2 Commits
v12.3 ... v11.6

Author SHA1 Message Date
Rutwik Choughule
7db525e24d update sample cudaNvSci 2022-01-25 17:18:04 +05:30
Rutwik Choughule
b9063abcf4 update lib path for conda 2022-01-18 19:03:54 +05:30
1572 changed files with 5441 additions and 311586 deletions

1
.gitignore vendored
View File

@@ -1 +0,0 @@
.vscode/*

View File

@@ -1,22 +1,5 @@
## Changelog ## Changelog
### CUDA 12.2
* libNVVM samples received updates
* Fixed jitLto Case issues
* Enabled HOST_COMPILER flag to the makefiles for GCC which is untested but may still work.
### CUDA 12.1
* Added new sample for Large Kernels
### CUDA 12.0
* Added new flags for JIT compiling
* Removed deprecated APIs in Hopper Architecture
### CUDA 11.6
* Added new folder structure for samples
* Added support of Visual Studio 2022 to all samples supported on [Windows](#windows-1).
* All CUDA samples are now only available on [GitHub](https://github.com/nvidia/cuda-samples). They are no longer available via CUDA toolkit.
### CUDA 11.5 ### CUDA 11.5
* Added `cuDLAHybridMode`. Demonstrate usage of cuDLA in hybrid mode. * Added `cuDLