samples/EGLStream_CUDA_Interop)** | **[vectorAddMMAP](./Samples/vectorAddMMAP)** |
|
||||||
**[bf16TensorCoreGemm](./Samples/bf16TensorCoreGemm)** | **[cuSolverDn_LinearSolver](./Samples/cuSolverDn_LinearSolver)** | **[vulkanImageCUDA](./Samples/vulkanImageCUDA)** | **[conjugateGradientMultiDeviceCG](./Samples/conjugateGradientMultiDeviceCG)** |
|
**[MersenneTwisterGP11213](./Samples/MersenneTwisterGP11213)** | **[streamOrderedAllocation](./Samples/streamOrderedAllocation)** | **[cuSolverDn_LinearSolver](./Samples/cuSolverDn_LinearSolver)** | **[cudaCompressibleMemory](./Samples/cudaCompressibleMemory)** |
|
||||||
**[matrixMulDrv](./Samples/matrixMulDrv)** | **[cuSolverSp_LinearSolver](./Samples/cuSolverSp_LinearSolver)** | **[simpleCUFFT](./Samples/simpleCUFFT)** | **[reduction](./Samples/reduction)** |
|
**[simpleAWBarrier](./Samples/simpleAWBarrier)** | **[simpleGL](./Samples/simpleGL)** | **[p2pBandwidthLatencyTest](./Samples/p2pBandwidthLatencyTest)** | **[conjugateGradientCudaGraphs](./Samples/conjugateGradientCudaGraphs)** |
|
||||||
**[nvJPEG_encoder](./Samples/nvJPEG_encoder)** | **[simpleDrvRuntime](./Samples/simpleDrvRuntime)** | **[MersenneTwisterGP11213](./Samples/MersenneTwisterGP11213)** | **[simpleAWBarrier](./Samples/simpleAWBarrier)** |
|
**[globalToShmemAsyncCopy](./Samples/globalToShmemAsyncCopy)** | **[deviceQuery](./Samples/deviceQuery)** | **[systemWideAtomics](./Samples/systemWideAtomics)** | **[matrixMulDrv](./Samples/matrixMulDrv)** |
|
||||||
**[immaTensorCoreGemm](./Samples/immaTensorCoreGemm)** | **[bandwidthTest](./Samples/bandwidthTest)** | **[concurrentKernels](./Samples/concurrentKernels)** | **[simpleCUBLAS](./Samples/simpleCUBLAS)** |
|
**[conjugateGradientMultiBlockCG](./Samples/conjugateGradientMultiBlockCG)** | **[simpleZeroCopy](./Samples/simpleZeroCopy)** | **[simpleAttributes](./Samples/simpleAttributes)** | **[batchedLabelMarkersAndLabelCompressionNPP](./Samples/batchedLabelMarkersAndLabelCompressionNPP)** |
|
||||||
**[NV12toBGRandResize](./Samples/NV12toBGRandResize)** | **[simpleGL](./Samples/simpleGL)** | **[cudaTensorCoreGemm](./Samples/cudaTensorCoreGemm)** | **[jacobiCudaGraphs](./Samples/jacobiCudaGraphs)** |
|
**[conjugateGradientMultiDeviceCG](./Samples/conjugateGradientMultiDeviceCG)** | **[simpleIPC](./Samples/simpleIPC)** | **[simpleVulkanMMAP](./Samples/simpleVulkanMMAP)** | **[cudaNvSciNvMedia](./Samples/cudaNvSciNvMedia)** |
|
||||||
**[simpleVulkan](./Samples/simpleVulkan)** | **[vectorAdd_nvrtc](./Samples/vectorAdd_nvrtc)** | **[cannyEdgeDetectorNPP](./Samples/cannyEdgeDetectorNPP)** | **[p2pBandwidthLatencyTest](./Samples/p2pBandwidthLatencyTest)** |
|
**[simpleVoteIntrinsics](./Samples/simpleVoteIntrinsics)** | **[simpleCUBLAS_LU](./Samples/simpleCUBLAS_LU)** | **[bandwidthTest](./Samples/bandwidthTest)** | **[cdpQuadtree](./Samples/cdpQuadtree)** |
|
||||||
**[simpleVulkanMMAP](./Samples/simpleVulkanMMAP)** | **[cudaOpenMP](./Samples/cudaOpenMP)** | **[matrixMul](./Samples/matrixMul)** | **[systemWideAtomics](./Samples/systemWideAtomics)** |
|
**[simpleCUBLASXT](./Samples/simpleCUBLASXT)** | **[tf32TensorCoreGemm](./Samples/tf32TensorCoreGemm)** | **[simpleDrvRuntime](./Samples/simpleDrvRuntime)** | **[reduction](./Samples/reduction)** |
|
||||||
|
**[memMapIPCDrv](./Samples/memMapIPCDrv)** | **[warpAggregatedAtomicsCG](./Samples/warpAggregatedAtomicsCG)** | **[cannyEdgeDetectorNPP](./Samples/cannyEdgeDetectorNPP)** |
|
||||||
|
|
||||||
#### Windows
|
#### Windows
|
||||||
**[warpAggregatedAtomicsCG](./Samples/warpAggregatedAtomicsCG)** | **[boxFilterNPP](./Samples/boxFilterNPP)** | **[streamOrderedAllocationP2P](./Samples/streamOrderedAllocationP2P)** | **[binaryPartitionCG](./Samples/binaryPartitionCG)** |
|
**[vectorAdd_nvrtc](./Samples/vectorAdd_nvrtc)** | **[shfl_scan](./Samples/shfl_scan)** | **[binaryPartitionCG](./Samples/binaryPartitionCG)** | **[concurrentKernels](./Samples/concurrentKernels)** |
|
||||||
---|---|---|---|
|
---|---|---|---|
|
||||||
**[dmmaTensorCoreGemm](./Samples/dmmaTensorCoreGemm)** | **[conjugateGradientMultiBlockCG](./Samples/conjugateGradientMultiBlockCG)** | **[simpleIPC](./Samples/simpleIPC)** | **[memMapIPCDrv](./Samples/memMapIPCDrv)** |
|
**[UnifiedMemoryPerf](./Samples/UnifiedMemoryPerf)** | **[bf16TensorCoreGemm](./Samples/bf16TensorCoreGemm)** | **[cuSolverSp_LinearSolver](./Samples/cuSolverSp_LinearSolver)** | **[nvJPEG](./Samples/nvJPEG)** |
|
||||||
**[vectorAddMMAP](./Samples/vectorAddMMAP)** | **[shfl_scan](./Samples/shfl_scan)** | **[simpleZeroCopy](./Samples/simpleZeroCopy)** | **[conjugateGradientCudaGraphs](./Samples/conjugateGradientCudaGraphs)** |
|
**[watershedSegmentationNPP](./Samples/watershedSegmentationNPP)** | **[vulkanImageCUDA](./Samples/vulkanImageCUDA)** | **[boxFilterNPP](./Samples/boxFilterNPP)** | **[matrixMul](./Samples/matrixMul)** |
|
||||||
**[globalToShmemAsyncCopy](./Samples/globalToShmemAsyncCopy)** | **[nvJPEG](./Samples/nvJPEG)** | **[batchedLabelMarkersAndLabelCompressionNPP](./Samples/batchedLabelMarkersAndLabelCompressionNPP)** | **[simpleD3D12](./Samples/simpleD3D12)** |
|
**[nvJPEG_encoder](./Samples/nvJPEG_encoder)** | **[simpleCudaGraphs](./Samples/simpleCudaGraphs)** | **[simpleCUFFT](./Samples/simpleCUFFT)** | **[cudaTensorCoreGemm](./Samples/cudaTensorCoreGemm)** |
|
||||||
**[watershedSegmentationNPP](./Samples/watershedSegmentationNPP)** | **[simpleCudaGraphs](./Samples/simpleCudaGraphs)** | **[streamOrderedAllocation](./Samples/streamOrderedAllocation)** | **[deviceQuery](./Samples/deviceQuery)** |
|
**[cudaOpenMP](./Samples/cudaOpenMP)** | **[NV12toBGRandResize](./Samples/NV12toBGRandResize)** | **[immaTensorCoreGemm](./Samples/immaTensorCoreGemm)** | **[simpleVulkan](./Samples/simpleVulkan)** |
|
||||||
**[simpleVoteIntrinsics](./Samples/simpleVoteIntrinsics)** | **[simpleCUBLASXT](./Samples/simpleCUBLASXT)** | **[simpleAttributes](./Samples/simpleAttributes)** | **[tf32TensorCoreGemm](./Samples/tf32TensorCoreGemm)** |
|
**[simpleCUBLAS](./Samples/simpleCUBLAS)** | **[jacobiCudaGraphs](./Samples/jacobiCudaGraphs)** | **[streamOrderedAllocationP2P](./Samples/streamOrderedAllocationP2P)** | **[dmmaTensorCoreGemm](./Samples/dmmaTensorCoreGemm)** |
|
||||||
**[UnifiedMemoryPerf](./Samples/UnifiedMemoryPerf)** | **[cudaCompressibleMemory](./Samples/cudaCompressibleMemory)** | **[bf16TensorCoreGemm](./Samples/bf16TensorCoreGemm)** | **[cuSolverDn_LinearSolver](./Samples/cuSolverDn_LinearSolver)** |
|
**[vectorAddMMAP](./Samples/vectorAddMMAP)** | **[MersenneTwisterGP11213](./Samples/MersenneTwisterGP11213)** | **[streamOrderedAllocation](./Samples/streamOrderedAllocation)** | **[cuSolverDn_LinearSolver](./Samples/cuSolverDn_LinearSolver)** |
|
||||||
**[vulkanImageCUDA](./Samples/vulkanImageCUDA)** | **[conjugateGradientMultiDeviceCG](./Samples/conjugateGradientMultiDeviceCG)** | **[matrixMulDrv](./Samples/matrixMulDrv)** | **[cuSolverSp_LinearSolver](./Samples/cuSolverSp_LinearSolver)** |
|
**[cudaCompressibleMemory](./Samples/cudaCompressibleMemory)** | **[simpleAWBarrier](./Samples/simpleAWBarrier)** | **[simpleGL](./Samples/simpleGL)** | **[p2pBandwidthLatencyTest](./Samples/p2pBandwidthLatencyTest)** |
|
||||||
**[simpleCUFFT](./Samples/simpleCUFFT)** | **[reduction](./Samples/reduction)** | **[nvJPEG_encoder](./Samples/nvJPEG_encoder)** | **[simpleDrvRuntime](./Samples/simpleDrvRuntime)** |
|
**[conjugateGradientCudaGraphs](./Samples/conjugateGradientCudaGraphs)** | **[simpleD3D11](./Samples/simpleD3D11)** | **[globalToShmemAsyncCopy](./Samples/globalToShmemAsyncCopy)** | **[deviceQuery](./Samples/deviceQuery)** |
|
||||||
**[simpleD3D11](./Samples/simpleD3D11)** | **[MersenneTwisterGP11213](./Samples/MersenneTwisterGP11213)** | **[simpleAWBarrier](./Samples/simpleAWBarrier)** | **[immaTensorCoreGemm](./Samples/immaTensorCoreGemm)** |
|
**[matrixMulDrv](./Samples/matrixMulDrv)** | **[conjugateGradientMultiBlockCG](./Samples/conjugateGradientMultiBlockCG)** | **[simpleZeroCopy](./Samples/simpleZeroCopy)** | **[simpleAttributes](./Samples/simpleAttributes)** |
|
||||||
**[bandwidthTest](./Samples/bandwidthTest)** | **[concurrentKernels](./Samples/concurrentKernels)** | **[simpleCUBLAS](./Samples/simpleCUBLAS)** | **[NV12toBGRandResize](./Samples/NV12toBGRandResize)** |
|
**[simpleD3D12](./Samples/simpleD3D12)** | **[batchedLabelMarkersAndLabelCompressionNPP](./Samples/batchedLabelMarkersAndLabelCompressionNPP)** | **[conjugateGradientMultiDeviceCG](./Samples/conjugateGradientMultiDeviceCG)** | **[simpleIPC](./Samples/simpleIPC)** |
|
||||||
**[simpleGL](./Samples/simpleGL)** | **[cudaTensorCoreGemm](./Samples/cudaTensorCoreGemm)** | **[jacobiCudaGraphs](./Samples/jacobiCudaGraphs)** | **[simpleVulkan](./Samples/simpleVulkan)** |
|
**[simpleVulkanMMAP](./Samples/simpleVulkanMMAP)** | **[simpleVoteIntrinsics](./Samples/simpleVoteIntrinsics)** | **[simpleCUBLAS_LU](./Samples/simpleCUBLAS_LU)** | **[bandwidthTest](./Samples/bandwidthTest)** |
|
||||||
**[vectorAdd_nvrtc](./Samples/vectorAdd_nvrtc)** | **[cannyEdgeDetectorNPP](./Samples/cannyEdgeDetectorNPP)** | **[p2pBandwidthLatencyTest](./Samples/p2pBandwidthLatencyTest)** | **[simpleVulkanMMAP](./Samples/simpleVulkanMMAP)** |
|
**[cdpQuadtree](./Samples/cdpQuadtree)** | **[simpleCUBLASXT](./Samples/simpleCUBLASXT)** | **[tf32TensorCoreGemm](./Samples/tf32TensorCoreGemm)** | **[simpleDrvRuntime](./Samples/simpleDrvRuntime)** |
|
||||||
**[cudaOpenMP](./Samples/cudaOpenMP)** | **[matrixMul](./Samples/matrixMul)** |
|
**[reduction](./Samples/reduction)** | **[memMapIPCDrv](./Samples/memMapIPCDrv)** | **[warpAggregatedAtomicsCG](./Samples/warpAggregatedAtomicsCG)** | **[cannyEdgeDetectorNPP](./Samples/cannyEdgeDetectorNPP)** |
|
||||||
|
|
||||||
## Dependencies
|
## Dependencies
|
||||||
|
|
||||||
|
|||||||
18
Samples/EGLStream_CUDA_Interop/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/EGLStream_CUDA_Interop/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"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
Samples/EGLStream_CUDA_Interop/.vscode/extensions.json
vendored
Normal file
7
Samples/EGLStream_CUDA_Interop/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/EGLStream_CUDA_Interop/.vscode/launch.json
vendored
Normal file
10
Samples/EGLStream_CUDA_Interop/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/EGLStream_CUDA_Interop"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/EGLStream_CUDA_Interop/.vscode/tasks.json
vendored
Normal file
15
Samples/EGLStream_CUDA_Interop/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@@ -30,7 +30,7 @@ cuDeviceGet, cuDeviceGetAttribute, cuDeviceComputeCapability, cuDeviceGetCount,
|
|||||||
|
|
||||||
## Prerequisites
|
## Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
||||||
|
|
||||||
## Build and Run
|
## Build and Run
|
||||||
|
|||||||
18
Samples/MersenneTwisterGP11213/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/MersenneTwisterGP11213/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"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
Samples/MersenneTwisterGP11213/.vscode/extensions.json
vendored
Normal file
7
Samples/MersenneTwisterGP11213/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/MersenneTwisterGP11213/.vscode/launch.json
vendored
Normal file
10
Samples/MersenneTwisterGP11213/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/MersenneTwisterGP11213"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/MersenneTwisterGP11213/.vscode/tasks.json
vendored
Normal file
15
Samples/MersenneTwisterGP11213/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@@ -47,138 +47,134 @@
|
|||||||
|
|
||||||
float compareResults(int rand_n, float *h_RandGPU, float *h_RandCPU);
|
float compareResults(int rand_n, float *h_RandGPU, float *h_RandCPU);
|
||||||
|
|
||||||
const int DEFAULT_RAND_N = 2400000;
|
const int DEFAULT_RAND_N = 2400000;
|
||||||
const unsigned int DEFAULT_SEED = 777;
|
const unsigned int DEFAULT_SEED = 777;
|
||||||
|
|
||||||
///////////////////////////////////////////////////////////////////////////////
|
///////////////////////////////////////////////////////////////////////////////
|
||||||
// Main program
|
// Main program
|
||||||
///////////////////////////////////////////////////////////////////////////////
|
///////////////////////////////////////////////////////////////////////////////
|
||||||
int main(int argc, char **argv)
|
int main(int argc, char **argv) {
|
||||||
{
|
// Start logs
|
||||||
// Start logs
|
printf("%s Starting...\n\n", argv[0]);
|
||||||
printf("%s Starting...\n\n", argv[0]);
|
|
||||||
|
|
||||||
// initialize the GPU, either identified by --device
|
// initialize the GPU, either identified by --device
|
||||||
// or by picking the device with highest flop rate.
|
// or by picking the device with highest flop rate.
|
||||||
int devID = findCudaDevice(argc, (const char **)argv);
|
int devID = findCudaDevice(argc, (const char **)argv);
|
||||||
|
|
||||||
// parsing the number of random numbers to generate
|
// parsing the number of random numbers to generate
|
||||||
int rand_n = DEFAULT_RAND_N;
|
int rand_n = DEFAULT_RAND_N;
|
||||||
|
|
||||||
if (checkCmdLineFlag(argc, (const char **) argv, "count"))
|
if (checkCmdLineFlag(argc, (const char **)argv, "count")) {
|
||||||
{
|
rand_n = getCmdLineArgumentInt(argc, (const char **)argv, "count");
|
||||||
rand_n = getCmdLineArgumentInt(argc, (const char **) argv, "count");
|
}
|
||||||
}
|
|
||||||
|
|
||||||
printf("Allocating data for %i samples...\n", rand_n);
|
printf("Allocating data for %i samples...\n", rand_n);
|
||||||
|
|
||||||
// parsing the seed
|
// parsing the seed
|
||||||
int seed = DEFAULT_SEED;
|
int seed = DEFAULT_SEED;
|
||||||
|
|
||||||
if (checkCmdLineFlag(argc, (const char **) argv, "seed"))
|
if (checkCmdLineFlag(argc, (const char **)argv, "seed")) {
|
||||||
{
|
seed = getCmdLineArgumentInt(argc, (const char **)argv, "seed");
|
||||||
seed = getCmdLineArgumentInt(argc, (const char **) argv, "seed");
|
}
|
||||||
}
|
|
||||||
|
|
||||||
printf("Seeding with %i ...\n", seed);
|
printf("Seeding with %i ...\n", seed);
|
||||||
|
|
||||||
cudaStream_t stream;
|
cudaStream_t stream;
|
||||||
checkCudaErrors(cudaStreamCreateWithFlags(&stream, cudaStreamNonBlocking));
|
checkCudaErrors(cudaStreamCreateWithFlags(&stream, cudaStreamNonBlocking));
|
||||||
|
|
||||||
float *d_Rand;
|
float *d_Rand;
|
||||||
checkCudaErrors(cudaMalloc((void **)&d_Rand, rand_n * sizeof(float)));
|
checkCudaErrors(cudaMalloc((void **)&d_Rand, rand_n * sizeof(float)));
|
||||||
|
|
||||||
curandGenerator_t prngGPU;
|
curandGenerator_t prngGPU;
|
||||||
checkCudaErrors(curandCreateGenerator(&prngGPU, CURAND_RNG_PSEUDO_MTGP32));
|
checkCudaErrors(curandCreateGenerator(&prngGPU, CURAND_RNG_PSEUDO_MTGP32));
|
||||||
checkCudaErrors(curandSetStream(prngGPU, stream));
|
checkCudaErrors(curandSetStream(prngGPU, stream));
|
||||||
checkCudaErrors(curandSetPseudoRandomGeneratorSeed(prngGPU, seed));
|
checkCudaErrors(curandSetPseudoRandomGeneratorSeed(prngGPU, seed));
|
||||||
|
|
||||||
curandGenerator_t prngCPU;
|
curandGenerator_t prngCPU;
|
||||||
checkCudaErrors(curandCreateGeneratorHost(&prngCPU, CURAND_RNG_PSEUDO_MTGP32));
|
checkCudaErrors(
|
||||||
checkCudaErrors(curandSetPseudoRandomGeneratorSeed(prngCPU, seed));
|
curandCreateGeneratorHost(&prngCPU, CURAND_RNG_PSEUDO_MTGP32));
|
||||||
|
checkCudaErrors(curandSetPseudoRandomGeneratorSeed(prngCPU, seed));
|
||||||
|
|
||||||
//
|
//
|
||||||
// Example 1: Compare random numbers generated on GPU and CPU
|
// Example 1: Compare random numbers generated on GPU and CPU
|
||||||
float *h_RandGPU = (float *)malloc(rand_n * sizeof(float));
|
float *h_RandGPU;
|
||||||
|
checkCudaErrors(cudaMallocHost(&h_RandGPU, rand_n * sizeof(float)));
|
||||||
|
|
||||||
printf("Generating random numbers on GPU...\n\n");
|
printf("Generating random numbers on GPU...\n\n");
|
||||||
checkCudaErrors(curandGenerateUniform(prngGPU, (float *) d_Rand, rand_n));
|
checkCudaErrors(curandGenerateUniform(prngGPU, (float *)d_Rand, rand_n));
|
||||||
|
|
||||||
printf("\nReading back the results...\n");
|
printf("\nReading back the results...\n");
|
||||||
checkCudaErrors(cudaMemcpyAsync(h_RandGPU, d_Rand, rand_n * sizeof(float), cudaMemcpyDeviceToHost, stream));
|
checkCudaErrors(cudaMemcpyAsync(h_RandGPU, d_Rand, rand_n * sizeof(float),
|
||||||
|
cudaMemcpyDeviceToHost, stream));
|
||||||
|
|
||||||
|
float *h_RandCPU = (float *)malloc(rand_n * sizeof(float));
|
||||||
|
|
||||||
float *h_RandCPU = (float *)malloc(rand_n * sizeof(float));
|
printf("Generating random numbers on CPU...\n\n");
|
||||||
|
checkCudaErrors(curandGenerateUniform(prngCPU, (float *)h_RandCPU, rand_n));
|
||||||
|
|
||||||
printf("Generating random numbers on CPU...\n\n");
|
checkCudaErrors(cudaStreamSynchronize(stream));
|
||||||
checkCudaErrors(curandGenerateUniform(prngCPU, (float *) h_RandCPU, rand_n));
|
printf("Comparing CPU/GPU random numbers...\n\n");
|
||||||
|
float L1norm = compareResults(rand_n, h_RandGPU, h_RandCPU);
|
||||||
|
|
||||||
checkCudaErrors(cudaStreamSynchronize(stream));
|
//
|
||||||
printf("Comparing CPU/GPU random numbers...\n\n");
|
// Example 2: Timing of random number generation on GPU
|
||||||
float L1norm = compareResults(rand_n, h_RandGPU, h_RandCPU);
|
const int numIterations = 10;
|
||||||
|
int i;
|
||||||
|
StopWatchInterface *hTimer;
|
||||||
|
|
||||||
//
|
sdkCreateTimer(&hTimer);
|
||||||
// Example 2: Timing of random number generation on GPU
|
sdkResetTimer(&hTimer);
|
||||||
const int numIterations = 10;
|
sdkStartTimer(&hTimer);
|
||||||
int i;
|
|
||||||
StopWatchInterface *hTimer;
|
|
||||||
|
|
||||||
sdkCreateTimer(&hTimer);
|
for (i = 0; i < numIterations; i++) {
|
||||||
sdkResetTimer(&hTimer);
|
checkCudaErrors(curandGenerateUniform(prngGPU, (float *)d_Rand, rand_n));
|
||||||
sdkStartTimer(&hTimer);
|
}
|
||||||
|
|
||||||
for (i = 0; i < numIterations; i++)
|
checkCudaErrors(cudaStreamSynchronize(stream));
|
||||||
{
|
sdkStopTimer(&hTimer);
|
||||||
checkCudaErrors(curandGenerateUniform(prngGPU, (float *) d_Rand, rand_n));
|
|
||||||
}
|
|
||||||
|
|
||||||
checkCudaErrors(cudaStreamSynchronize(stream));
|
double gpuTime = 1.0e-3 * sdkGetTimerValue(&hTimer) / (double)numIterations;
|
||||||
sdkStopTimer(&hTimer);
|
|
||||||
|
|
||||||
double gpuTime = 1.0e-3 * sdkGetTimerValue(&hTimer)/(double)numIterations;
|
printf(
|
||||||
|
"MersenneTwisterGP11213, Throughput = %.4f GNumbers/s, Time = %.5f s, "
|
||||||
|
"Size = %u Numbers\n",
|
||||||
|
1.0e-9 * rand_n / gpuTime, gpuTime, rand_n);
|
||||||
|
|
||||||
printf("MersenneTwisterGP11213, Throughput = %.4f GNumbers/s, Time = %.5f s, Size = %u Numbers\n",
|
printf("Shutting down...\n");
|
||||||
1.0e-9 * rand_n / gpuTime, gpuTime, rand_n);
|
|
||||||
|
|
||||||
printf("Shutting down...\n");
|
checkCudaErrors(curandDestroyGenerator(prngGPU));
|
||||||
|
checkCudaErrors(curandDestroyGenerator(prngCPU));
|
||||||
|
checkCudaErrors(cudaStreamDestroy(stream));
|
||||||
|
checkCudaErrors(cudaFree(d_Rand));
|
||||||
|
sdkDeleteTimer(&hTimer);
|
||||||
|
checkCudaErrors(cudaFreeHost(h_RandGPU));
|
||||||
|
free(h_RandCPU);
|
||||||
|
|
||||||
checkCudaErrors(curandDestroyGenerator(prngGPU));
|
exit(L1norm < 1e-6 ? EXIT_SUCCESS : EXIT_FAILURE);
|
||||||
checkCudaErrors(curandDestroyGenerator(prngCPU));
|
|
||||||
checkCudaErrors(cudaStreamDestroy(stream));
|
|
||||||
checkCudaErrors(cudaFree(d_Rand));
|
|
||||||
sdkDeleteTimer(&hTimer);
|
|
||||||
free(h_RandGPU);
|
|
||||||
free(h_RandCPU);
|
|
||||||
|
|
||||||
exit(L1norm < 1e-6 ? EXIT_SUCCESS : EXIT_FAILURE);
|
|
||||||
}
|
}
|
||||||
|
|
||||||
|
float compareResults(int rand_n, float *h_RandGPU, float *h_RandCPU) {
|
||||||
|
int i;
|
||||||
|
float rCPU, rGPU, delta;
|
||||||
|
float max_delta = 0.;
|
||||||
|
float sum_delta = 0.;
|
||||||
|
float sum_ref = 0.;
|
||||||
|
|
||||||
float compareResults(int rand_n, float *h_RandGPU, float *h_RandCPU)
|
for (i = 0; i < rand_n; i++) {
|
||||||
{
|
rCPU = h_RandCPU[i];
|
||||||
int i;
|
rGPU = h_RandGPU[i];
|
||||||
float rCPU, rGPU, delta;
|
delta = fabs(rCPU - rGPU);
|
||||||
float max_delta = 0.;
|
sum_delta += delta;
|
||||||
float sum_delta = 0.;
|
sum_ref += fabs(rCPU);
|
||||||
float sum_ref = 0.;
|
|
||||||
|
|
||||||
for (i = 0; i < rand_n; i++)
|
if (delta >= max_delta) {
|
||||||
{
|
max_delta = delta;
|
||||||
rCPU = h_RandCPU[i];
|
|
||||||
rGPU = h_RandGPU[i];
|
|
||||||
delta = fabs(rCPU - rGPU);
|
|
||||||
sum_delta += delta;
|
|
||||||
sum_ref += fabs(rCPU);
|
|
||||||
|
|
||||||
if (delta >= max_delta)
|
|
||||||
{
|
|
||||||
max_delta = delta;
|
|
||||||
}
|
|
||||||
}
|
}
|
||||||
|
}
|
||||||
|
|
||||||
float L1norm = (float)(sum_delta / sum_ref);
|
float L1norm = (float)(sum_delta / sum_ref);
|
||||||
printf("Max absolute error: %E\n", max_delta);
|
printf("Max absolute error: %E\n", max_delta);
|
||||||
printf("L1 norm: %E\n\n", L1norm);
|
printf("L1 norm: %E\n\n", L1norm);
|
||||||
|
|
||||||
return L1norm;
|
return L1norm;
|
||||||
}
|
}
|
||||||
|
|||||||
@@ -38,7 +38,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -108,6 +108,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -34,7 +34,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -104,6 +104,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -27,7 +27,7 @@ x86_64, ppc64le, armv7l
|
|||||||
|
|
||||||
## Prerequisites
|
## Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
||||||
|
|
||||||
## Build and Run
|
## Build and Run
|
||||||
|
|||||||
18
Samples/NV12toBGRandResize/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/NV12toBGRandResize/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"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
Samples/NV12toBGRandResize/.vscode/extensions.json
vendored
Normal file
7
Samples/NV12toBGRandResize/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/NV12toBGRandResize/.vscode/launch.json
vendored
Normal file
10
Samples/NV12toBGRandResize/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/NV12toBGRandResize"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/NV12toBGRandResize/.vscode/tasks.json
vendored
Normal file
15
Samples/NV12toBGRandResize/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@@ -38,7 +38,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -113,6 +113,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -34,7 +34,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -109,6 +109,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -27,7 +27,7 @@ cudaMemcpy2D, cudaMallocManaged
|
|||||||
|
|
||||||
## Prerequisites
|
## Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
|
|
||||||
## Build and Run
|
## Build and Run
|
||||||
|
|
||||||
|
|||||||
18
Samples/UnifiedMemoryPerf/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/UnifiedMemoryPerf/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"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
Samples/UnifiedMemoryPerf/.vscode/extensions.json
vendored
Normal file
7
Samples/UnifiedMemoryPerf/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/UnifiedMemoryPerf/.vscode/launch.json
vendored
Normal file
10
Samples/UnifiedMemoryPerf/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/UnifiedMemoryPerf"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/UnifiedMemoryPerf/.vscode/tasks.json
vendored
Normal file
15
Samples/UnifiedMemoryPerf/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@@ -28,7 +28,7 @@ cudaMallocManaged, cudaStreamAttachMemAsync, cudaMemcpyAsync, cudaMallocHost, cu
|
|||||||
|
|
||||||
## Prerequisites
|
## Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
||||||
|
|
||||||
## Build and Run
|
## Build and Run
|
||||||
|
|||||||
@@ -38,7 +38,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -111,6 +111,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -34,7 +34,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -107,6 +107,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
18
Samples/bandwidthTest/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/bandwidthTest/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"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
Samples/bandwidthTest/.vscode/extensions.json
vendored
Normal file
7
Samples/bandwidthTest/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/bandwidthTest/.vscode/launch.json
vendored
Normal file
10
Samples/bandwidthTest/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/bandwidthTest"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/bandwidthTest/.vscode/tasks.json
vendored
Normal file
15
Samples/bandwidthTest/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@@ -27,7 +27,7 @@ cudaSetDevice, cudaHostAlloc, cudaFree, cudaMallocHost, cudaFreeHost, cudaMemcpy
|
|||||||
|
|
||||||
## Prerequisites
|
## Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
|
|
||||||
## Build and Run
|
## Build and Run
|
||||||
|
|
||||||
|
|||||||
@@ -38,7 +38,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -108,6 +108,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -34,7 +34,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -104,6 +104,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
18
Samples/batchedLabelMarkersAndLabelCompressionNPP/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/batchedLabelMarkersAndLabelCompressionNPP/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"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
Samples/batchedLabelMarkersAndLabelCompressionNPP/.vscode/extensions.json
vendored
Normal file
7
Samples/batchedLabelMarkersAndLabelCompressionNPP/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/batchedLabelMarkersAndLabelCompressionNPP/.vscode/launch.json
vendored
Normal file
10
Samples/batchedLabelMarkersAndLabelCompressionNPP/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/batchedLabelMarkersAndLabelCompressionNPP"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/batchedLabelMarkersAndLabelCompressionNPP/.vscode/tasks.json
vendored
Normal file
15
Samples/batchedLabelMarkersAndLabelCompressionNPP/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@@ -28,7 +28,7 @@ x86_64, ppc64le, armv7l
|
|||||||
|
|
||||||
## Prerequisites
|
## Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
||||||
|
|
||||||
## Build and Run
|
## Build and Run
|
||||||
|
|||||||
@@ -36,11 +36,13 @@
|
|||||||
#include <string.h>
|
#include <string.h>
|
||||||
#include <fstream>
|
#include <fstream>
|
||||||
|
|
||||||
|
#include <cuda_runtime.h>
|
||||||
#include <helper_cuda.h>
|
#include <helper_cuda.h>
|
||||||
|
#include <helper_string.h>
|
||||||
#include <npp.h>
|
#include <npp.h>
|
||||||
|
|
||||||
// Note: If you want to view these images we HIGHLY recommend using imagej
|
// Note: If you want to view these images we HIGHLY recommend using imagej
|
||||||
// which is free on the internet and works on most platforms
|
// which is free on the internet and works on most platforms
|
||||||
// because it is one of the few image viewing apps that can display 32
|
// because it is one of the few image viewing apps that can display 32
|
||||||
// bit integer image data. While it normalizes the data to floating
|
// bit integer image data. While it normalizes the data to floating
|
||||||
// point values for viewing it still provides a good representation of
|
// point values for viewing it still provides a good representation of
|
||||||
@@ -102,11 +104,12 @@ void tearDown() // Clean up and tear down
|
|||||||
if (pUFBatchSrcDstImageListDev != 0) cudaFree(pUFBatchSrcDstImageListDev);
|
if (pUFBatchSrcDstImageListDev != 0) cudaFree(pUFBatchSrcDstImageListDev);
|
||||||
if (pUFBatchSrcImageListDev != 0) cudaFree(pUFBatchSrcImageListDev);
|
if (pUFBatchSrcImageListDev != 0) cudaFree(pUFBatchSrcImageListDev);
|
||||||
if (pUFBatchPerImageCompressedCountListHost != 0)
|
if (pUFBatchPerImageCompressedCountListHost != 0)
|
||||||
free(pUFBatchPerImageCompressedCountListHost);
|
cudaFreeHost(pUFBatchPerImageCompressedCountListHost);
|
||||||
if (pUFBatchSrcDstScratchBufferListHost != 0)
|
if (pUFBatchSrcDstScratchBufferListHost != 0)
|
||||||
free(pUFBatchSrcDstScratchBufferListHost);
|
cudaFreeHost(pUFBatchSrcDstScratchBufferListHost);
|
||||||
if (pUFBatchSrcDstImageListHost != 0) free(pUFBatchSrcDstImageListHost);
|
if (pUFBatchSrcDstImageListHost != 0)
|
||||||
if (pUFBatchSrcImageListHost != 0) free(pUFBatchSrcImageListHost);
|
cudaFreeHost(pUFBatchSrcDstImageListHost);
|
||||||
|
if (pUFBatchSrcImageListHost != 0) cudaFreeHost(pUFBatchSrcImageListHost);
|
||||||
|
|
||||||
for (int j = 0; j < NUMBER_OF_IMAGES; j++) {
|
for (int j = 0; j < NUMBER_OF_IMAGES; j++) {
|
||||||
if (pUFCompressedLabelsScratchBufferDev[j] != 0)
|
if (pUFCompressedLabelsScratchBufferDev[j] != 0)
|
||||||
@@ -115,8 +118,8 @@ void tearDown() // Clean up and tear down
|
|||||||
cudaFree(pUFGenerateLabelsScratchBufferDev[j]);
|
cudaFree(pUFGenerateLabelsScratchBufferDev[j]);
|
||||||
if (pUFLabelDev[j] != 0) cudaFree(pUFLabelDev[j]);
|
if (pUFLabelDev[j] != 0) cudaFree(pUFLabelDev[j]);
|
||||||
if (pInputImageDev[j] != 0) cudaFree(pInputImageDev[j]);
|
if (pInputImageDev[j] != 0) cudaFree(pInputImageDev[j]);
|
||||||
if (pUFLabelHost[j] != 0) free(pUFLabelHost[j]);
|
if (pUFLabelHost[j] != 0) cudaFreeHost(pUFLabelHost[j]);
|
||||||
if (pInputImageHost[j] != 0) free(pInputImageHost[j]);
|
if (pInputImageHost[j] != 0) cudaFreeHost(pInputImageHost[j]);
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
@@ -177,7 +180,7 @@ int loadRaw8BitImage(Npp8u *pImage, int nWidth, int nHeight, int nImage) {
|
|||||||
exit(EXIT_WAIVED);
|
exit(EXIT_WAIVED);
|
||||||
}
|
}
|
||||||
|
|
||||||
bmpFile = fopen(InputFile, "rb");
|
FOPEN(bmpFile, InputFile, "rb");
|
||||||
} else if (nImage == 1) {
|
} else if (nImage == 1) {
|
||||||
if (nWidth != 512 || nHeight != 512) return -1;
|
if (nWidth != 512 || nHeight != 512) return -1;
|
||||||
const char *fileName = "CT_skull_512x512_8u.raw";
|
const char *fileName = "CT_skull_512x512_8u.raw";
|
||||||
@@ -187,7 +190,7 @@ int loadRaw8BitImage(Npp8u *pImage, int nWidth, int nHeight, int nImage) {
|
|||||||
exit(EXIT_WAIVED);
|
exit(EXIT_WAIVED);
|
||||||
}
|
}
|
||||||
|
|
||||||
bmpFile = fopen(InputFile, "rb");
|
FOPEN(bmpFile, InputFile, "rb");
|
||||||
} else if (nImage == 2) {
|
} else if (nImage == 2) {
|
||||||
if (nWidth != 509 || nHeight != 335) return -1;
|
if (nWidth != 509 || nHeight != 335) return -1;
|
||||||
const char *fileName = "PCB_METAL_509x335_8u.raw";
|
const char *fileName = "PCB_METAL_509x335_8u.raw";
|
||||||
@@ -197,7 +200,7 @@ int loadRaw8BitImage(Npp8u *pImage, int nWidth, int nHeight, int nImage) {
|
|||||||
exit(EXIT_WAIVED);
|
exit(EXIT_WAIVED);
|
||||||
}
|
}
|
||||||
|
|
||||||
bmpFile = fopen(InputFile, "rb");
|
FOPEN(bmpFile, InputFile, "rb");
|
||||||
} else if (nImage == 3) {
|
} else if (nImage == 3) {
|
||||||
if (nWidth != 1024 || nHeight != 683) return -1;
|
if (nWidth != 1024 || nHeight != 683) return -1;
|
||||||
const char *fileName = "PCB2_1024x683_8u.raw";
|
const char *fileName = "PCB2_1024x683_8u.raw";
|
||||||
@@ -207,7 +210,7 @@ int loadRaw8BitImage(Npp8u *pImage, int nWidth, int nHeight, int nImage) {
|
|||||||
exit(EXIT_WAIVED);
|
exit(EXIT_WAIVED);
|
||||||
}
|
}
|
||||||
|
|
||||||
bmpFile = fopen(InputFile, "rb");
|
FOPEN(bmpFile, InputFile, "rb");
|
||||||
} else if (nImage == 4) {
|
} else if (nImage == 4) {
|
||||||
if (nWidth != 1280 || nHeight != 720) return -1;
|
if (nWidth != 1280 || nHeight != 720) return -1;
|
||||||
const char *fileName = "PCB_1280x720_8u.raw";
|
const char *fileName = "PCB_1280x720_8u.raw";
|
||||||
@@ -217,7 +220,7 @@ int loadRaw8BitImage(Npp8u *pImage, int nWidth, int nHeight, int nImage) {
|
|||||||
exit(EXIT_WAIVED);
|
exit(EXIT_WAIVED);
|
||||||
}
|
}
|
||||||
|
|
||||||
bmpFile = fopen(InputFile, "rb");
|
FOPEN(bmpFile, InputFile, "rb");
|
||||||
} else {
|
} else {
|
||||||
printf("Input file load failed.\n");
|
printf("Input file load failed.\n");
|
||||||
return -1;
|
return -1;
|
||||||
@@ -347,9 +350,11 @@ int main(int argc, char **argv) {
|
|||||||
oSizeROI[nImage].width * sizeof(Npp32u) * oSizeROI[nImage].height);
|
oSizeROI[nImage].width * sizeof(Npp32u) * oSizeROI[nImage].height);
|
||||||
if (cudaError != cudaSuccess) return NPP_MEMORY_ALLOCATION_ERR;
|
if (cudaError != cudaSuccess) return NPP_MEMORY_ALLOCATION_ERR;
|
||||||
|
|
||||||
pInputImageHost[nImage] = reinterpret_cast<Npp8u *>(malloc(
|
checkCudaErrors(cudaMallocHost(
|
||||||
|
&(pInputImageHost[nImage]),
|
||||||
oSizeROI[nImage].width * sizeof(Npp8u) * oSizeROI[nImage].height));
|
oSizeROI[nImage].width * sizeof(Npp8u) * oSizeROI[nImage].height));
|
||||||
pUFLabelHost[nImage] = reinterpret_cast<Npp32u *>(malloc(
|
checkCudaErrors(cudaMallocHost(
|
||||||
|
&(pUFLabelHost[nImage]),
|
||||||
oSizeROI[nImage].width * sizeof(Npp32u) * oSizeROI[nImage].height));
|
oSizeROI[nImage].width * sizeof(Npp32u) * oSizeROI[nImage].height));
|
||||||
|
|
||||||
// Use UF functions throughout this sample.
|
// Use UF functions throughout this sample.
|
||||||
@@ -409,15 +414,15 @@ int main(int argc, char **argv) {
|
|||||||
}
|
}
|
||||||
|
|
||||||
if (nImage == 0)
|
if (nImage == 0)
|
||||||
bmpFile = fopen(LabelMarkersOutputFile0.c_str(), "wb");
|
FOPEN(bmpFile, LabelMarkersOutputFile0.c_str(), "wb");
|
||||||
else if (nImage == 1)
|
else if (nImage == 1)
|
||||||
bmpFile = fopen(LabelMarkersOutputFile1.c_str(), "wb");
|
FOPEN(bmpFile, LabelMarkersOutputFile1.c_str(), "wb");
|
||||||
else if (nImage == 2)
|
else if (nImage == 2)
|
||||||
bmpFile = fopen(LabelMarkersOutputFile2.c_str(), "wb");
|
FOPEN(bmpFile, LabelMarkersOutputFile2.c_str(), "wb");
|
||||||
else if (nImage == 3)
|
else if (nImage == 3)
|
||||||
bmpFile = fopen(LabelMarkersOutputFile3.c_str(), "wb");
|
FOPEN(bmpFile, LabelMarkersOutputFile3.c_str(), "wb");
|
||||||
else if (nImage == 4)
|
else if (nImage == 4)
|
||||||
bmpFile = fopen(LabelMarkersOutputFile4.c_str(), "wb");
|
FOPEN(bmpFile, LabelMarkersOutputFile4.c_str(), "wb");
|
||||||
|
|
||||||
if (bmpFile == NULL) return -1;
|
if (bmpFile == NULL) return -1;
|
||||||
size_t nSize = 0;
|
size_t nSize = 0;
|
||||||
@@ -478,15 +483,15 @@ int main(int argc, char **argv) {
|
|||||||
}
|
}
|
||||||
|
|
||||||
if (nImage == 0)
|
if (nImage == 0)
|
||||||
bmpFile = fopen(CompressedMarkerLabelsOutputFile0.c_str(), "wb");
|
FOPEN(bmpFile, CompressedMarkerLabelsOutputFile0.c_str(), "wb");
|
||||||
else if (nImage == 1)
|
else if (nImage == 1)
|
||||||
bmpFile = fopen(CompressedMarkerLabelsOutputFile1.c_str(), "wb");
|
FOPEN(bmpFile, CompressedMarkerLabelsOutputFile1.c_str(), "wb");
|
||||||
else if (nImage == 2)
|
else if (nImage == 2)
|
||||||
bmpFile = fopen(CompressedMarkerLabelsOutputFile2.c_str(), "wb");
|
FOPEN(bmpFile, CompressedMarkerLabelsOutputFile2.c_str(), "wb");
|
||||||
else if (nImage == 3)
|
else if (nImage == 3)
|
||||||
bmpFile = fopen(CompressedMarkerLabelsOutputFile3.c_str(), "wb");
|
FOPEN(bmpFile, CompressedMarkerLabelsOutputFile3.c_str(), "wb");
|
||||||
else if (nImage == 4)
|
else if (nImage == 4)
|
||||||
bmpFile = fopen(CompressedMarkerLabelsOutputFile4.c_str(), "wb");
|
FOPEN(bmpFile, CompressedMarkerLabelsOutputFile4.c_str(), "wb");
|
||||||
|
|
||||||
if (bmpFile == NULL) return -1;
|
if (bmpFile == NULL) return -1;
|
||||||
nSize = 0;
|
nSize = 0;
|
||||||
@@ -554,10 +559,11 @@ int main(int argc, char **argv) {
|
|||||||
cudaMalloc((void **)&pUFBatchSrcDstImageListDev, nBatchImageListBytes);
|
cudaMalloc((void **)&pUFBatchSrcDstImageListDev, nBatchImageListBytes);
|
||||||
if (cudaError != cudaSuccess) return NPP_MEMORY_ALLOCATION_ERR;
|
if (cudaError != cudaSuccess) return NPP_MEMORY_ALLOCATION_ERR;
|
||||||
|
|
||||||
pUFBatchSrcImageListHost =
|
checkCudaErrors(
|
||||||
reinterpret_cast<NppiImageDescriptor *>(malloc(nBatchImageListBytes));
|
cudaMallocHost((void **)&pUFBatchSrcImageListHost, nBatchImageListBytes));
|
||||||
pUFBatchSrcDstImageListHost =
|
|
||||||
reinterpret_cast<NppiImageDescriptor *>(malloc(nBatchImageListBytes));
|
checkCudaErrors(cudaMallocHost((void **)&pUFBatchSrcDstImageListHost,
|
||||||
|
nBatchImageListBytes));
|
||||||
|
|
||||||
NppiSize oMaxROISize = {0, 0};
|
NppiSize oMaxROISize = {0, 0};
|
||||||
|
|
||||||
@@ -620,15 +626,15 @@ int main(int argc, char **argv) {
|
|||||||
// Save output to files
|
// Save output to files
|
||||||
for (int nImage = 0; nImage < NUMBER_OF_IMAGES; nImage++) {
|
for (int nImage = 0; nImage < NUMBER_OF_IMAGES; nImage++) {
|
||||||
if (nImage == 0)
|
if (nImage == 0)
|
||||||
bmpFile = fopen(LabelMarkersBatchOutputFile0.c_str(), "wb");
|
FOPEN(bmpFile, LabelMarkersBatchOutputFile0.c_str(), "wb");
|
||||||
else if (nImage == 1)
|
else if (nImage == 1)
|
||||||
bmpFile = fopen(LabelMarkersBatchOutputFile1.c_str(), "wb");
|
FOPEN(bmpFile, LabelMarkersBatchOutputFile1.c_str(), "wb");
|
||||||
else if (nImage == 2)
|
else if (nImage == 2)
|
||||||
bmpFile = fopen(LabelMarkersBatchOutputFile2.c_str(), "wb");
|
FOPEN(bmpFile, LabelMarkersBatchOutputFile2.c_str(), "wb");
|
||||||
else if (nImage == 3)
|
else if (nImage == 3)
|
||||||
bmpFile = fopen(LabelMarkersBatchOutputFile3.c_str(), "wb");
|
FOPEN(bmpFile, LabelMarkersBatchOutputFile3.c_str(), "wb");
|
||||||
else if (nImage == 4)
|
else if (nImage == 4)
|
||||||
bmpFile = fopen(LabelMarkersBatchOutputFile4.c_str(), "wb");
|
FOPEN(bmpFile, LabelMarkersBatchOutputFile4.c_str(), "wb");
|
||||||
|
|
||||||
if (bmpFile == NULL) return -1;
|
if (bmpFile == NULL) return -1;
|
||||||
size_t nSize = 0;
|
size_t nSize = 0;
|
||||||
@@ -652,12 +658,13 @@ int main(int argc, char **argv) {
|
|||||||
|
|
||||||
// Allocate host side scratch buffer point and size list and initialize with
|
// Allocate host side scratch buffer point and size list and initialize with
|
||||||
// device scratch buffer pointers
|
// device scratch buffer pointers
|
||||||
pUFBatchSrcDstScratchBufferListHost =
|
checkCudaErrors(
|
||||||
reinterpret_cast<NppiBufferDescriptor *>(
|
cudaMallocHost((void **)&pUFBatchSrcDstScratchBufferListHost,
|
||||||
malloc(NUMBER_OF_IMAGES * sizeof(NppiBufferDescriptor)));
|
NUMBER_OF_IMAGES * sizeof(NppiBufferDescriptor)));
|
||||||
|
|
||||||
pUFBatchPerImageCompressedCountListHost =
|
checkCudaErrors(
|
||||||
reinterpret_cast<Npp32u *>(malloc(NUMBER_OF_IMAGES * sizeof(Npp32u)));
|
cudaMallocHost((void **)&pUFBatchPerImageCompressedCountListHost,
|
||||||
|
+NUMBER_OF_IMAGES * sizeof(Npp32u)));
|
||||||
|
|
||||||
// Start buffer pointer at beginning of full per image buffer list sized
|
// Start buffer pointer at beginning of full per image buffer list sized
|
||||||
// pUFCompressedLabelsScratchBufferDev[0]
|
// pUFCompressedLabelsScratchBufferDev[0]
|
||||||
@@ -728,15 +735,15 @@ int main(int argc, char **argv) {
|
|||||||
// Save compressed label images into files
|
// Save compressed label images into files
|
||||||
for (int nImage = 0; nImage < NUMBER_OF_IMAGES; nImage++) {
|
for (int nImage = 0; nImage < NUMBER_OF_IMAGES; nImage++) {
|
||||||
if (nImage == 0)
|
if (nImage == 0)
|
||||||
bmpFile = fopen(CompressedMarkerLabelsBatchOutputFile0.c_str(), "wb");
|
FOPEN(bmpFile, CompressedMarkerLabelsBatchOutputFile0.c_str(), "wb");
|
||||||
else if (nImage == 1)
|
else if (nImage == 1)
|
||||||
bmpFile = fopen(CompressedMarkerLabelsBatchOutputFile1.c_str(), "wb");
|
FOPEN(bmpFile, CompressedMarkerLabelsBatchOutputFile1.c_str(), "wb");
|
||||||
else if (nImage == 2)
|
else if (nImage == 2)
|
||||||
bmpFile = fopen(CompressedMarkerLabelsBatchOutputFile2.c_str(), "wb");
|
FOPEN(bmpFile, CompressedMarkerLabelsBatchOutputFile2.c_str(), "wb");
|
||||||
else if (nImage == 3)
|
else if (nImage == 3)
|
||||||
bmpFile = fopen(CompressedMarkerLabelsBatchOutputFile3.c_str(), "wb");
|
FOPEN(bmpFile, CompressedMarkerLabelsBatchOutputFile3.c_str(), "wb");
|
||||||
else if (nImage == 4)
|
else if (nImage == 4)
|
||||||
bmpFile = fopen(CompressedMarkerLabelsBatchOutputFile4.c_str(), "wb");
|
FOPEN(bmpFile, CompressedMarkerLabelsBatchOutputFile4.c_str(), "wb");
|
||||||
|
|
||||||
if (bmpFile == NULL) return -1;
|
if (bmpFile == NULL) return -1;
|
||||||
size_t nSize = 0;
|
size_t nSize = 0;
|
||||||
|
|||||||
@@ -38,7 +38,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -108,6 +108,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -34,7 +34,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -104,6 +104,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
18
Samples/bf16TensorCoreGemm/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/bf16TensorCoreGemm/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"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
Samples/bf16TensorCoreGemm/.vscode/extensions.json
vendored
Normal file
7
Samples/bf16TensorCoreGemm/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/bf16TensorCoreGemm/.vscode/launch.json
vendored
Normal file
10
Samples/bf16TensorCoreGemm/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/bf16TensorCoreGemm"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/bf16TensorCoreGemm/.vscode/tasks.json
vendored
Normal file
15
Samples/bf16TensorCoreGemm/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@@ -27,7 +27,7 @@ cudaMallocManaged, cudaDeviceSynchronize, cudaFuncSetAttribute, cudaEventCreate,
|
|||||||
|
|
||||||
## Prerequisites
|
## Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
|
|
||||||
## Build and Run
|
## Build and Run
|
||||||
|
|
||||||
|
|||||||
@@ -38,7 +38,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -108,6 +108,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -34,7 +34,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -104,6 +104,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
18
Samples/binaryPartitionCG/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/binaryPartitionCG/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"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
Samples/binaryPartitionCG/.vscode/extensions.json
vendored
Normal file
7
Samples/binaryPartitionCG/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/binaryPartitionCG/.vscode/launch.json
vendored
Normal file
10
Samples/binaryPartitionCG/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/binaryPartitionCG"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/binaryPartitionCG/.vscode/tasks.json
vendored
Normal file
15
Samples/binaryPartitionCG/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@@ -24,7 +24,7 @@ x86_64, ppc64le, armv7l
|
|||||||
|
|
||||||
## Prerequisites
|
## Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
|
|
||||||
## Build and Run
|
## Build and Run
|
||||||
|
|
||||||
|
|||||||
@@ -31,14 +31,16 @@
|
|||||||
* 1.) Each thread loads a value from random array.
|
* 1.) Each thread loads a value from random array.
|
||||||
* 2.) then checks if it is odd or even.
|
* 2.) then checks if it is odd or even.
|
||||||
* 3.) create binary partition group based on the above predicate
|
* 3.) create binary partition group based on the above predicate
|
||||||
* 4.) we count the number of odd/even in the group based on size of the binary groups
|
* 4.) we count the number of odd/even in the group based on size of the binary
|
||||||
|
groups
|
||||||
* 5.) write it global counter of odd.
|
* 5.) write it global counter of odd.
|
||||||
* 6.) sum the values loaded by individual threads(using reduce) and write it to global
|
* 6.) sum the values loaded by individual threads(using reduce) and write it to
|
||||||
* even & odd elements sum.
|
global even & odd elements sum.
|
||||||
*
|
*
|
||||||
* **NOTE** : binary_partition results in splitting warp into divergent thread groups
|
* **NOTE** :
|
||||||
this is not good from performance perspective, but in cases where warp
|
* binary_partition results in splitting warp into divergent thread groups
|
||||||
divergence is inevitable one can use binary_partition group.
|
* this is not good from performance perspective, but in cases where warp
|
||||||
|
* divergence is inevitable one can use binary_partition group.
|
||||||
*/
|
*/
|
||||||
|
|
||||||
#include <stdio.h>
|
#include <stdio.h>
|
||||||
@@ -48,108 +50,110 @@
|
|||||||
|
|
||||||
namespace cg = cooperative_groups;
|
namespace cg = cooperative_groups;
|
||||||
|
|
||||||
void initOddEvenArr(int *inputArr, unsigned int size)
|
void initOddEvenArr(int *inputArr, unsigned int size) {
|
||||||
{
|
for (int i = 0; i < size; i++) {
|
||||||
for (int i=0; i < size; i++)
|
inputArr[i] = rand() % 50;
|
||||||
{
|
}
|
||||||
inputArr[i] = rand() % 50;
|
|
||||||
}
|
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
||||||
/**
|
/**
|
||||||
* CUDA kernel device code
|
* CUDA kernel device code
|
||||||
*
|
*
|
||||||
* Creates cooperative groups and performs odd/even counting & summation.
|
* Creates cooperative groups and performs odd/even counting & summation.
|
||||||
*/
|
*/
|
||||||
__global__ void oddEvenCountAndSumCG(int *inputArr, int *numOfOdds, int *sumOfOddAndEvens, unsigned int size)
|
__global__ void oddEvenCountAndSumCG(int *inputArr, int *numOfOdds,
|
||||||
{
|
int *sumOfOddAndEvens, unsigned int size) {
|
||||||
cg::thread_block cta = cg::this_thread_block();
|
cg::thread_block cta = cg::this_thread_block();
|
||||||
cg::grid_group grid = cg::this_grid();
|
cg::grid_group grid = cg::this_grid();
|
||||||
cg::thread_block_tile<32> tile32 = cg::tiled_partition<32>(cta);
|
cg::thread_block_tile<32> tile32 = cg::tiled_partition<32>(cta);
|
||||||
|
|
||||||
for (int i = grid.thread_rank(); i < size; i += grid.size())
|
for (int i = grid.thread_rank(); i < size; i += grid.size()) {
|
||||||
|
int elem = inputArr[i];
|
||||||
|
auto subTile = cg::binary_partition(tile32, elem & 1);
|
||||||
|
if (elem & 1) // Odd numbers group
|
||||||
{
|
{
|
||||||
int elem = inputArr[i];
|
int oddGroupSum = cg::reduce(subTile, elem, cg::plus<int>());
|
||||||
auto subTile = cg::binary_partition(tile32, elem & 1);
|
|
||||||
if (elem & 1) // Odd numbers group
|
|
||||||
{
|
|
||||||
int oddGroupSum = cg::reduce(subTile, elem, cg::plus<int>());
|
|
||||||
|
|
||||||
if (subTile.thread_rank() == 0)
|
if (subTile.thread_rank() == 0) {
|
||||||
{
|
// Add number of odds present in this group of Odds.
|
||||||
// Add number of odds present in this group of Odds.
|
atomicAdd(numOfOdds, subTile.size());
|
||||||
atomicAdd(numOfOdds, subTile.size());
|
|
||||||
|
|
||||||
// Add local reduction of odds present in this group of Odds.
|
// Add local reduction of odds present in this group of Odds.
|
||||||
atomicAdd(&sumOfOddAndEvens[0], oddGroupSum);
|
atomicAdd(&sumOfOddAndEvens[0], oddGroupSum);
|
||||||
|
}
|
||||||
|
} else // Even numbers group
|
||||||
|
{
|
||||||
|
int evenGroupSum = cg::reduce(subTile, elem, cg::plus<int>());
|
||||||
|
|
||||||
}
|
if (subTile.thread_rank() == 0) {
|
||||||
}
|
// Add local reduction of even present in this group of evens.
|
||||||
else // Even numbers group
|
atomicAdd(&sumOfOddAndEvens[1], evenGroupSum);
|
||||||
{
|
}
|
||||||
int evenGroupSum = cg::reduce(subTile, elem, cg::plus<int>());
|
|
||||||
|
|
||||||
if (subTile.thread_rank() == 0)
|
|
||||||
{
|
|
||||||
// Add local reduction of even present in this group of evens.
|
|
||||||
atomicAdd(&sumOfOddAndEvens[1], evenGroupSum);
|
|
||||||
}
|
|
||||||
}
|
|
||||||
// reconverge warp so for next loop iteration we ensure convergence of
|
|
||||||
// above diverged threads to perform coalesced loads of inputArr.
|
|
||||||
cg::sync(tile32);
|
|
||||||
}
|
}
|
||||||
|
// reconverge warp so for next loop iteration we ensure convergence of
|
||||||
|
// above diverged threads to perform coalesced loads of inputArr.
|
||||||
|
cg::sync(tile32);
|
||||||
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
|
|
||||||
/**
|
/**
|
||||||
* Host main routine
|
* Host main routine
|
||||||
*/
|
*/
|
||||||
int main(int argc, const char **argv)
|
int main(int argc, const char **argv) {
|
||||||
{
|
int deviceId = findCudaDevice(argc, argv);
|
||||||
int deviceId = findCudaDevice(argc, argv);
|
int *h_inputArr, *d_inputArr;
|
||||||
int *h_inputArr, *d_inputArr;
|
int *h_numOfOdds, *d_numOfOdds;
|
||||||
int *h_numOfOdds, *d_numOfOdds;
|
int *h_sumOfOddEvenElems, *d_sumOfOddEvenElems;
|
||||||
int *h_sumOfOddEvenElems, *d_sumOfOddEvenElems;
|
unsigned int arrSize = 1024 * 100;
|
||||||
unsigned int arrSize = 1024 * 100;
|
|
||||||
|
|
||||||
h_inputArr = new int[arrSize];
|
checkCudaErrors(cudaMallocHost(&h_inputArr, sizeof(int) * arrSize));
|
||||||
h_numOfOdds = new int[1];
|
checkCudaErrors(cudaMallocHost(&h_numOfOdds, sizeof(int)));
|
||||||
h_sumOfOddEvenElems = new int[2];
|
checkCudaErrors(cudaMallocHost(&h_sumOfOddEvenElems, sizeof(int) * 2));
|
||||||
initOddEvenArr(h_inputArr, arrSize);
|
initOddEvenArr(h_inputArr, arrSize);
|
||||||
|
|
||||||
cudaStream_t stream;
|
|
||||||
checkCudaErrors(cudaStreamCreateWithFlags(&stream, cudaStreamNonBlocking));
|
|
||||||
checkCudaErrors(cudaMalloc(&d_inputArr, sizeof(int)*arrSize));
|
|
||||||
checkCudaErrors(cudaMalloc(&d_numOfOdds, sizeof(int)));
|
|
||||||
checkCudaErrors(cudaMalloc(&d_sumOfOddEvenElems, sizeof(int)*2));
|
|
||||||
|
|
||||||
checkCudaErrors(cudaMemcpyAsync(d_inputArr, h_inputArr, sizeof(int)*arrSize, cudaMemcpyHostToDevice, stream));
|
cudaStream_t stream;
|
||||||
checkCudaErrors(cudaMemsetAsync(d_numOfOdds, 0, sizeof(int), stream));
|
checkCudaErrors(cudaStreamCreateWithFlags(&stream, cudaStreamNonBlocking));
|
||||||
checkCudaErrors(cudaMemsetAsync(d_sumOfOddEvenElems, 0, 2*sizeof(int), stream));
|
checkCudaErrors(cudaMalloc(&d_inputArr, sizeof(int) * arrSize));
|
||||||
|
checkCudaErrors(cudaMalloc(&d_numOfOdds, sizeof(int)));
|
||||||
|
checkCudaErrors(cudaMalloc(&d_sumOfOddEvenElems, sizeof(int) * 2));
|
||||||
|
|
||||||
//Launch the kernel
|
checkCudaErrors(cudaMemcpyAsync(d_inputArr, h_inputArr, sizeof(int) * arrSize,
|
||||||
int threadsPerBlock=1024;
|
cudaMemcpyHostToDevice, stream));
|
||||||
int blocksPerGrid = arrSize / threadsPerBlock;
|
checkCudaErrors(cudaMemsetAsync(d_numOfOdds, 0, sizeof(int), stream));
|
||||||
|
checkCudaErrors(
|
||||||
|
cudaMemsetAsync(d_sumOfOddEvenElems, 0, 2 * sizeof(int), stream));
|
||||||
|
|
||||||
printf("\nLaunching %d blocks with %d threads...\n\n",blocksPerGrid, threadsPerBlock);
|
// Launch the kernel
|
||||||
|
int threadsPerBlock = 0;
|
||||||
|
int blocksPerGrid = 0;
|
||||||
|
checkCudaErrors(cudaOccupancyMaxPotentialBlockSize(
|
||||||
|
&blocksPerGrid, &threadsPerBlock, oddEvenCountAndSumCG, 0, 0));
|
||||||
|
|
||||||
oddEvenCountAndSumCG<<<blocksPerGrid, threadsPerBlock, 0, stream>>>(d_inputArr, d_numOfOdds, d_sumOfOddEvenElems, arrSize);
|
printf("\nLaunching %d blocks with %d threads...\n\n", blocksPerGrid,
|
||||||
|
threadsPerBlock);
|
||||||
|
|
||||||
checkCudaErrors(cudaMemcpyAsync(h_numOfOdds, d_numOfOdds, sizeof(int), cudaMemcpyDeviceToHost, stream));
|
oddEvenCountAndSumCG<<<blocksPerGrid, threadsPerBlock, 0, stream>>>(
|
||||||
checkCudaErrors(cudaMemcpyAsync(h_sumOfOddEvenElems, d_sumOfOddEvenElems, 2*sizeof(int), cudaMemcpyDeviceToHost, stream));
|
d_inputArr, d_numOfOdds, d_sumOfOddEvenElems, arrSize);
|
||||||
|
|
||||||
printf("Array size = %d Num of Odds = %d Sum of Odds = %d Sum of Evens %d\n", arrSize, h_numOfOdds[0], h_sumOfOddEvenElems[0], h_sumOfOddEvenElems[1]);
|
|
||||||
printf("\n...Done.\n\n");
|
|
||||||
|
|
||||||
delete[] h_inputArr;
|
checkCudaErrors(cudaMemcpyAsync(h_numOfOdds, d_numOfOdds, sizeof(int),
|
||||||
delete[] h_numOfOdds;
|
cudaMemcpyDeviceToHost, stream));
|
||||||
delete[] h_sumOfOddEvenElems;
|
checkCudaErrors(cudaMemcpyAsync(h_sumOfOddEvenElems, d_sumOfOddEvenElems,
|
||||||
|
2 * sizeof(int), cudaMemcpyDeviceToHost,
|
||||||
|
stream));
|
||||||
|
checkCudaErrors(cudaStreamSynchronize(stream));
|
||||||
|
|
||||||
checkCudaErrors(cudaFree(d_inputArr));
|
printf("Array size = %d Num of Odds = %d Sum of Odds = %d Sum of Evens %d\n",
|
||||||
checkCudaErrors(cudaFree(d_numOfOdds));
|
arrSize, h_numOfOdds[0], h_sumOfOddEvenElems[0],
|
||||||
checkCudaErrors(cudaFree(d_sumOfOddEvenElems));
|
h_sumOfOddEvenElems[1]);
|
||||||
|
printf("\n...Done.\n\n");
|
||||||
|
|
||||||
return EXIT_SUCCESS;
|
checkCudaErrors(cudaFreeHost(h_inputArr));
|
||||||
|
checkCudaErrors(cudaFreeHost(h_numOfOdds));
|
||||||
|
checkCudaErrors(cudaFreeHost(h_sumOfOddEvenElems));
|
||||||
|
|
||||||
|
checkCudaErrors(cudaFree(d_inputArr));
|
||||||
|
checkCudaErrors(cudaFree(d_numOfOdds));
|
||||||
|
checkCudaErrors(cudaFree(d_sumOfOddEvenElems));
|
||||||
|
|
||||||
|
return EXIT_SUCCESS;
|
||||||
}
|
}
|
||||||
|
|||||||
@@ -38,7 +38,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -108,6 +108,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -34,7 +34,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -104,6 +104,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
18
Samples/boxFilterNPP/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/boxFilterNPP/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"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
Samples/boxFilterNPP/.vscode/extensions.json
vendored
Normal file
7
Samples/boxFilterNPP/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/boxFilterNPP/.vscode/launch.json
vendored
Normal file
10
Samples/boxFilterNPP/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/boxFilterNPP"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/boxFilterNPP/.vscode/tasks.json
vendored
Normal file
15
Samples/boxFilterNPP/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@@ -27,7 +27,7 @@ x86_64, ppc64le, armv7l
|
|||||||
|
|
||||||
## Prerequisites
|
## Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
||||||
|
|
||||||
## Build and Run
|
## Build and Run
|
||||||
|
|||||||
@@ -38,7 +38,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -118,6 +118,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -34,7 +34,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -114,6 +114,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
18
Samples/cannyEdgeDetectorNPP/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/cannyEdgeDetectorNPP/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"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
Samples/cannyEdgeDetectorNPP/.vscode/extensions.json
vendored
Normal file
7
Samples/cannyEdgeDetectorNPP/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/cannyEdgeDetectorNPP/.vscode/launch.json
vendored
Normal file
10
Samples/cannyEdgeDetectorNPP/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/cannyEdgeDetectorNPP"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/cannyEdgeDetectorNPP/.vscode/tasks.json
vendored
Normal file
15
Samples/cannyEdgeDetectorNPP/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@@ -27,7 +27,7 @@ x86_64, ppc64le, armv7l
|
|||||||
|
|
||||||
## Prerequisites
|
## Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
||||||
|
|
||||||
## Build and Run
|
## Build and Run
|
||||||
|
|||||||
@@ -38,7 +38,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -118,6 +118,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -34,7 +34,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -114,6 +114,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
18
Samples/cdpQuadtree/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/cdpQuadtree/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"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
Samples/cdpQuadtree/.vscode/extensions.json
vendored
Normal file
7
Samples/cdpQuadtree/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/cdpQuadtree/.vscode/launch.json
vendored
Normal file
10
Samples/cdpQuadtree/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/cdpQuadtree"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/cdpQuadtree/.vscode/tasks.json
vendored
Normal file
15
Samples/cdpQuadtree/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
370
Samples/cdpQuadtree/Makefile
Normal file
370
Samples/cdpQuadtree/Makefile
Normal file
@@ -0,0 +1,370 @@
|
|||||||
|
################################################################################
|
||||||
|
# Copyright (c) 2019, 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.
|
||||||
|
#
|
||||||
|
################################################################################
|
||||||
|
#
|
||||||
|
# Makefile project only supported on Mac OS X and Linux Platforms)
|
||||||
|
#
|
||||||
|
################################################################################
|
||||||
|
|
||||||
|
# Location of the CUDA Toolkit
|
||||||
|
CUDA_PATH ?= /usr/local/cuda
|
||||||
|
|
||||||
|
##############################
|
||||||
|
# start deprecated interface #
|
||||||
|
##############################
|
||||||
|
ifeq ($(x86_64),1)
|
||||||
|
$(info WARNING - x86_64 variable has been deprecated)
|
||||||
|
$(info WARNING - please use TARGET_ARCH=x86_64 instead)
|
||||||
|
TARGET_ARCH ?= x86_64
|
||||||
|
endif
|
||||||
|
ifeq ($(ARMv7),1)
|
||||||
|
$(info WARNING - ARMv7 variable has been deprecated)
|
||||||
|
$(info WARNING - please use TARGET_ARCH=armv7l instead)
|
||||||
|
TARGET_ARCH ?= armv7l
|
||||||
|
endif
|
||||||
|
ifeq ($(aarch64),1)
|
||||||
|
$(info WARNING - aarch64 variable has been deprecated)
|
||||||
|
$(info WARNING - please use TARGET_ARCH=aarch64 instead)
|
||||||
|
TARGET_ARCH ?= aarch64
|
||||||
|
endif
|
||||||
|
ifeq ($(ppc64le),1)
|
||||||
|
$(info WARNING - ppc64le variable has been deprecated)
|
||||||
|
$(info WARNING - please use TARGET_ARCH=ppc64le instead)
|
||||||
|
TARGET_ARCH ?= ppc64le
|
||||||
|
endif
|
||||||
|
ifneq ($(GCC),)
|
||||||
|
$(info WARNING - GCC variable has been deprecated)
|
||||||
|
$(info WARNING - please use HOST_COMPILER=$(GCC) instead)
|
||||||
|
HOST_COMPILER ?= $(GCC)
|
||||||
|
endif
|
||||||
|
ifneq ($(abi),)
|
||||||
|
$(error ERROR - abi variable has been removed)
|
||||||
|
endif
|
||||||
|
############################
|
||||||
|
# end deprecated interface #
|
||||||
|
############################
|
||||||
|
|
||||||
|
# architecture
|
||||||
|
HOST_ARCH := $(shell uname -m)
|
||||||
|
TARGET_ARCH ?= $(HOST_ARCH)
|
||||||
|
ifneq (,$(filter $(TARGET_ARCH),x86_64 aarch64 sbsa ppc64le armv7l))
|
||||||
|
ifneq ($(TARGET_ARCH),$(HOST_ARCH))
|
||||||
|
ifneq (,$(filter $(TARGET_ARCH),x86_64 aarch64 sbsa ppc64le))
|
||||||
|
TARGET_SIZE := 64
|
||||||
|
else ifneq (,$(filter $(TARGET_ARCH),armv7l))
|
||||||
|
TARGET_SIZE := 32
|
||||||
|
endif
|
||||||
|
else
|
||||||
|
TARGET_SIZE := $(shell getconf LONG_BIT)
|
||||||
|
endif
|
||||||
|
else
|
||||||
|
$(error ERROR - unsupported value $(TARGET_ARCH) for TARGET_ARCH!)
|
||||||
|
endif
|
||||||
|
|
||||||
|
# sbsa and aarch64 systems look similar. Need to differentiate them at host level for now.
|
||||||
|
ifeq ($(HOST_ARCH),aarch64)
|
||||||
|
ifeq ($(CUDA_PATH)/targets/sbsa-linux,$(shell ls -1d $(CUDA_PATH)/targets/sbsa-linux 2>/dev/null))
|
||||||
|
HOST_ARCH := sbsa
|
||||||
|
TARGET_ARCH := sbsa
|
||||||
|
endif
|
||||||
|
endif
|
||||||
|
|
||||||
|
ifneq ($(TARGET_ARCH),$(HOST_ARCH))
|
||||||
|
ifeq (,$(filter $(HOST_ARCH)-$(TARGET_ARCH),aarch64-armv7l x86_64-armv7l x86_64-aarch64 x86_64-sbsa x86_64-ppc64le))
|
||||||
|
$(error ERROR - cross compiling from $(HOST_ARCH) to $(TARGET_ARCH) is not supported!)
|
||||||
|
endif
|
||||||
|
endif
|
||||||
|
|
||||||
|
# When on native aarch64 system with userspace of 32-bit, change TARGET_ARCH to armv7l
|
||||||
|
ifeq ($(HOST_ARCH)-$(TARGET_ARCH)-$(TARGET_SIZE),aarch64-aarch64-32)
|
||||||
|
TARGET_ARCH = armv7l
|
||||||
|
endif
|
||||||
|
|
||||||
|
# operating system
|
||||||
|
HOST_OS := $(shell uname -s 2>/dev/null | tr "[:upper:]" "[:lower:]")
|
||||||
|
TARGET_OS ?= $(HOST_OS)
|
||||||
|
ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
|
||||||
|
$(error ERROR - unsupported value $(TARGET_OS) for TARGET_OS!)
|
||||||
|
endif
|
||||||
|
|
||||||
|
# host compiler
|
||||||
|
ifeq ($(TARGET_OS),darwin)
|
||||||
|
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
|
||||||
|
HOST_COMPILER ?= clang++
|
||||||
|
endif
|
||||||
|
else ifneq ($(TARGET_ARCH),$(HOST_ARCH))
|
||||||
|
ifeq ($(HOST_ARCH)-$(TARGET_ARCH),x86_64-armv7l)
|
||||||
|
ifeq ($(TARGET_OS),linux)
|
||||||
|
HOST_COMPILER ?= arm-linux-gnueabihf-g++
|
||||||
|
else ifeq ($(TARGET_OS),qnx)
|
||||||
|
ifeq ($(QNX_HOST),)
|
||||||
|
$(error ERROR - QNX_HOST must be passed to the QNX host toolchain)
|
||||||
|
endif
|
||||||
|
ifeq ($(QNX_TARGET),)
|
||||||
|
$(error ERROR - QNX_TARGET must be passed to the QNX target toolchain)
|
||||||
|
endif
|
||||||
|
export QNX_HOST
|
||||||
|
export QNX_TARGET
|
||||||
|
HOST_COMPILER ?= $(QNX_HOST)/usr/bin/arm-unknown-nto-qnx6.6.0eabi-g++
|
||||||
|
else ifeq ($(TARGET_OS),android)
|
||||||
|
HOST_COMPILER ?= arm-linux-androideabi-g++
|
||||||
|
endif
|
||||||
|
else ifeq ($(TARGET_ARCH),aarch64)
|
||||||
|
ifeq ($(TARGET_OS), linux)
|
||||||
|
HOST_COMPILER ?= aarch64-linux-gnu-g++
|
||||||
|
else ifeq ($(TARGET_OS),qnx)
|
||||||
|
ifeq ($(QNX_HOST),)
|
||||||
|
$(error ERROR - QNX_HOST must be passed to the QNX host toolchain)
|
||||||
|
endif
|
||||||
|
ifeq ($(QNX_TARGET),)
|
||||||
|
$(error ERROR - QNX_TARGET must be passed to the QNX target toolchain)
|
||||||
|
endif
|
||||||
|
export QNX_HOST
|
||||||
|
export QNX_TARGET
|
||||||
|
HOST_COMPILER ?= $(QNX_HOST)/usr/bin/q++
|
||||||
|
else ifeq ($(TARGET_OS), android)
|
||||||
|
HOST_COMPILER ?= aarch64-linux-android-clang++
|
||||||
|
endif
|
||||||
|
else ifeq ($(TARGET_ARCH),sbsa)
|
||||||
|
HOST_COMPILER ?= aarch64-linux-gnu-g++
|
||||||
|
else ifeq ($(TARGET_ARCH),ppc64le)
|
||||||
|
HOST_COMPILER ?= powerpc64le-linux-gnu-g++
|
||||||
|
endif
|
||||||
|
endif
|
||||||
|
HOST_COMPILER ?= g++
|
||||||
|
NVCC := $(CUDA_PATH)/bin/nvcc -ccbin $(HOST_COMPILER)
|
||||||
|
|
||||||
|
# internal flags
|
||||||
|
NVCCFLAGS := -m${TARGET_SIZE}
|
||||||
|
CCFLAGS :=
|
||||||
|
LDFLAGS :=
|
||||||
|
|
||||||
|
# build flags
|
||||||
|
ifeq ($(TARGET_OS),darwin)
|
||||||
|
LDFLAGS += -rpath $(CUDA_PATH)/lib
|
||||||
|
CCFLAGS += -arch $(HOST_ARCH)
|
||||||
|
else ifeq ($(HOST_ARCH)-$(TARGET_ARCH)-$(TARGET_OS),x86_64-armv7l-linux)
|
||||||
|
LDFLAGS += --dynamic-linker=/lib/ld-linux-armhf.so.3
|
||||||
|
CCFLAGS += -mfloat-abi=hard
|
||||||
|
else ifeq ($(TARGET_OS),android)
|
||||||
|
LDFLAGS += -pie
|
||||||
|
CCFLAGS += -fpie -fpic -fexceptions
|
||||||
|
endif
|
||||||
|
|
||||||
|
ifneq ($(TARGET_ARCH),$(HOST_ARCH))
|
||||||
|
ifeq ($(TARGET_ARCH)-$(TARGET_OS),armv7l-linux)
|
||||||
|
ifneq ($(TARGET_FS),)
|
||||||
|
GCCVERSIONLTEQ46 := $(shell expr `$(HOST_COMPILER) -dumpversion` \<= 4.6)
|
||||||
|
ifeq ($(GCCVERSIONLTEQ46),1)
|
||||||
|
CCFLAGS += --sysroot=$(TARGET_FS)
|
||||||
|
endif
|
||||||
|
LDFLAGS += --sysroot=$(TARGET_FS)
|
||||||
|
LDFLAGS += -rpath-link=$(TARGET_FS)/lib
|
||||||
|
LDFLAGS += -rpath-link=$(TARGET_FS)/usr/lib
|
||||||
|
LDFLAGS += -rpath-link=$(TARGET_FS)/usr/lib/arm-linux-gnueabihf
|
||||||
|
endif
|
||||||
|
endif
|
||||||
|
ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-linux)
|
||||||
|
ifneq ($(TARGET_FS),)
|
||||||
|
GCCVERSIONLTEQ46 := $(shell expr `$(HOST_COMPILER) -dumpversion` \<= 4.6)
|
||||||
|
ifeq ($(GCCVERSIONLTEQ46),1)
|
||||||
|
CCFLAGS += --sysroot=$(TARGET_FS)
|
||||||
|
endif
|
||||||
|
LDFLAGS += --sysroot=$(TARGET_FS)
|
||||||
|
LDFLAGS += -rpath-link=$(TARGET_FS)/lib -L$(TARGET_FS)/lib
|
||||||
|
LDFLAGS += -rpath-link=$(TARGET_FS)/lib/aarch64-linux-gnu -L$(TARGET_FS)/lib/aarch64-linux-gnu
|
||||||
|
LDFLAGS += -rpath-link=$(TARGET_FS)/usr/lib -L$(TARGET_FS)/usr/lib
|
||||||
|
LDFLAGS += -rpath-link=$(TARGET_FS)/usr/lib/aarch64-linux-gnu -L$(TARGET_FS)/usr/lib/aarch64-linux-gnu
|
||||||
|
LDFLAGS += --unresolved-symbols=ignore-in-shared-libs
|
||||||
|
CCFLAGS += -isystem=$(TARGET_FS)/usr/include -I$(TARGET_FS)/usr/include -I$(TARGET_FS)/usr/include/libdrm
|
||||||
|
CCFLAGS += -isystem=$(TARGET_FS)/usr/include/aarch64-linux-gnu -I$(TARGET_FS)/usr/include/aarch64-linux-gnu
|
||||||
|
endif
|
||||||
|
endif
|
||||||
|
ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-qnx)
|
||||||
|
NVCCFLAGS += --qpp-config 5.4.0,gcc_ntoaarch64le
|
||||||
|
CCFLAGS += -DWIN_INTERFACE_CUSTOM -I/usr/include/aarch64-qnx-gnu
|
||||||
|
LDFLAGS += -lsocket
|
||||||
|
LDFLAGS += -L/usr/lib/aarch64-qnx-gnu
|
||||||
|
CCFLAGS += "-Wl\,-rpath-link\,/usr/lib/aarch64-qnx-gnu"
|
||||||
|
ifdef TARGET_OVERRIDE
|
||||||
|
LDFLAGS += -lslog2
|
||||||
|
endif
|
||||||
|
|
||||||
|
ifneq ($(TARGET_FS),)
|
||||||
|
LDFLAGS += -L$(TARGET_FS)/usr/lib
|
||||||
|
CCFLAGS += "-Wl\,-rpath-link\,$(TARGET_FS)/usr/lib"
|
||||||
|
LDFLAGS += -L$(TARGET_FS)/usr/libnvidia
|
||||||
|
CCFLAGS += "-Wl\,-rpath-link\,$(TARGET_FS)/usr/libnvidia"
|
||||||
|
CCFLAGS += -I$(TARGET_FS)/../include
|
||||||
|
endif
|
||||||
|
endif
|
||||||
|
endif
|
||||||
|
|
||||||
|
ifdef TARGET_OVERRIDE # cuda toolkit targets override
|
||||||
|
NVCCFLAGS += -target-dir $(TARGET_OVERRIDE)
|
||||||
|
endif
|
||||||
|
|
||||||
|
# Install directory of different arch
|
||||||
|
CUDA_INSTALL_TARGET_DIR :=
|
||||||
|
ifeq ($(TARGET_ARCH)-$(TARGET_OS),armv7l-linux)
|
||||||
|
CUDA_INSTALL_TARGET_DIR = targets/armv7-linux-gnueabihf/
|
||||||
|
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-linux)
|
||||||
|
CUDA_INSTALL_TARGET_DIR = targets/aarch64-linux/
|
||||||
|
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),sbsa-linux)
|
||||||
|
CUDA_INSTALL_TARGET_DIR = targets/sbsa-linux/
|
||||||
|
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),armv7l-android)
|
||||||
|
CUDA_INSTALL_TARGET_DIR = targets/armv7-linux-androideabi/
|
||||||
|
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-android)
|
||||||
|
CUDA_INSTALL_TARGET_DIR = targets/aarch64-linux-androideabi/
|
||||||
|
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),armv7l-qnx)
|
||||||
|
CUDA_INSTALL_TARGET_DIR = targets/ARMv7-linux-QNX/
|
||||||
|
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-qnx)
|
||||||
|
CUDA_INSTALL_TARGET_DIR = targets/aarch64-qnx/
|
||||||
|
else ifeq ($(TARGET_ARCH),ppc64le)
|
||||||
|
CUDA_INSTALL_TARGET_DIR = targets/ppc64le-linux/
|
||||||
|
endif
|
||||||
|
|
||||||
|
# Debug build flags
|
||||||
|
ifeq ($(dbg),1)
|
||||||
|
NVCCFLAGS += -g -G
|
||||||
|
BUILD_TYPE := debug
|
||||||
|
else
|
||||||
|
BUILD_TYPE := release
|
||||||
|
endif
|
||||||
|
|
||||||
|
ALL_CCFLAGS :=
|
||||||
|
ALL_CCFLAGS += $(NVCCFLAGS)
|
||||||
|
ALL_CCFLAGS += $(EXTRA_NVCCFLAGS)
|
||||||
|
ALL_CCFLAGS += $(addprefix -Xcompiler ,$(CCFLAGS))
|
||||||
|
ALL_CCFLAGS += $(addprefix -Xcompiler ,$(EXTRA_CCFLAGS))
|
||||||
|
|
||||||
|
SAMPLE_ENABLED := 1
|
||||||
|
|
||||||
|
# This sample is not supported on QNX
|
||||||
|
ifeq ($(TARGET_OS),qnx)
|
||||||
|
$(info >>> WARNING - cdpQuadtree is not supported on QNX - waiving sample <<<)
|
||||||
|
SAMPLE_ENABLED := 0
|
||||||
|
endif
|
||||||
|
|
||||||
|
ALL_LDFLAGS :=
|
||||||
|
ALL_LDFLAGS += $(ALL_CCFLAGS)
|
||||||
|
ALL_LDFLAGS += $(addprefix -Xlinker ,$(LDFLAGS))
|
||||||
|
ALL_LDFLAGS += $(addprefix -Xlinker ,$(EXTRA_LDFLAGS))
|
||||||
|
|
||||||
|
# Common includes and paths for CUDA
|
||||||
|
INCLUDES := -I../../Common
|
||||||
|
LIBRARIES :=
|
||||||
|
|
||||||
|
################################################################################
|
||||||
|
|
||||||
|
#Detect if installed version of GCC supports required C++14
|
||||||
|
ifeq ($(TARGET_OS),linux)
|
||||||
|
empty :=
|
||||||
|
space := $(empty) $(empty)
|
||||||
|
GCCVERSIONSTRING := $(shell expr `$(HOST_COMPILER) -dumpversion`)
|
||||||
|
#Create version number without "."
|
||||||
|
GCCVERSION := $(shell expr `echo $(GCCVERSIONSTRING)` | cut -f1 -d.)
|
||||||
|
GCCVERSION += $(shell expr `echo $(GCCVERSIONSTRING)` | cut -f2 -d.)
|
||||||
|
GCCVERSION += $(shell expr `echo $(GCCVERSIONSTRING)` | cut -f3 -d.)
|
||||||
|
# Make sure the version number has at least 3 decimals
|
||||||
|
GCCVERSION += 00
|
||||||
|
# Remove spaces from the version number
|
||||||
|
GCCVERSION := $(subst $(space),$(empty),$(GCCVERSION))
|
||||||
|
#$(warning $(GCCVERSION))
|
||||||
|
|
||||||
|
IS_MIN_VERSION := $(shell expr `echo $(GCCVERSION)` \>= 50000)
|
||||||
|
|
||||||
|
ifeq ($(IS_MIN_VERSION), 1)
|
||||||
|
$(info >>> GCC Version is greater or equal to 5.0.0 <<<)
|
||||||
|
else
|
||||||
|
$(info >>> Waiving build. Minimum GCC version required is 5.0.0<<<)
|
||||||
|
SAMPLE_ENABLED := 0
|
||||||
|
endif
|
||||||
|
endif
|
||||||
|
|
||||||
|
# Gencode arguments
|
||||||
|
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64))
|
||||||
|
SMS ?= 35 37 50 52 60 61 70 72 75 80 86
|
||||||
|
else
|
||||||
|
SMS ?= 35 37 50 52 60 61 70 75 80 86
|
||||||
|
endif
|
||||||
|
|
||||||
|
ifeq ($(SMS),)
|
||||||
|
$(info >>> WARNING - no SM architectures have been specified - waiving sample <<<)
|
||||||
|
SAMPLE_ENABLED := 0
|
||||||
|
endif
|
||||||
|
|
||||||
|
ifeq ($(GENCODE_FLAGS),)
|
||||||
|
# Generate SASS code for each SM architecture listed in $(SMS)
|
||||||
|
$(foreach sm,$(SMS),$(eval GENCODE_FLAGS += -gencode arch=compute_$(sm),code=sm_$(sm)))
|
||||||
|
|
||||||
|
# Generate PTX code from the highest SM architecture in $(SMS) to guarantee forward-compatibility
|
||||||
|
HIGHEST_SM := $(lastword $(sort $(SMS)))
|
||||||
|
ifneq ($(HIGHEST_SM),)
|
||||||
|
GENCODE_FLAGS += -gencode arch=compute_$(HIGHEST_SM),code=compute_$(HIGHEST_SM)
|
||||||
|
endif
|
||||||
|
endif
|
||||||
|
|
||||||
|
ALL_CCFLAGS += -dc --std=c++14 --threads 0
|
||||||
|
|
||||||
|
LIBRARIES += -lcudadevrt
|
||||||
|
|
||||||
|
ifeq ($(SAMPLE_ENABLED),0)
|
||||||
|
EXEC ?= @echo "[@]"
|
||||||
|
endif
|
||||||
|
|
||||||
|
################################################################################
|
||||||
|
|
||||||
|
# Target rules
|
||||||
|
all: build
|
||||||
|
|
||||||
|
build: cdpQuadtree
|
||||||
|
|
||||||
|
check.deps:
|
||||||
|
ifeq ($(SAMPLE_ENABLED),0)
|
||||||
|
@echo "Sample will be waived due to the above missing dependencies"
|
||||||
|
else
|
||||||
|
@echo "Sample is ready - all dependencies have been met"
|
||||||
|
endif
|
||||||
|
|
||||||
|
cdpQuadtree.o:cdpQuadtree.cu
|
||||||
|
$(EXEC) $(NVCC) $(INCLUDES) $(ALL_CCFLAGS) $(GENCODE_FLAGS) -o $@ -c $<
|
||||||
|
|
||||||
|
cdpQuadtree: cdpQuadtree.o
|
||||||
|
$(EXEC) $(NVCC) $(ALL_LDFLAGS) $(GENCODE_FLAGS) -o $@ $+ $(LIBRARIES)
|
||||||
|
$(EXEC) mkdir -p ../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)
|
||||||
|
$(EXEC) cp $@ ../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)
|
||||||
|
|
||||||
|
run: build
|
||||||
|
$(EXEC) ./cdpQuadtree
|
||||||
|
|
||||||
|
clean:
|
||||||
|
rm -f cdpQuadtree cdpQuadtree.o
|
||||||
|
rm -rf ../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)/cdpQuadtree
|
||||||
|
|
||||||
|
clobber: clean
|
||||||
72
Samples/cdpQuadtree/NsightEclipse.xml
Normal file
72
Samples/cdpQuadtree/NsightEclipse.xml
Normal file
@@ -0,0 +1,72 @@
|
|||||||
|
<?xml version="1.0" encoding="UTF-8"?>
|
||||||
|
<!DOCTYPE entry SYSTEM "SamplesInfo.dtd">
|
||||||
|
<entry>
|
||||||
|
<name>cdpQuadtree</name>
|
||||||
|
<cflags>
|
||||||
|
<flag>-dc</flag>
|
||||||
|
<flag>--std=c++14</flag>
|
||||||
|
</cflags>
|
||||||
|
<description><![CDATA[This sample demonstrates Quad Trees implemented using CUDA Dynamic Parallelism. This sample requires devices with compute capability 3.5 or higher.]]></description>
|
||||||
|
<includepaths>
|
||||||
|
<path>./</path>
|
||||||
|
<path>../</path>
|
||||||
|
<path>../../common/inc</path>
|
||||||
|
</includepaths>
|
||||||
|
<keyconcepts>
|
||||||
|
<concept level="basic">Cooperative Groups</concept>
|
||||||
|
<concept level="advanced">CUDA Dynamic Parallelism</concept>
|
||||||
|
</keyconcepts>
|
||||||
|
<keywords>
|
||||||
|
<keyword>GPGPU</keyword>
|
||||||
|
<keyword>CPP14</keyword>
|
||||||
|
</keywords>
|
||||||
|
<libraries>
|
||||||
|
<library>cudadevrt</library>
|
||||||
|
</libraries>
|
||||||
|
<librarypaths>
|
||||||
|
</librarypaths>
|
||||||
|
<nsight_eclipse>true</nsight_eclipse>
|
||||||
|
<primary_file>cdpQuadTree.cu</primary_file>
|
||||||
|
<required_dependencies>
|
||||||
|
<dependency>CDP</dependency>
|
||||||
|
</required_dependencies>
|
||||||
|
<scopes>
|
||||||
|
<scope>1:CUDA Advanced Topics</scope>
|
||||||
|
</scopes>
|
||||||
|
<sm-arch>sm35</sm-arch>
|
||||||
|
<sm-arch>sm37</sm-arch>
|
||||||
|
<sm-arch>sm50</sm-arch>
|
||||||
|
<sm-arch>sm52</sm-arch>
|
||||||
|
<sm-arch>sm60</sm-arch>
|
||||||
|
<sm-arch>sm61</sm-arch>
|
||||||
|
<sm-arch>sm70</sm-arch>
|
||||||
|
<sm-arch>sm72</sm-arch>
|
||||||
|
<sm-arch>sm75</sm-arch>
|
||||||
|
<sm-arch>sm80</sm-arch>
|
||||||
|
<sm-arch>sm86</sm-arch>
|
||||||
|
<supported_envs>
|
||||||
|
<env>
|
||||||
|
<arch>x86_64</arch>
|
||||||
|
<platform>linux</platform>
|
||||||
|
</env>
|
||||||
|
<env>
|
||||||
|
<platform>windows7</platform>
|
||||||
|
</env>
|
||||||
|
<env>
|
||||||
|
<arch>x86_64</arch>
|
||||||
|
<platform>macosx</platform>
|
||||||
|
</env>
|
||||||
|
<env>
|
||||||
|
<arch>arm</arch>
|
||||||
|
</env>
|
||||||
|
<env>
|
||||||
|
<arch>ppc64le</arch>
|
||||||
|
<platform>linux</platform>
|
||||||
|
</env>
|
||||||
|
</supported_envs>
|
||||||
|
<supported_sm_architectures>
|
||||||
|
<from>3.5</from>
|
||||||
|
</supported_sm_architectures>
|
||||||
|
<title>Quad Tree (CUDA Dynamic Parallelism)</title>
|
||||||
|
<type>exe</type>
|
||||||
|
</entry>
|
||||||
71
Samples/cdpQuadtree/README.md
Normal file
71
Samples/cdpQuadtree/README.md
Normal file
@@ -0,0 +1,71 @@
|
|||||||
|
# cdpQuadtree - Quad Tree (CUDA Dynamic Parallelism)
|
||||||
|
|
||||||
|
## Description
|
||||||
|
|
||||||
|
This sample demonstrates Quad Trees implemented using CUDA Dynamic Parallelism. This sample requires devices with compute capability 3.5 or higher.
|
||||||
|
|
||||||
|
## Key Concepts
|
||||||
|
|
||||||
|
Cooperative Groups, CUDA Dynamic Parallelism
|
||||||
|
|
||||||
|
## Supported SM Architectures
|
||||||
|
|
||||||
|
[SM 3.5 ](https://developer.nvidia.com/cuda-gpus) [SM 3.7 ](https://developer.nvidia.com/cuda-gpus) [SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](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)
|
||||||
|
|
||||||
|
## Supported OSes
|
||||||
|
|
||||||
|
Linux, Windows
|
||||||
|
|
||||||
|
## Supported CPU Architecture
|
||||||
|
|
||||||
|
x86_64, ppc64le, armv7l
|
||||||
|
|
||||||
|
## CUDA APIs involved
|
||||||
|
|
||||||
|
## Dependencies needed to build/run
|
||||||
|
[CDP](../../README.md#cdp)
|
||||||
|
|
||||||
|
## Prerequisites
|
||||||
|
|
||||||
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
|
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
||||||
|
|
||||||
|
## Build and Run
|
||||||
|
|
||||||
|
### Windows
|
||||||
|
The Windows samples are built using the Visual Studio IDE. Solution files (.sln) are provided for each supported version of Visual Studio, using the format:
|
||||||
|
```
|
||||||
|
*_vs<version>.sln - for Visual Studio <version>
|
||||||
|
```
|
||||||
|
Each individual sample has its own set of solution files in its directory:
|
||||||
|
|
||||||
|
To build/examine all the samples at once, the complete solution files should be used. To build/examine a single sample, the individual sample solution files should be used.
|
||||||
|
> **Note:** Some samples require that the Microsoft DirectX SDK (June 2010 or newer) be installed and that the VC++ directory paths are properly set up (**Tools > Options...**). Check DirectX Dependencies section for details."
|
||||||
|
|
||||||
|
### Linux
|
||||||
|
The Linux samples are built using makefiles. To use the makefiles, change the current directory to the sample directory you wish to build, and run make:
|
||||||
|
```
|
||||||
|
$ cd <sample_dir>
|
||||||
|
$ make
|
||||||
|
```
|
||||||
|
The samples makefiles can take advantage of certain options:
|
||||||
|
* **TARGET_ARCH=<arch>** - cross-compile targeting a specific architecture. Allowed architectures are x86_64, ppc64le, armv7l.
|
||||||
|
By default, TARGET_ARCH is set to HOST_ARCH. On a x86_64 machine, not setting TARGET_ARCH is the equivalent of setting TARGET_ARCH=x86_64.<br/>
|
||||||
|
`$ make TARGET_ARCH=x86_64` <br/> `$ make TARGET_ARCH=ppc64le` <br/> `$ make TARGET_ARCH=armv7l` <br/>
|
||||||
|
See [here](http://docs.nvidia.com/cuda/cuda-samples/index.html#cross-samples) for more details.
|
||||||
|
* **dbg=1** - build with debug symbols
|
||||||
|
```
|
||||||
|
$ make dbg=1
|
||||||
|
```
|
||||||
|
* **SMS="A B ..."** - override the SM architectures for which the sample will be built, where `"A B ..."` is a space-delimited list of SM architectures. For example, to generate SASS for SM 50 and SM 60, use `SMS="50 60"`.
|
||||||
|
```
|
||||||
|
$ make SMS="50 60"
|
||||||
|
```
|
||||||
|
|
||||||
|
* **HOST_COMPILER=<host_compiler>** - override the default g++ host compiler. See the [Linux Installation Guide](http://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#system-requirements) for a list of supported host compilers.
|
||||||
|
```
|
||||||
|
$ make HOST_COMPILER=g++
|
||||||
|
```
|
||||||
|
|
||||||
|
## References (for more details)
|
||||||
|
|
||||||
742
Samples/cdpQuadtree/cdpQuadtree.cu
Normal file
742
Samples/cdpQuadtree/cdpQuadtree.cu
Normal file
@@ -0,0 +1,742 @@
|
|||||||
|
/* Copyright (c) 2021, 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.
|
||||||
|
*/
|
||||||
|
|
||||||
|
#include <thrust/random.h>
|
||||||
|
#include <thrust/device_vector.h>
|
||||||
|
#include <thrust/host_vector.h>
|
||||||
|
#include <cooperative_groups.h>
|
||||||
|
|
||||||
|
namespace cg = cooperative_groups;
|
||||||
|
#include <helper_cuda.h>
|
||||||
|
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
// A structure of 2D points (structure of arrays).
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
class Points {
|
||||||
|
float *m_x;
|
||||||
|
float *m_y;
|
||||||
|
|
||||||
|
public:
|
||||||
|
// Constructor.
|
||||||
|
__host__ __device__ Points() : m_x(NULL), m_y(NULL) {}
|
||||||
|
|
||||||
|
// Constructor.
|
||||||
|
__host__ __device__ Points(float *x, float *y) : m_x(x), m_y(y) {}
|
||||||
|
|
||||||
|
// Get a point.
|
||||||
|
__host__ __device__ __forceinline__ float2 get_point(int idx) const {
|
||||||
|
return make_float2(m_x[idx], m_y[idx]);
|
||||||
|
}
|
||||||
|
|
||||||
|
// Set a point.
|
||||||
|
__host__ __device__ __forceinline__ void set_point(int idx, const float2 &p) {
|
||||||
|
m_x[idx] = p.x;
|
||||||
|
m_y[idx] = p.y;
|
||||||
|
}
|
||||||
|
|
||||||
|
// Set the pointers.
|
||||||
|
__host__ __device__ __forceinline__ void set(float *x, float *y) {
|
||||||
|
m_x = x;
|
||||||
|
m_y = y;
|
||||||
|
}
|
||||||
|
};
|
||||||
|
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
// A 2D bounding box
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
class Bounding_box {
|
||||||
|
// Extreme points of the bounding box.
|
||||||
|
float2 m_p_min;
|
||||||
|
float2 m_p_max;
|
||||||
|
|
||||||
|
public:
|
||||||
|
// Constructor. Create a unit box.
|
||||||
|
__host__ __device__ Bounding_box() {
|
||||||
|
m_p_min = make_float2(0.0f, 0.0f);
|
||||||
|
m_p_max = make_float2(1.0f, 1.0f);
|
||||||
|
}
|
||||||
|
|
||||||
|
// Compute the center of the bounding-box.
|
||||||
|
__host__ __device__ void compute_center(float2 ¢er) const {
|
||||||
|
center.x = 0.5f * (m_p_min.x + m_p_max.x);
|
||||||
|
center.y = 0.5f * (m_p_min.y + m_p_max.y);
|
||||||
|
}
|
||||||
|
|
||||||
|
// The points of the box.
|
||||||
|
__host__ __device__ __forceinline__ const float2 &get_max() const {
|
||||||
|
return m_p_max;
|
||||||
|
}
|
||||||
|
|
||||||
|
__host__ __device__ __forceinline__ const float2 &get_min() const {
|
||||||
|
return m_p_min;
|
||||||
|
}
|
||||||
|
|
||||||
|
// Does a box contain a point.
|
||||||
|
__host__ __device__ bool contains(const float2 &p) const {
|
||||||
|
return p.x >= m_p_min.x && p.x < m_p_max.x && p.y >= m_p_min.y &&
|
||||||
|
p.y < m_p_max.y;
|
||||||
|
}
|
||||||
|
|
||||||
|
// Define the bounding box.
|
||||||
|
__host__ __device__ void set(float min_x, float min_y, float max_x,
|
||||||
|
float max_y) {
|
||||||
|
m_p_min.x = min_x;
|
||||||
|
m_p_min.y = min_y;
|
||||||
|
m_p_max.x = max_x;
|
||||||
|
m_p_max.y = max_y;
|
||||||
|
}
|
||||||
|
};
|
||||||
|
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
// A node of a quadree.
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
class Quadtree_node {
|
||||||
|
// The identifier of the node.
|
||||||
|
int m_id;
|
||||||
|
// The bounding box of the tree.
|
||||||
|
Bounding_box m_bounding_box;
|
||||||
|
// The range of points.
|
||||||
|
int m_begin, m_end;
|
||||||
|
|
||||||
|
public:
|
||||||
|
// Constructor.
|
||||||
|
__host__ __device__ Quadtree_node() : m_id(0), m_begin(0), m_end(0) {}
|
||||||
|
|
||||||
|
// The ID of a node at its level.
|
||||||
|
__host__ __device__ int id() const { return m_id; }
|
||||||
|
|
||||||
|
// The ID of a node at its level.
|
||||||
|
__host__ __device__ void set_id(int new_id) { m_id = new_id; }
|
||||||
|
|
||||||
|
// The bounding box.
|
||||||
|
__host__ __device__ __forceinline__ const Bounding_box &bounding_box() const {
|
||||||
|
return m_bounding_box;
|
||||||
|
}
|
||||||
|
|
||||||
|
// Set the bounding box.
|
||||||
|
__host__ __device__ __forceinline__ void set_bounding_box(float min_x,
|
||||||
|
float min_y,
|
||||||
|
float max_x,
|
||||||
|
float max_y) {
|
||||||
|
m_bounding_box.set(min_x, min_y, max_x, max_y);
|
||||||
|
}
|
||||||
|
|
||||||
|
// The number of points in the tree.
|
||||||
|
__host__ __device__ __forceinline__ int num_points() const {
|
||||||
|
return m_end - m_begin;
|
||||||
|
}
|
||||||
|
|
||||||
|
// The range of points in the tree.
|
||||||
|
__host__ __device__ __forceinline__ int points_begin() const {
|
||||||
|
return m_begin;
|
||||||
|
}
|
||||||
|
|
||||||
|
__host__ __device__ __forceinline__ int points_end() const { return m_end; }
|
||||||
|
|
||||||
|
// Define the range for that node.
|
||||||
|
__host__ __device__ __forceinline__ void set_range(int begin, int end) {
|
||||||
|
m_begin = begin;
|
||||||
|
m_end = end;
|
||||||
|
}
|
||||||
|
};
|
||||||
|
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
// Algorithm parameters.
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
struct Parameters {
|
||||||
|
// Choose the right set of points to use as in/out.
|
||||||
|
int point_selector;
|
||||||
|
// The number of nodes at a given level (2^k for level k).
|
||||||
|
int num_nodes_at_this_level;
|
||||||
|
// The recursion depth.
|
||||||
|
int depth;
|
||||||
|
// The max value for depth.
|
||||||
|
const int max_depth;
|
||||||
|
// The minimum number of points in a node to stop recursion.
|
||||||
|
const int min_points_per_node;
|
||||||
|
|
||||||
|
// Constructor set to default values.
|
||||||
|
__host__ __device__ Parameters(int max_depth, int min_points_per_node)
|
||||||
|
: point_selector(0),
|
||||||
|
num_nodes_at_this_level(1),
|
||||||
|
depth(0),
|
||||||
|
max_depth(max_depth),
|
||||||
|
min_points_per_node(min_points_per_node) {}
|
||||||
|
|
||||||
|
// Copy constructor. Changes the values for next iteration.
|
||||||
|
__host__ __device__ Parameters(const Parameters ¶ms, bool)
|
||||||
|
: point_selector((params.point_selector + 1) % 2),
|
||||||
|
num_nodes_at_this_level(4 * params.num_nodes_at_this_level),
|
||||||
|
depth(params.depth + 1),
|
||||||
|
max_depth(params.max_depth),
|
||||||
|
min_points_per_node(params.min_points_per_node) {}
|
||||||
|
};
|
||||||
|
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
// Build a quadtree on the GPU. Use CUDA Dynamic Parallelism.
|
||||||
|
//
|
||||||
|
// The algorithm works as follows. The host (CPU) launches one block of
|
||||||
|
// NUM_THREADS_PER_BLOCK threads. That block will do the following steps:
|
||||||
|
//
|
||||||
|
// 1- Check the number of points and its depth.
|
||||||
|
//
|
||||||
|
// We impose a maximum depth to the tree and a minimum number of points per
|
||||||
|
// node. If the maximum depth is exceeded or the minimum number of points is
|
||||||
|
// reached. The threads in the block exit.
|
||||||
|
//
|
||||||
|
// Before exiting, they perform a buffer swap if it is needed. Indeed, the
|
||||||
|
// algorithm uses two buffers to permute the points and make sure they are
|
||||||
|
// properly distributed in the quadtree. By design we want all points to be
|
||||||
|
// in the first buffer of points at the end of the algorithm. It is the reason
|
||||||
|
// why we may have to swap the buffer before leavin (if the points are in the
|
||||||
|
// 2nd buffer).
|
||||||
|
//
|
||||||
|
// 2- Count the number of points in each child.
|
||||||
|
//
|
||||||
|
// If the depth is not too high and the number of points is sufficient, the
|
||||||
|
// block has to dispatch the points into four geometrical buckets: Its
|
||||||
|
// children. For that purpose, we compute the center of the bounding box and
|
||||||
|
// count the number of points in each quadrant.
|
||||||
|
//
|
||||||
|
// The set of points is divided into sections. Each section is given to a
|
||||||
|
// warp of threads (32 threads). Warps use __ballot and __popc intrinsics
|
||||||
|
// to count the points. See the Programming Guide for more information about
|
||||||
|
// those functions.
|
||||||
|
//
|
||||||
|
// 3- Scan the warps' results to know the "global" numbers.
|
||||||
|
//
|
||||||
|
// Warps work independently from each other. At the end, each warp knows the
|
||||||
|
// number of points in its section. To know the numbers for the block, the
|
||||||
|
// block has to run a scan/reduce at the block level. It's a traditional
|
||||||
|
// approach. The implementation in that sample is not as optimized as what
|
||||||
|
// could be found in fast radix sorts, for example, but it relies on the same
|
||||||
|
// idea.
|
||||||
|
//
|
||||||
|
// 4- Move points.
|
||||||
|
//
|
||||||
|
// Now that the block knows how many points go in each of its 4 children, it
|
||||||
|
// remains to dispatch the points. It is straightforward.
|
||||||
|
//
|
||||||
|
// 5- Launch new blocks.
|
||||||
|
//
|
||||||
|
// The block launches four new blocks: One per children. Each of the four blocks
|
||||||
|
// will apply the same algorithm.
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
template <int NUM_THREADS_PER_BLOCK>
|
||||||
|
__global__ void build_quadtree_kernel(Quadtree_node *nodes, Points *points,
|
||||||
|
Parameters params) {
|
||||||
|
// Handle to thread block group
|
||||||
|
cg::thread_block cta = cg::this_thread_block();
|
||||||
|
// The number of warps in a block.
|
||||||
|
const int NUM_WARPS_PER_BLOCK = NUM_THREADS_PER_BLOCK / warpSize;
|
||||||
|
|
||||||
|
// Shared memory to store the number of points.
|
||||||
|
extern __shared__ int smem[];
|
||||||
|
|
||||||
|
// s_num_pts[4][NUM_WARPS_PER_BLOCK];
|
||||||
|
// Addresses of shared memory.
|
||||||
|
volatile int *s_num_pts[4];
|
||||||
|
|
||||||
|
for (int i = 0; i < 4; ++i)
|
||||||
|
s_num_pts[i] = (volatile int *)&smem[i * NUM_WARPS_PER_BLOCK];
|
||||||
|
|
||||||
|
// Compute the coordinates of the threads in the block.
|
||||||
|
const int warp_id = threadIdx.x / warpSize;
|
||||||
|
const int lane_id = threadIdx.x % warpSize;
|
||||||
|
|
||||||
|
// Mask for compaction.
|
||||||
|
// Same as: asm( "mov.u32 %0, %%lanemask_lt;" : "=r"(lane_mask_lt) );
|
||||||
|
int lane_mask_lt = (1 << lane_id) - 1;
|
||||||
|
|
||||||
|
// The current node.
|
||||||
|
Quadtree_node &node = nodes[blockIdx.x];
|
||||||
|
|
||||||
|
// The number of points in the node.
|
||||||
|
int num_points = node.num_points();
|
||||||
|
|
||||||
|
float2 center;
|
||||||
|
int range_begin, range_end;
|
||||||
|
int warp_cnts[4] = {0, 0, 0, 0};
|
||||||
|
//
|
||||||
|
// 1- Check the number of points and its depth.
|
||||||
|
//
|
||||||
|
|
||||||
|
// Stop the recursion here. Make sure points[0] contains all the points.
|
||||||
|
if (params.depth >= params.max_depth ||
|
||||||
|
num_points <= params.min_points_per_node) {
|
||||||
|
if (params.point_selector == 1) {
|
||||||
|
int it = node.points_begin(), end = node.points_end();
|
||||||
|
|
||||||
|
for (it += threadIdx.x; it < end; it += NUM_THREADS_PER_BLOCK)
|
||||||
|
if (it < end) points[0].set_point(it, points[1].get_point(it));
|
||||||
|
}
|
||||||
|
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
|
// Compute the center of the bounding box of the points.
|
||||||
|
const Bounding_box &bbox = node.bounding_box();
|
||||||
|
|
||||||
|
bbox.compute_center(center);
|
||||||
|
|
||||||
|
// Find how many points to give to each warp.
|
||||||
|
int num_points_per_warp = max(
|
||||||
|
warpSize, (num_points + NUM_WARPS_PER_BLOCK - 1) / NUM_WARPS_PER_BLOCK);
|
||||||
|
|
||||||
|
// Each warp of threads will compute the number of points to move to each
|
||||||
|
// quadrant.
|
||||||
|
range_begin = node.points_begin() + warp_id * num_points_per_warp;
|
||||||
|
range_end = min(range_begin + num_points_per_warp, node.points_end());
|
||||||
|
|
||||||
|
//
|
||||||
|
// 2- Count the number of points in each child.
|
||||||
|
//
|
||||||
|
|
||||||
|
// Input points.
|
||||||
|
const Points &in_points = points[params.point_selector];
|
||||||
|
|
||||||
|
cg::thread_block_tile<32> tile32 = cg::tiled_partition<32>(cta);
|
||||||
|
// Compute the number of points.
|
||||||
|
for (int range_it = range_begin + tile32.thread_rank();
|
||||||
|
tile32.any(range_it < range_end); range_it += warpSize) {
|
||||||
|
// Is it still an active thread?
|
||||||
|
bool is_active = range_it < range_end;
|
||||||
|
|
||||||
|
// Load the coordinates of the point.
|
||||||
|
float2 p =
|
||||||
|
is_active ? in_points.get_point(range_it) : make_float2(0.0f, 0.0f);
|
||||||
|
|
||||||
|
// Count top-left points.
|
||||||
|
int num_pts =
|
||||||
|
__popc(tile32.ballot(is_active && p.x < center.x && p.y >= center.y));
|
||||||
|
warp_cnts[0] += tile32.shfl(num_pts, 0);
|
||||||
|
|
||||||
|
// Count top-right points.
|
||||||
|
num_pts =
|
||||||
|
__popc(tile32.ballot(is_active && p.x >= center.x && p.y >= center.y));
|
||||||
|
warp_cnts[1] += tile32.shfl(num_pts, 0);
|
||||||
|
|
||||||
|
// Count bottom-left points.
|
||||||
|
num_pts =
|
||||||
|
__popc(tile32.ballot(is_active && p.x < center.x && p.y < center.y));
|
||||||
|
warp_cnts[2] += tile32.shfl(num_pts, 0);
|
||||||
|
|
||||||
|
// Count bottom-right points.
|
||||||
|
num_pts =
|
||||||
|
__popc(tile32.ballot(is_active && p.x >= center.x && p.y < center.y));
|
||||||
|
warp_cnts[3] += tile32.shfl(num_pts, 0);
|
||||||
|
}
|
||||||
|
|
||||||
|
if (tile32.thread_rank() == 0) {
|
||||||
|
s_num_pts[0][warp_id] = warp_cnts[0];
|
||||||
|
s_num_pts[1][warp_id] = warp_cnts[1];
|
||||||
|
s_num_pts[2][warp_id] = warp_cnts[2];
|
||||||
|
s_num_pts[3][warp_id] = warp_cnts[3];
|
||||||
|
}
|
||||||
|
|
||||||
|
// Make sure warps have finished counting.
|
||||||
|
cg::sync(cta);
|
||||||
|
|
||||||
|
//
|
||||||
|
// 3- Scan the warps' results to know the "global" numbers.
|
||||||
|
//
|
||||||
|
|
||||||
|
// First 4 warps scan the numbers of points per child (inclusive scan).
|
||||||
|
if (warp_id < 4) {
|
||||||
|
int num_pts = tile32.thread_rank() < NUM_WARPS_PER_BLOCK
|
||||||
|
? s_num_pts[warp_id][tile32.thread_rank()]
|
||||||
|
: 0;
|
||||||
|
#pragma unroll
|
||||||
|
|
||||||
|
for (int offset = 1; offset < NUM_WARPS_PER_BLOCK; offset *= 2) {
|
||||||
|
int n = tile32.shfl_up(num_pts, offset);
|
||||||
|
|
||||||
|
if (tile32.thread_rank() >= offset) num_pts += n;
|
||||||
|
}
|
||||||
|
|
||||||
|
if (tile32.thread_rank() < NUM_WARPS_PER_BLOCK)
|
||||||
|
s_num_pts[warp_id][tile32.thread_rank()] = num_pts;
|
||||||
|
}
|
||||||
|
|
||||||
|
cg::sync(cta);
|
||||||
|
|
||||||
|
// Compute global offsets.
|
||||||
|
if (warp_id == 0) {
|
||||||
|
int sum = s_num_pts[0][NUM_WARPS_PER_BLOCK - 1];
|
||||||
|
|
||||||
|
for (int row = 1; row < 4; ++row) {
|
||||||
|
int tmp = s_num_pts[row][NUM_WARPS_PER_BLOCK - 1];
|
||||||
|
cg::sync(tile32);
|
||||||
|
|
||||||
|
if (tile32.thread_rank() < NUM_WARPS_PER_BLOCK)
|
||||||
|
s_num_pts[row][tile32.thread_rank()] += sum;
|
||||||
|
|
||||||
|
cg::sync(tile32);
|
||||||
|
sum += tmp;
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
cg::sync(cta);
|
||||||
|
|
||||||
|
// Make the scan exclusive.
|
||||||
|
int val = 0;
|
||||||
|
if (threadIdx.x < 4 * NUM_WARPS_PER_BLOCK) {
|
||||||
|
val = threadIdx.x == 0 ? 0 : smem[threadIdx.x - 1];
|
||||||
|
val += node.points_begin();
|
||||||
|
}
|
||||||
|
|
||||||
|
cg::sync(cta);
|
||||||
|
|
||||||
|
if (threadIdx.x < 4 * NUM_WARPS_PER_BLOCK) {
|
||||||
|
smem[threadIdx.x] = val;
|
||||||
|
}
|
||||||
|
|
||||||
|
cg::sync(cta);
|
||||||
|
|
||||||
|
//
|
||||||
|
// 4- Move points.
|
||||||
|
//
|
||||||
|
if (!(params.depth >= params.max_depth ||
|
||||||
|
num_points <= params.min_points_per_node)) {
|
||||||
|
// Output points.
|
||||||
|
Points &out_points = points[(params.point_selector + 1) % 2];
|
||||||
|
|
||||||
|
warp_cnts[0] = s_num_pts[0][warp_id];
|
||||||
|
warp_cnts[1] = s_num_pts[1][warp_id];
|
||||||
|
warp_cnts[2] = s_num_pts[2][warp_id];
|
||||||
|
warp_cnts[3] = s_num_pts[3][warp_id];
|
||||||
|
|
||||||
|
const Points &in_points = points[params.point_selector];
|
||||||
|
// Reorder points.
|
||||||
|
for (int range_it = range_begin + tile32.thread_rank();
|
||||||
|
tile32.any(range_it < range_end); range_it += warpSize) {
|
||||||
|
// Is it still an active thread?
|
||||||
|
bool is_active = range_it < range_end;
|
||||||
|
|
||||||
|
// Load the coordinates of the point.
|
||||||
|
float2 p =
|
||||||
|
is_active ? in_points.get_point(range_it) : make_float2(0.0f, 0.0f);
|
||||||
|
|
||||||
|
// Count top-left points.
|
||||||
|
bool pred = is_active && p.x < center.x && p.y >= center.y;
|
||||||
|
int vote = tile32.ballot(pred);
|
||||||
|
int dest = warp_cnts[0] + __popc(vote & lane_mask_lt);
|
||||||
|
|
||||||
|
if (pred) out_points.set_point(dest, p);
|
||||||
|
|
||||||
|
warp_cnts[0] += tile32.shfl(__popc(vote), 0);
|
||||||
|
|
||||||
|
// Count top-right points.
|
||||||
|
pred = is_active && p.x >= center.x && p.y >= center.y;
|
||||||
|
vote = tile32.ballot(pred);
|
||||||
|
dest = warp_cnts[1] + __popc(vote & lane_mask_lt);
|
||||||
|
|
||||||
|
if (pred) out_points.set_point(dest, p);
|
||||||
|
|
||||||
|
warp_cnts[1] += tile32.shfl(__popc(vote), 0);
|
||||||
|
|
||||||
|
// Count bottom-left points.
|
||||||
|
pred = is_active && p.x < center.x && p.y < center.y;
|
||||||
|
vote = tile32.ballot(pred);
|
||||||
|
dest = warp_cnts[2] + __popc(vote & lane_mask_lt);
|
||||||
|
|
||||||
|
if (pred) out_points.set_point(dest, p);
|
||||||
|
|
||||||
|
warp_cnts[2] += tile32.shfl(__popc(vote), 0);
|
||||||
|
|
||||||
|
// Count bottom-right points.
|
||||||
|
pred = is_active && p.x >= center.x && p.y < center.y;
|
||||||
|
vote = tile32.ballot(pred);
|
||||||
|
dest = warp_cnts[3] + __popc(vote & lane_mask_lt);
|
||||||
|
|
||||||
|
if (pred) out_points.set_point(dest, p);
|
||||||
|
|
||||||
|
warp_cnts[3] += tile32.shfl(__popc(vote), 0);
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
cg::sync(cta);
|
||||||
|
|
||||||
|
if (tile32.thread_rank() == 0) {
|
||||||
|
s_num_pts[0][warp_id] = warp_cnts[0];
|
||||||
|
s_num_pts[1][warp_id] = warp_cnts[1];
|
||||||
|
s_num_pts[2][warp_id] = warp_cnts[2];
|
||||||
|
s_num_pts[3][warp_id] = warp_cnts[3];
|
||||||
|
}
|
||||||
|
|
||||||
|
cg::sync(cta);
|
||||||
|
|
||||||
|
//
|
||||||
|
// 5- Launch new blocks.
|
||||||
|
//
|
||||||
|
if (!(params.depth >= params.max_depth ||
|
||||||
|
num_points <= params.min_points_per_node)) {
|
||||||
|
// The last thread launches new blocks.
|
||||||
|
if (threadIdx.x == NUM_THREADS_PER_BLOCK - 1) {
|
||||||
|
// The children.
|
||||||
|
Quadtree_node *children =
|
||||||
|
&nodes[params.num_nodes_at_this_level - (node.id() & ~3)];
|
||||||
|
|
||||||
|
// The offsets of the children at their level.
|
||||||
|
int child_offset = 4 * node.id();
|
||||||
|
|
||||||
|
// Set IDs.
|
||||||
|
children[child_offset + 0].set_id(4 * node.id() + 0);
|
||||||
|
children[child_offset + 1].set_id(4 * node.id() + 1);
|
||||||
|
children[child_offset + 2].set_id(4 * node.id() + 2);
|
||||||
|
children[child_offset + 3].set_id(4 * node.id() + 3);
|
||||||
|
|
||||||
|
const Bounding_box &bbox = node.bounding_box();
|
||||||
|
// Points of the bounding-box.
|
||||||
|
const float2 &p_min = bbox.get_min();
|
||||||
|
const float2 &p_max = bbox.get_max();
|
||||||
|
|
||||||
|
// Set the bounding boxes of the children.
|
||||||
|
children[child_offset + 0].set_bounding_box(p_min.x, center.y, center.x,
|
||||||
|
p_max.y); // Top-left.
|
||||||
|
children[child_offset + 1].set_bounding_box(center.x, center.y, p_max.x,
|
||||||
|
p_max.y); // Top-right.
|
||||||
|
children[child_offset + 2].set_bounding_box(p_min.x, p_min.y, center.x,
|
||||||
|
center.y); // Bottom-left.
|
||||||
|
children[child_offset + 3].set_bounding_box(center.x, p_min.y, p_max.x,
|
||||||
|
center.y); // Bottom-right.
|
||||||
|
|
||||||
|
// Set the ranges of the children.
|
||||||
|
|
||||||
|
children[child_offset + 0].set_range(node.points_begin(),
|
||||||
|
s_num_pts[0][warp_id]);
|
||||||
|
children[child_offset + 1].set_range(s_num_pts[0][warp_id],
|
||||||
|
s_num_pts[1][warp_id]);
|
||||||
|
children[child_offset + 2].set_range(s_num_pts[1][warp_id],
|
||||||
|
s_num_pts[2][warp_id]);
|
||||||
|
children[child_offset + 3].set_range(s_num_pts[2][warp_id],
|
||||||
|
s_num_pts[3][warp_id]);
|
||||||
|
|
||||||
|
// Launch 4 children.
|
||||||
|
build_quadtree_kernel<NUM_THREADS_PER_BLOCK><<<
|
||||||
|
4, NUM_THREADS_PER_BLOCK, 4 * NUM_WARPS_PER_BLOCK * sizeof(int)>>>(
|
||||||
|
&children[child_offset], points, Parameters(params, true));
|
||||||
|
}
|
||||||
|
}
|
||||||
|
}
|
||||||
|
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
// Make sure a Quadtree is properly defined.
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
bool check_quadtree(const Quadtree_node *nodes, int idx, int num_pts,
|
||||||
|
Points *pts, Parameters params) {
|
||||||
|
const Quadtree_node &node = nodes[idx];
|
||||||
|
int num_points = node.num_points();
|
||||||
|
|
||||||
|
if (!(params.depth == params.max_depth ||
|
||||||
|
num_points <= params.min_points_per_node)) {
|
||||||
|
int num_points_in_children = 0;
|
||||||
|
|
||||||
|
num_points_in_children +=
|
||||||
|
nodes[params.num_nodes_at_this_level + 4 * idx + 0].num_points();
|
||||||
|
num_points_in_children +=
|
||||||
|
nodes[params.num_nodes_at_this_level + 4 * idx + 1].num_points();
|
||||||
|
num_points_in_children +=
|
||||||
|
nodes[params.num_nodes_at_this_level + 4 * idx + 2].num_points();
|
||||||
|
num_points_in_children +=
|
||||||
|
nodes[params.num_nodes_at_this_level + 4 * idx + 3].num_points();
|
||||||
|
|
||||||
|
if (num_points_in_children != node.num_points()) return false;
|
||||||
|
|
||||||
|
return check_quadtree(&nodes[params.num_nodes_at_this_level], 4 * idx + 0,
|
||||||
|
num_pts, pts, Parameters(params, true)) &&
|
||||||
|
check_quadtree(&nodes[params.num_nodes_at_this_level], 4 * idx + 1,
|
||||||
|
num_pts, pts, Parameters(params, true)) &&
|
||||||
|
check_quadtree(&nodes[params.num_nodes_at_this_level], 4 * idx + 2,
|
||||||
|
num_pts, pts, Parameters(params, true)) &&
|
||||||
|
check_quadtree(&nodes[params.num_nodes_at_this_level], 4 * idx + 3,
|
||||||
|
num_pts, pts, Parameters(params, true));
|
||||||
|
}
|
||||||
|
|
||||||
|
const Bounding_box &bbox = node.bounding_box();
|
||||||
|
|
||||||
|
for (int it = node.points_begin(); it < node.points_end(); ++it) {
|
||||||
|
if (it >= num_pts) return false;
|
||||||
|
|
||||||
|
float2 p = pts->get_point(it);
|
||||||
|
|
||||||
|
if (!bbox.contains(p)) return false;
|
||||||
|
}
|
||||||
|
|
||||||
|
return true;
|
||||||
|
}
|
||||||
|
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
// Parallel random number generator.
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
struct Random_generator {
|
||||||
|
int count;
|
||||||
|
|
||||||
|
__host__ __device__ Random_generator() : count(0) {}
|
||||||
|
__host__ __device__ unsigned int hash(unsigned int a) {
|
||||||
|
a = (a + 0x7ed55d16) + (a << 12);
|
||||||
|
a = (a ^ 0xc761c23c) ^ (a >> 19);
|
||||||
|
a = (a + 0x165667b1) + (a << 5);
|
||||||
|
a = (a + 0xd3a2646c) ^ (a << 9);
|
||||||
|
a = (a + 0xfd7046c5) + (a << 3);
|
||||||
|
a = (a ^ 0xb55a4f09) ^ (a >> 16);
|
||||||
|
return a;
|
||||||
|
}
|
||||||
|
|
||||||
|
__host__ __device__ __forceinline__ thrust::tuple<float, float> operator()() {
|
||||||
|
#ifdef __CUDA_ARCH__
|
||||||
|
unsigned seed = hash(blockIdx.x * blockDim.x + threadIdx.x + count);
|
||||||
|
// thrust::generate may call operator() more than once per thread.
|
||||||
|
// Hence, increment count by grid size to ensure uniqueness of seed
|
||||||
|
count += blockDim.x * gridDim.x;
|
||||||
|
#else
|
||||||
|
unsigned seed = hash(0);
|
||||||
|
#endif
|
||||||
|
thrust::default_random_engine rng(seed);
|
||||||
|
thrust::random::uniform_real_distribution<float> distrib;
|
||||||
|
return thrust::make_tuple(distrib(rng), distrib(rng));
|
||||||
|
}
|
||||||
|
};
|
||||||
|
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
// Allocate GPU structs, launch kernel and clean up
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
bool cdpQuadtree(int warp_size) {
|
||||||
|
// Constants to control the algorithm.
|
||||||
|
const int num_points = 1024;
|
||||||
|
const int max_depth = 8;
|
||||||
|
const int min_points_per_node = 16;
|
||||||
|
|
||||||
|
// Allocate memory for points.
|
||||||
|
thrust::device_vector<float> x_d0(num_points);
|
||||||
|
thrust::device_vector<float> x_d1(num_points);
|
||||||
|
thrust::device_vector<float> y_d0(num_points);
|
||||||
|
thrust::device_vector<float> y_d1(num_points);
|
||||||
|
|
||||||
|
// Generate random points.
|
||||||
|
Random_generator rnd;
|
||||||
|
thrust::generate(
|
||||||
|
thrust::make_zip_iterator(thrust::make_tuple(x_d0.begin(), y_d0.begin())),
|
||||||
|
thrust::make_zip_iterator(thrust::make_tuple(x_d0.end(), y_d0.end())),
|
||||||
|
rnd);
|
||||||
|
|
||||||
|
// Host structures to analyze the device ones.
|
||||||
|
Points points_init[2];
|
||||||
|
points_init[0].set(thrust::raw_pointer_cast(&x_d0[0]),
|
||||||
|
thrust::raw_pointer_cast(&y_d0[0]));
|
||||||
|
points_init[1].set(thrust::raw_pointer_cast(&x_d1[0]),
|
||||||
|
thrust::raw_pointer_cast(&y_d1[0]));
|
||||||
|
|
||||||
|
// Allocate memory to store points.
|
||||||
|
Points *points;
|
||||||
|
checkCudaErrors(cudaMalloc((void **)&points, 2 * sizeof(Points)));
|
||||||
|
checkCudaErrors(cudaMemcpy(points, points_init, 2 * sizeof(Points),
|
||||||
|
cudaMemcpyHostToDevice));
|
||||||
|
|
||||||
|
// We could use a close form...
|
||||||
|
int max_nodes = 0;
|
||||||
|
|
||||||
|
for (int i = 0, num_nodes_at_level = 1; i < max_depth;
|
||||||
|
++i, num_nodes_at_level *= 4)
|
||||||
|
max_nodes += num_nodes_at_level;
|
||||||
|
|
||||||
|
// Allocate memory to store the tree.
|
||||||
|
Quadtree_node root;
|
||||||
|
root.set_range(0, num_points);
|
||||||
|
Quadtree_node *nodes;
|
||||||
|
checkCudaErrors(
|
||||||
|
cudaMalloc((void **)&nodes, max_nodes * sizeof(Quadtree_node)));
|
||||||
|
checkCudaErrors(
|
||||||
|
cudaMemcpy(nodes, &root, sizeof(Quadtree_node), cudaMemcpyHostToDevice));
|
||||||
|
|
||||||
|
// We set the recursion limit for CDP to max_depth.
|
||||||
|
cudaDeviceSetLimit(cudaLimitDevRuntimeSyncDepth, max_depth);
|
||||||
|
|
||||||
|
// Build the quadtree.
|
||||||
|
Parameters params(max_depth, min_points_per_node);
|
||||||
|
std::cout << "Launching CDP kernel to build the quadtree" << std::endl;
|
||||||
|
const int NUM_THREADS_PER_BLOCK = 128; // Do not use less than 128 threads.
|
||||||
|
const int NUM_WARPS_PER_BLOCK = NUM_THREADS_PER_BLOCK / warp_size;
|
||||||
|
const size_t smem_size = 4 * NUM_WARPS_PER_BLOCK * sizeof(int);
|
||||||
|
build_quadtree_kernel<
|
||||||
|
NUM_THREADS_PER_BLOCK><<<1, NUM_THREADS_PER_BLOCK, smem_size>>>(
|
||||||
|
nodes, points, params);
|
||||||
|
checkCudaErrors(cudaGetLastError());
|
||||||
|
|
||||||
|
// Copy points to CPU.
|
||||||
|
thrust::host_vector<float> x_h(x_d0);
|
||||||
|
thrust::host_vector<float> y_h(y_d0);
|
||||||
|
Points host_points;
|
||||||
|
host_points.set(thrust::raw_pointer_cast(&x_h[0]),
|
||||||
|
thrust::raw_pointer_cast(&y_h[0]));
|
||||||
|
|
||||||
|
// Copy nodes to CPU.
|
||||||
|
Quadtree_node *host_nodes = new Quadtree_node[max_nodes];
|
||||||
|
checkCudaErrors(cudaMemcpy(host_nodes, nodes,
|
||||||
|
max_nodes * sizeof(Quadtree_node),
|
||||||
|
cudaMemcpyDeviceToHost));
|
||||||
|
|
||||||
|
// Validate the results.
|
||||||
|
bool ok = check_quadtree(host_nodes, 0, num_points, &host_points, params);
|
||||||
|
std::cout << "Results: " << (ok ? "OK" : "FAILED") << std::endl;
|
||||||
|
|
||||||
|
// Free CPU memory.
|
||||||
|
delete[] host_nodes;
|
||||||
|
|
||||||
|
// Free memory.
|
||||||
|
checkCudaErrors(cudaFree(nodes));
|
||||||
|
checkCudaErrors(cudaFree(points));
|
||||||
|
|
||||||
|
return ok;
|
||||||
|
}
|
||||||
|
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
// Main entry point.
|
||||||
|
////////////////////////////////////////////////////////////////////////////////
|
||||||
|
int main(int argc, char **argv) {
|
||||||
|
// Find/set the device.
|
||||||
|
// The test requires an architecture SM35 or greater (CDP capable).
|
||||||
|
int cuda_device = findCudaDevice(argc, (const char **)argv);
|
||||||
|
cudaDeviceProp deviceProps;
|
||||||
|
checkCudaErrors(cudaGetDeviceProperties(&deviceProps, cuda_device));
|
||||||
|
int cdpCapable = (deviceProps.major == 3 && deviceProps.minor >= 5) ||
|
||||||
|
deviceProps.major >= 4;
|
||||||
|
|
||||||
|
printf("GPU device %s has compute capabilities (SM %d.%d)\n",
|
||||||
|
deviceProps.name, deviceProps.major, deviceProps.minor);
|
||||||
|
|
||||||
|
if (!cdpCapable) {
|
||||||
|
std::cerr << "cdpQuadTree requires SM 3.5 or higher to use CUDA Dynamic "
|
||||||
|
"Parallelism. Exiting...\n"
|
||||||
|
<< std::endl;
|
||||||
|
exit(EXIT_WAIVED);
|
||||||
|
}
|
||||||
|
|
||||||
|
bool ok = cdpQuadtree(deviceProps.warpSize);
|
||||||
|
|
||||||
|
return (ok ? EXIT_SUCCESS : EXIT_FAILURE);
|
||||||
|
}
|
||||||
20
Samples/cdpQuadtree/cdpQuadtree_vs2017.sln
Normal file
20
Samples/cdpQuadtree/cdpQuadtree_vs2017.sln
Normal file
@@ -0,0 +1,20 @@
|
|||||||
|
|
||||||
|
Microsoft Visual Studio Solution File, Format Version 12.00
|
||||||
|
# Visual Studio 2017
|
||||||
|
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "cdpQuadtree", "cdpQuadtree_vs2017.vcxproj", "{997E0757-EA74-4A4E-A0FC-47D8C8831A15}"
|
||||||
|
EndProject
|
||||||
|
Global
|
||||||
|
GlobalSection(SolutionConfigurationPlatforms) = preSolution
|
||||||
|
Debug|x64 = Debug|x64
|
||||||
|
Release|x64 = Release|x64
|
||||||
|
EndGlobalSection
|
||||||
|
GlobalSection(ProjectConfigurationPlatforms) = postSolution
|
||||||
|
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.ActiveCfg = Debug|x64
|
||||||
|
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.Build.0 = Debug|x64
|
||||||
|
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.ActiveCfg = Release|x64
|
||||||
|
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.Build.0 = Release|x64
|
||||||
|
EndGlobalSection
|
||||||
|
GlobalSection(SolutionProperties) = preSolution
|
||||||
|
HideSolutionNode = FALSE
|
||||||
|
EndGlobalSection
|
||||||
|
EndGlobal
|
||||||
114
Samples/cdpQuadtree/cdpQuadtree_vs2017.vcxproj
Normal file
114
Samples/cdpQuadtree/cdpQuadtree_vs2017.vcxproj
Normal file
@@ -0,0 +1,114 @@
|
|||||||
|
<?xml version="1.0" encoding="utf-8"?>
|
||||||
|
<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
|
||||||
|
<PropertyGroup>
|
||||||
|
<CUDAPropsPath Condition="'$(CUDAPropsPath)'==''">$(VCTargetsPath)\BuildCustomizations</CUDAPropsPath>
|
||||||
|
</PropertyGroup>
|
||||||
|
<ItemGroup Label="ProjectConfigurations">
|
||||||
|
<ProjectConfiguration Include="Debug|x64">
|
||||||
|
<Configuration>Debug</Configuration>
|
||||||
|
<Platform>x64</Platform>
|
||||||
|
</ProjectConfiguration>
|
||||||
|
<ProjectConfiguration Include="Release|x64">
|
||||||
|
<Configuration>Release</Configuration>
|
||||||
|
<Platform>x64</Platform>
|
||||||
|
</ProjectConfiguration>
|
||||||
|
</ItemGroup>
|
||||||
|
<PropertyGroup Label="Globals">
|
||||||
|
<ProjectGuid>{997E0757-EA74-4A4E-A0FC-47D8C8831A15}</ProjectGuid>
|
||||||
|
<RootNamespace>cdpQuadtree_vs2017</RootNamespace>
|
||||||
|
<ProjectName>cdpQuadtree</ProjectName>
|
||||||
|
<CudaToolkitCustomDir />
|
||||||
|
</PropertyGroup>
|
||||||
|
<PropertyGroup Condition="'$(WindowsTargetPlatformVersion)'==''">
|
||||||
|
<LatestTargetPlatformVersion>$([Microsoft.Build.Utilities.ToolLocationHelper]::GetLatestSDKTargetPlatformVersion('Windows', '10.0'))</LatestTargetPlatformVersion>
|
||||||
|
<WindowsTargetPlatformVersion Condition="'$(WindowsTargetPlatformVersion)' == ''">$(LatestTargetPlatformVersion)</WindowsTargetPlatformVersion>
|
||||||
|
<TargetPlatformVersion>$(WindowsTargetPlatformVersion)</TargetPlatformVersion>
|
||||||
|
</PropertyGroup>
|
||||||
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
|
||||||
|
<PropertyGroup>
|
||||||
|
<ConfigurationType>Application</ConfigurationType>
|
||||||
|
<CharacterSet>MultiByte</CharacterSet>
|
||||||
|
<PlatformToolset>v141</PlatformToolset>
|
||||||
|
</PropertyGroup>
|
||||||
|
<PropertyGroup Condition="'$(Configuration)'=='Debug'">
|
||||||
|
<UseDebugLibraries>true</UseDebugLibraries>
|
||||||
|
</PropertyGroup>
|
||||||
|
<PropertyGroup Condition="'$(Configuration)'=='Release'">
|
||||||
|
<WholeProgramOptimization>true</WholeProgramOptimization>
|
||||||
|
</PropertyGroup>
|
||||||
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
|
<ImportGroup Label="ExtensionSettings">
|
||||||
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
|
</ImportGroup>
|
||||||
|
<ImportGroup Label="PropertySheets">
|
||||||
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
|
</ImportGroup>
|
||||||
|
<PropertyGroup Label="UserMacros" />
|
||||||
|
<PropertyGroup>
|
||||||
|
<IntDir>$(Platform)/$(Configuration)/</IntDir>
|
||||||
|
<IncludePath>$(IncludePath)</IncludePath>
|
||||||
|
<CodeAnalysisRuleSet>AllRules.ruleset</CodeAnalysisRuleSet>
|
||||||
|
<CodeAnalysisRules />
|
||||||
|
<CodeAnalysisRuleAssemblies />
|
||||||
|
</PropertyGroup>
|
||||||
|
<PropertyGroup Condition="'$(Platform)'=='x64'">
|
||||||
|
<OutDir>../../bin/win64/$(Configuration)/</OutDir>
|
||||||
|
</PropertyGroup>
|
||||||
|
<ItemDefinitionGroup>
|
||||||
|
<ClCompile>
|
||||||
|
<WarningLevel>Level3</WarningLevel>
|
||||||
|
<PreprocessorDefinitions>WIN32;_MBCS;%(PreprocessorDefinitions)</PreprocessorDefinitions>
|
||||||
|
<AdditionalIncludeDirectories>./;$(CudaToolkitDir)/include;../../Common;$(CudaToolkitIncludeDir);</AdditionalIncludeDirectories>
|
||||||
|
</ClCompile>
|
||||||
|
<Link>
|
||||||
|
<SubSystem>Console</SubSystem>
|
||||||
|
<AdditionalDependencies>cudadevrt.lib;cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||||
|
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||||
|
<OutputFile>$(OutDir)/cdpQuadtree.exe</OutputFile>
|
||||||
|
</Link>
|
||||||
|
<CudaCompile>
|
||||||
|
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||||
|
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
|
||||||
|
<Include>./;../../Common</Include>
|
||||||
|
<Defines>WIN32</Defines>
|
||||||
|
<GenerateRelocatableDeviceCode>true</GenerateRelocatableDeviceCode>
|
||||||
|
<AdditionalOptions>--threads 0</AdditionalOptions>
|
||||||
|
</CudaCompile>
|
||||||
|
</ItemDefinitionGroup>
|
||||||
|
<ItemDefinitionGroup Condition="'$(Configuration)'=='Debug'">
|
||||||
|
<ClCompile>
|
||||||
|
<Optimization>Disabled</Optimization>
|
||||||
|
<RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
|
||||||
|
</ClCompile>
|
||||||
|
<Link>
|
||||||
|
<GenerateDebugInformation>true</GenerateDebugInformation>
|
||||||
|
<LinkTimeCodeGeneration>Default</LinkTimeCodeGeneration>
|
||||||
|
</Link>
|
||||||
|
<CudaCompile>
|
||||||
|
<Runtime>MTd</Runtime>
|
||||||
|
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||||
|
</CudaCompile>
|
||||||
|
</ItemDefinitionGroup>
|
||||||
|
<ItemDefinitionGroup Condition="'$(Configuration)'=='Release'">
|
||||||
|
<ClCompile>
|
||||||
|
<Optimization>MaxSpeed</Optimization>
|
||||||
|
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||||
|
</ClCompile>
|
||||||
|
<Link>
|
||||||
|
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||||
|
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||||
|
</Link>
|
||||||
|
<CudaCompile>
|
||||||
|
<Runtime>MT</Runtime>
|
||||||
|
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||||
|
</CudaCompile>
|
||||||
|
</ItemDefinitionGroup>
|
||||||
|
<ItemGroup>
|
||||||
|
<CudaCompile Include="cdpQuadtree.cu" />
|
||||||
|
|
||||||
|
</ItemGroup>
|
||||||
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
|
<ImportGroup Label="ExtensionTargets">
|
||||||
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
|
</ImportGroup>
|
||||||
|
</Project>
|
||||||
20
Samples/cdpQuadtree/cdpQuadtree_vs2019.sln
Normal file
20
Samples/cdpQuadtree/cdpQuadtree_vs2019.sln
Normal file
@@ -0,0 +1,20 @@
|
|||||||
|
|
||||||
|
Microsoft Visual Studio Solution File, Format Version 12.00
|
||||||
|
# Visual Studio 2019
|
||||||
|
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "cdpQuadtree", "cdpQuadtree_vs2019.vcxproj", "{997E0757-EA74-4A4E-A0FC-47D8C8831A15}"
|
||||||
|
EndProject
|
||||||
|
Global
|
||||||
|
GlobalSection(SolutionConfigurationPlatforms) = preSolution
|
||||||
|
Debug|x64 = Debug|x64
|
||||||
|
Release|x64 = Release|x64
|
||||||
|
EndGlobalSection
|
||||||
|
GlobalSection(ProjectConfigurationPlatforms) = postSolution
|
||||||
|
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.ActiveCfg = Debug|x64
|
||||||
|
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.Build.0 = Debug|x64
|
||||||
|
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.ActiveCfg = Release|x64
|
||||||
|
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.Build.0 = Release|x64
|
||||||
|
EndGlobalSection
|
||||||
|
GlobalSection(SolutionProperties) = preSolution
|
||||||
|
HideSolutionNode = FALSE
|
||||||
|
EndGlobalSection
|
||||||
|
EndGlobal
|
||||||
110
Samples/cdpQuadtree/cdpQuadtree_vs2019.vcxproj
Normal file
110
Samples/cdpQuadtree/cdpQuadtree_vs2019.vcxproj
Normal file
@@ -0,0 +1,110 @@
|
|||||||
|
<?xml version="1.0" encoding="utf-8"?>
|
||||||
|
<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
|
||||||
|
<PropertyGroup>
|
||||||
|
<CUDAPropsPath Condition="'$(CUDAPropsPath)'==''">$(VCTargetsPath)\BuildCustomizations</CUDAPropsPath>
|
||||||
|
</PropertyGroup>
|
||||||
|
<ItemGroup Label="ProjectConfigurations">
|
||||||
|
<ProjectConfiguration Include="Debug|x64">
|
||||||
|
<Configuration>Debug</Configuration>
|
||||||
|
<Platform>x64</Platform>
|
||||||
|
</ProjectConfiguration>
|
||||||
|
<ProjectConfiguration Include="Release|x64">
|
||||||
|
<Configuration>Release</Configuration>
|
||||||
|
<Platform>x64</Platform>
|
||||||
|
</ProjectConfiguration>
|
||||||
|
</ItemGroup>
|
||||||
|
<PropertyGroup Label="Globals">
|
||||||
|
<ProjectGuid>{997E0757-EA74-4A4E-A0FC-47D8C8831A15}</ProjectGuid>
|
||||||
|
<RootNamespace>cdpQuadtree_vs2019</RootNamespace>
|
||||||
|
<ProjectName>cdpQuadtree</ProjectName>
|
||||||
|
<CudaToolkitCustomDir />
|
||||||
|
</PropertyGroup>
|
||||||
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
|
||||||
|
<PropertyGroup>
|
||||||
|
<ConfigurationType>Application</ConfigurationType>
|
||||||
|
<CharacterSet>MultiByte</CharacterSet>
|
||||||
|
<PlatformToolset>v142</PlatformToolset>
|
||||||
|
<WindowsTargetPlatformVersion>10.0</WindowsTargetPlatformVersion>
|
||||||
|
</PropertyGroup>
|
||||||
|
<PropertyGroup Condition="'$(Configuration)'=='Debug'">
|
||||||
|
<UseDebugLibraries>true</UseDebugLibraries>
|
||||||
|
</PropertyGroup>
|
||||||
|
<PropertyGroup Condition="'$(Configuration)'=='Release'">
|
||||||
|
<WholeProgramOptimization>true</WholeProgramOptimization>
|
||||||
|
</PropertyGroup>
|
||||||
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
|
<ImportGroup Label="ExtensionSettings">
|
||||||
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
|
</ImportGroup>
|
||||||
|
<ImportGroup Label="PropertySheets">
|
||||||
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
|
</ImportGroup>
|
||||||
|
<PropertyGroup Label="UserMacros" />
|
||||||
|
<PropertyGroup>
|
||||||
|
<IntDir>$(Platform)/$(Configuration)/</IntDir>
|
||||||
|
<IncludePath>$(IncludePath)</IncludePath>
|
||||||
|
<CodeAnalysisRuleSet>AllRules.ruleset</CodeAnalysisRuleSet>
|
||||||
|
<CodeAnalysisRules />
|
||||||
|
<CodeAnalysisRuleAssemblies />
|
||||||
|
</PropertyGroup>
|
||||||
|
<PropertyGroup Condition="'$(Platform)'=='x64'">
|
||||||
|
<OutDir>../../bin/win64/$(Configuration)/</OutDir>
|
||||||
|
</PropertyGroup>
|
||||||
|
<ItemDefinitionGroup>
|
||||||
|
<ClCompile>
|
||||||
|
<WarningLevel>Level3</WarningLevel>
|
||||||
|
<PreprocessorDefinitions>WIN32;_MBCS;%(PreprocessorDefinitions)</PreprocessorDefinitions>
|
||||||
|
<AdditionalIncludeDirectories>./;$(CudaToolkitDir)/include;../../Common;$(CudaToolkitIncludeDir);</AdditionalIncludeDirectories>
|
||||||
|
</ClCompile>
|
||||||
|
<Link>
|
||||||
|
<SubSystem>Console</SubSystem>
|
||||||
|
<AdditionalDependencies>cudadevrt.lib;cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||||
|
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||||
|
<OutputFile>$(OutDir)/cdpQuadtree.exe</OutputFile>
|
||||||
|
</Link>
|
||||||
|
<CudaCompile>
|
||||||
|
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||||
|
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
|
||||||
|
<Include>./;../../Common</Include>
|
||||||
|
<Defines>WIN32</Defines>
|
||||||
|
<GenerateRelocatableDeviceCode>true</GenerateRelocatableDeviceCode>
|
||||||
|
<AdditionalOptions>--threads 0</AdditionalOptions>
|
||||||
|
</CudaCompile>
|
||||||
|
</ItemDefinitionGroup>
|
||||||
|
<ItemDefinitionGroup Condition="'$(Configuration)'=='Debug'">
|
||||||
|
<ClCompile>
|
||||||
|
<Optimization>Disabled</Optimization>
|
||||||
|
<RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
|
||||||
|
</ClCompile>
|
||||||
|
<Link>
|
||||||
|
<GenerateDebugInformation>true</GenerateDebugInformation>
|
||||||
|
<LinkTimeCodeGeneration>Default</LinkTimeCodeGeneration>
|
||||||
|
</Link>
|
||||||
|
<CudaCompile>
|
||||||
|
<Runtime>MTd</Runtime>
|
||||||
|
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||||
|
</CudaCompile>
|
||||||
|
</ItemDefinitionGroup>
|
||||||
|
<ItemDefinitionGroup Condition="'$(Configuration)'=='Release'">
|
||||||
|
<ClCompile>
|
||||||
|
<Optimization>MaxSpeed</Optimization>
|
||||||
|
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||||
|
</ClCompile>
|
||||||
|
<Link>
|
||||||
|
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||||
|
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||||
|
</Link>
|
||||||
|
<CudaCompile>
|
||||||
|
<Runtime>MT</Runtime>
|
||||||
|
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||||
|
</CudaCompile>
|
||||||
|
</ItemDefinitionGroup>
|
||||||
|
<ItemGroup>
|
||||||
|
<CudaCompile Include="cdpQuadtree.cu" />
|
||||||
|
|
||||||
|
</ItemGroup>
|
||||||
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
|
<ImportGroup Label="ExtensionTargets">
|
||||||
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
|
</ImportGroup>
|
||||||
|
</Project>
|
||||||
18
Samples/concurrentKernels/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/concurrentKernels/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "Linux",
|
||||||
|
"includePath": [
|
||||||
|
"${workspaceFolder}/**",
|
||||||
|
"${workspaceFolder}/../../Common"
|
||||||
|
],
|
||||||
|
"defines": [],
|
||||||
|
"compilerPath": "/usr/local/cuda/bin/nvcc",
|
||||||
|
"cStandard": "gnu17",
|
||||||
|
"cppStandard": "gnu++14",
|
||||||
|
"intelliSenseMode": "linux-gcc-x64",
|
||||||
|
"configurationProvider": "ms-vscode.makefile-tools"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"version": 4
|
||||||
|
}
|
||||||
7
Samples/concurrentKernels/.vscode/extensions.json
vendored
Normal file
7
Samples/concurrentKernels/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/concurrentKernels/.vscode/launch.json
vendored
Normal file
10
Samples/concurrentKernels/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/concurrentKernels"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/concurrentKernels/.vscode/tasks.json
vendored
Normal file
15
Samples/concurrentKernels/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@@ -24,7 +24,7 @@ x86_64, ppc64le, armv7l
|
|||||||
|
|
||||||
## Prerequisites
|
## Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
|
|
||||||
## Build and Run
|
## Build and Run
|
||||||
|
|
||||||
|
|||||||
@@ -38,7 +38,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -108,6 +108,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -34,7 +34,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -104,6 +104,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
18
Samples/conjugateGradientCudaGraphs/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/conjugateGradientCudaGraphs/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "Linux",
|
||||||
|
"includePath": [
|
||||||
|
"${workspaceFolder}/**",
|
||||||
|
"${workspaceFolder}/../../Common"
|
||||||
|
],
|
||||||
|
"defines": [],
|
||||||
|
"compilerPath": "/usr/local/cuda/bin/nvcc",
|
||||||
|
"cStandard": "gnu17",
|
||||||
|
"cppStandard": "gnu++14",
|
||||||
|
"intelliSenseMode": "linux-gcc-x64",
|
||||||
|
"configurationProvider": "ms-vscode.makefile-tools"
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"version": 4
|
||||||
|
}
|
||||||
7
Samples/conjugateGradientCudaGraphs/.vscode/extensions.json
vendored
Normal file
7
Samples/conjugateGradientCudaGraphs/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
|||||||
|
{
|
||||||
|
"recommendations": [
|
||||||
|
"nvidia.nsight-vscode-edition",
|
||||||
|
"ms-vscode.cpptools",
|
||||||
|
"ms-vscode.makefile-tools"
|
||||||
|
]
|
||||||
|
}
|
||||||
10
Samples/conjugateGradientCudaGraphs/.vscode/launch.json
vendored
Normal file
10
Samples/conjugateGradientCudaGraphs/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "CUDA C++: Launch",
|
||||||
|
"type": "cuda-gdb",
|
||||||
|
"request": "launch",
|
||||||
|
"program": "${workspaceFolder}/conjugateGradientCudaGraphs"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
15
Samples/conjugateGradientCudaGraphs/.vscode/tasks.json
vendored
Normal file
15
Samples/conjugateGradientCudaGraphs/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
|||||||
|
{
|
||||||
|
"version": "2.0.0",
|
||||||
|
"tasks": [
|
||||||
|
{
|
||||||
|
"label": "sample",
|
||||||
|
"type": "shell",
|
||||||
|
"command": "make dbg=1",
|
||||||
|
"problemMatcher": ["$nvcc"],
|
||||||
|
"group": {
|
||||||
|
"kind": "build",
|
||||||
|
"isDefault": true
|
||||||
|
}
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
@@ -30,7 +30,7 @@ cudaStreamBeginCapture, cudaStreamEndCapture, cudaGraphCreate, cudaGraphLaunch,
|
|||||||
|
|
||||||
## Prerequisites
|
## Prerequisites
|
||||||
|
|
||||||
Download and install the [CUDA Toolkit 11.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
Download and install the [CUDA Toolkit 11.4](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||||
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
||||||
|
|
||||||
## Build and Run
|
## Build and Run
|
||||||
|
|||||||
@@ -25,7 +25,6 @@
|
|||||||
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||||
*/
|
*/
|
||||||
|
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* This sample implements a conjugate gradient solver on GPU
|
* This sample implements a conjugate gradient solver on GPU
|
||||||
* using CUBLAS and CUSPARSE with CUDA Graphs
|
* using CUBLAS and CUSPARSE with CUDA Graphs
|
||||||
@@ -46,7 +45,6 @@
|
|||||||
#include <helper_cuda.h> // helper function CUDA error checking and initialization
|
#include <helper_cuda.h> // helper function CUDA error checking and initialization
|
||||||
#include <helper_functions.h> // helper for shared functions common to CUDA Samples
|
#include <helper_functions.h> // helper for shared functions common to CUDA Samples
|
||||||
|
|
||||||
|
|
||||||
const char *sSDKname = "conjugateGradientCudaGraphs";
|
const char *sSDKname = "conjugateGradientCudaGraphs";
|
||||||
|
|
||||||
#ifndef WITH_GRAPH
|
#ifndef WITH_GRAPH
|
||||||
@@ -145,12 +143,12 @@ int main(int argc, char **argv) {
|
|||||||
/* Generate a random tridiagonal symmetric matrix in CSR format */
|
/* Generate a random tridiagonal symmetric matrix in CSR format */
|
||||||
N = 1048576;
|
N = 1048576;
|
||||||
nz = (N - 2) * 3 + 4;
|
nz = (N - 2) * 3 + 4;
|
||||||
I = (int *)malloc(sizeof(int) * (N + 1));
|
checkCudaErrors(cudaMallocHost(&I, sizeof(int) * (N + 1)));
|
||||||
J = (int *)malloc(sizeof(int) * nz);
|
checkCudaErrors(cudaMallocHost(&J, sizeof(int) * nz));
|
||||||
val = (float *)malloc(sizeof(float) * nz);
|
checkCudaErrors(cudaMallocHost(&val, sizeof(float) * nz));
|
||||||
genTridiag(I, J, val, N, nz);
|
genTridiag(I, J, val, N, nz);
|
||||||
|
|
||||||
x = (float *)malloc(sizeof(float) * N);
|
checkCudaErrors(cudaMallocHost(&x, sizeof(float) * N));
|
||||||
rhs = (float *)malloc(sizeof(float) * N);
|
rhs = (float *)malloc(sizeof(float) * N);
|
||||||
|
|
||||||
for (int i = 0; i < N; i++) {
|
for (int i = 0; i < N; i++) {
|
||||||
@@ -192,9 +190,9 @@ int main(int argc, char **argv) {
|
|||||||
|
|
||||||
/* Wrap raw data into cuSPARSE generic API objects */
|
/* Wrap raw data into cuSPARSE generic API objects */
|
||||||
cusparseSpMatDescr_t matA = NULL;
|
cusparseSpMatDescr_t matA = NULL;
|
||||||
checkCudaErrors(cusparseCreateCsr(
|
checkCudaErrors(cusparseCreateCsr(&matA, N, N, nz, d_row, d_col, d_val,
|
||||||
&matA, N, N, nz, d_row, d_col, d_val, CUSPARSE_INDEX_32I,
|
CUSPARSE_INDEX_32I, CUSPARSE_INDEX_32I,
|
||||||
CUSPARSE_INDEX_32I, CUSPARSE_INDEX_BASE_ZERO, CUDA_R_32F));
|
CUSPARSE_INDEX_BASE_ZERO, CUDA_R_32F));
|
||||||
cusparseDnVecDescr_t vecx = NULL;
|
cusparseDnVecDescr_t vecx = NULL;
|
||||||
checkCudaErrors(cusparseCreateDnVec(&vecx, N, d_x, CUDA_R_32F));
|
checkCudaErrors(cusparseCreateDnVec(&vecx, N, d_x, CUDA_R_32F));
|
||||||
cusparseDnVecDescr_t vecp = NULL;
|
cusparseDnVecDescr_t vecp = NULL;
|
||||||
@@ -206,7 +204,7 @@ int main(int argc, char **argv) {
|
|||||||
size_t bufferSize = 0;
|
size_t bufferSize = 0;
|
||||||
checkCudaErrors(cusparseSpMV_bufferSize(
|
checkCudaErrors(cusparseSpMV_bufferSize(
|
||||||
cusparseHandle, CUSPARSE_OPERATION_NON_TRANSPOSE, &alpha, matA, vecx,
|
cusparseHandle, CUSPARSE_OPERATION_NON_TRANSPOSE, &alpha, matA, vecx,
|
||||||
&beta, vecAx, CUDA_R_32F, CUSPARSE_MV_ALG_DEFAULT, &bufferSize));
|
&beta, vecAx, CUDA_R_32F, CUSPARSE_SPMV_ALG_DEFAULT, &bufferSize));
|
||||||
void *buffer = NULL;
|
void *buffer = NULL;
|
||||||
checkCudaErrors(cudaMalloc(&buffer, bufferSize));
|
checkCudaErrors(cudaMalloc(&buffer, bufferSize));
|
||||||
|
|
||||||
@@ -234,9 +232,9 @@ int main(int argc, char **argv) {
|
|||||||
beta = 0.0;
|
beta = 0.0;
|
||||||
|
|
||||||
checkCudaErrors(cusparseSetStream(cusparseHandle, stream1));
|
checkCudaErrors(cusparseSetStream(cusparseHandle, stream1));
|
||||||
checkCudaErrors(cusparseSpMV(
|
checkCudaErrors(cusparseSpMV(cusparseHandle, CUSPARSE_OPERATION_NON_TRANSPOSE,
|
||||||
cusparseHandle, CUSPARSE_OPERATION_NON_TRANSPOSE, &alpha, matA, vecx,
|
&alpha, matA, vecx, &beta, vecAx, CUDA_R_32F,
|
||||||
&beta, vecAx, CUDA_R_32F, CUSPARSE_MV_ALG_DEFAULT, &buffer));
|
CUSPARSE_SPMV_ALG_DEFAULT, buffer));
|
||||||
|
|
||||||
checkCudaErrors(cublasSetStream(cublasHandle, stream1));
|
checkCudaErrors(cublasSetStream(cublasHandle, stream1));
|
||||||
checkCudaErrors(cublasSaxpy(cublasHandle, N, &alpham1, d_Ax, 1, d_r, 1));
|
checkCudaErrors(cublasSaxpy(cublasHandle, N, &alpham1, d_Ax, 1, d_r, 1));
|
||||||
@@ -248,9 +246,9 @@ int main(int argc, char **argv) {
|
|||||||
k = 1;
|
k = 1;
|
||||||
// First Iteration when k=1 starts
|
// First Iteration when k=1 starts
|
||||||
checkCudaErrors(cublasScopy(cublasHandle, N, d_r, 1, d_p, 1));
|
checkCudaErrors(cublasScopy(cublasHandle, N, d_r, 1, d_p, 1));
|
||||||
checkCudaErrors(cusparseSpMV(
|
checkCudaErrors(cusparseSpMV(cusparseHandle, CUSPARSE_OPERATION_NON_TRANSPOSE,
|
||||||
cusparseHandle, CUSPARSE_OPERATION_NON_TRANSPOSE, &alpha, matA, vecp,
|
&alpha, matA, vecp, &beta, vecAx, CUDA_R_32F,
|
||||||
&beta, vecAx, CUDA_R_32F, CUSPARSE_MV_ALG_DEFAULT, &buffer));
|
CUSPARSE_SPMV_ALG_DEFAULT, buffer));
|
||||||
|
|
||||||
checkCudaErrors(cublasSdot(cublasHandle, N, d_p, 1, d_Ax, 1, d_dot));
|
checkCudaErrors(cublasSdot(cublasHandle, N, d_p, 1, d_Ax, 1, d_dot));
|
||||||
|
|
||||||
@@ -290,9 +288,9 @@ int main(int argc, char **argv) {
|
|||||||
|
|
||||||
checkCudaErrors(
|
checkCudaErrors(
|
||||||
cusparseSetPointerMode(cusparseHandle, CUSPARSE_POINTER_MODE_HOST));
|
cusparseSetPointerMode(cusparseHandle, CUSPARSE_POINTER_MODE_HOST));
|
||||||
checkCudaErrors(cusparseSpMV(
|
checkCudaErrors(cusparseSpMV(cusparseHandle, CUSPARSE_OPERATION_NON_TRANSPOSE,
|
||||||
cusparseHandle, CUSPARSE_OPERATION_NON_TRANSPOSE, &alpha, matA, vecp,
|
&alpha, matA, vecp, &beta, vecAx, CUDA_R_32F,
|
||||||
&beta, vecAx, CUDA_R_32F, CUSPARSE_MV_ALG_DEFAULT, &buffer));
|
CUSPARSE_SPMV_ALG_DEFAULT, buffer));
|
||||||
|
|
||||||
checkCudaErrors(cudaMemsetAsync(d_dot, 0, sizeof(float), stream1));
|
checkCudaErrors(cudaMemsetAsync(d_dot, 0, sizeof(float), stream1));
|
||||||
checkCudaErrors(cublasSdot(cublasHandle, N, d_p, 1, d_Ax, 1, d_dot));
|
checkCudaErrors(cublasSdot(cublasHandle, N, d_p, 1, d_Ax, 1, d_dot));
|
||||||
@@ -335,8 +333,8 @@ int main(int argc, char **argv) {
|
|||||||
checkCudaErrors(cublasSaxpy(cublasHandle, N, &alpha, d_r, 1, d_p, 1));
|
checkCudaErrors(cublasSaxpy(cublasHandle, N, &alpha, d_r, 1, d_p, 1));
|
||||||
|
|
||||||
checkCudaErrors(cusparseSpMV(
|
checkCudaErrors(cusparseSpMV(
|
||||||
cusparseHandle, CUSPARSE_OPERATION_NON_TRANSPOSE, &alpha, matA, vecp,
|
cusparseHandle, CUSPARSE_OPERATION_NON_TRANSPOSE, &alpha, matA, vecp,
|
||||||
&beta, vecAx, CUDA_R_32F, CUSPARSE_MV_ALG_DEFAULT, &buffer));
|
&beta, vecAx, CUDA_R_32F, CUSPARSE_SPMV_ALG_DEFAULT, buffer));
|
||||||
|
|
||||||
cublasSetPointerMode(cublasHandle, CUBLAS_POINTER_MODE_DEVICE);
|
cublasSetPointerMode(cublasHandle, CUBLAS_POINTER_MODE_DEVICE);
|
||||||
checkCudaErrors(cublasSdot(cublasHandle, N, d_p, 1, d_Ax, 1, d_dot));
|
checkCudaErrors(cublasSdot(cublasHandle, N, d_p, 1, d_Ax, 1, d_dot));
|
||||||
@@ -395,23 +393,31 @@ int main(int argc, char **argv) {
|
|||||||
cusparseDestroy(cusparseHandle);
|
cusparseDestroy(cusparseHandle);
|
||||||
cublasDestroy(cublasHandle);
|
cublasDestroy(cublasHandle);
|
||||||
|
|
||||||
if (matA ) { checkCudaErrors(cusparseDestroySpMat(matA)); }
|
if (matA) {
|
||||||
if (vecx ) { checkCudaErrors(cusparseDestroyDnVec(vecx)); }
|
checkCudaErrors(cusparseDestroySpMat(matA));
|
||||||
if (vecAx ) { checkCudaErrors(cusparseDestroyDnVec(vecAx)); }
|
}
|
||||||
if (vecp ) { checkCudaErrors(cusparseDestroyDnVec(vecp)); }
|
if (vecx) {
|
||||||
|
checkCudaErrors(cusparseDestroyDnVec(vecx));
|
||||||
|
}
|
||||||
|
if (vecAx) {
|
||||||
|
checkCudaErrors(cusparseDestroyDnVec(vecAx));
|
||||||
|
}
|
||||||
|
if (vecp) {
|
||||||
|
checkCudaErrors(cusparseDestroyDnVec(vecp));
|
||||||
|
}
|
||||||
|
|
||||||
free(I);
|
checkCudaErrors(cudaFreeHost(I));
|
||||||
free(J);
|
checkCudaErrors(cudaFreeHost(J));
|
||||||
free(val);
|
checkCudaErrors(cudaFreeHost(val));
|
||||||
free(x);
|
checkCudaErrors(cudaFreeHost(x));
|
||||||
free(rhs);
|
free(rhs);
|
||||||
cudaFree(d_col);
|
checkCudaErrors(cudaFree(d_col));
|
||||||
cudaFree(d_row);
|
checkCudaErrors(cudaFree(d_row));
|
||||||
cudaFree(d_val);
|
checkCudaErrors(cudaFree(d_val));
|
||||||
cudaFree(d_x);
|
checkCudaErrors(cudaFree(d_x));
|
||||||
cudaFree(d_r);
|
checkCudaErrors(cudaFree(d_r));
|
||||||
cudaFree(d_p);
|
checkCudaErrors(cudaFree(d_p));
|
||||||
cudaFree(d_Ax);
|
checkCudaErrors(cudaFree(d_Ax));
|
||||||
|
|
||||||
printf("Test Summary: Error amount = %f\n", err);
|
printf("Test Summary: Error amount = %f\n", err);
|
||||||
exit((k <= max_iter) ? 0 : 1);
|
exit((k <= max_iter) ? 0 : 1);
|
||||||
|
|||||||
@@ -38,7 +38,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -108,6 +108,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
@@ -34,7 +34,7 @@
|
|||||||
</PropertyGroup>
|
</PropertyGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||||
<ImportGroup Label="ExtensionSettings">
|
<ImportGroup Label="ExtensionSettings">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.props" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.props" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
<ImportGroup Label="PropertySheets">
|
<ImportGroup Label="PropertySheets">
|
||||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||||
@@ -104,6 +104,6 @@
|
|||||||
</ItemGroup>
|
</ItemGroup>
|
||||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||||
<ImportGroup Label="ExtensionTargets">
|
<ImportGroup Label="ExtensionTargets">
|
||||||
<Import Project="$(CUDAPropsPath)\CUDA 11.2.targets" />
|
<Import Project="$(CUDAPropsPath)\CUDA 11.4.targets" />
|
||||||
</ImportGroup>
|
</ImportGroup>
|
||||||
</Project>
|
</Project>
|
||||||
|
|||||||
18
Samples/conjugateGradientMultiBlockCG/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/conjugateGradientMultiBlockCG/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
|||||||
|
{
|
||||||
|
"configurations": [
|
||||||
|
{
|
||||||
|
"name": "Linux",
|
||||||
|
"includePath": [
|
||||||
|
"${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
|
||||||
|
}
|
||||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user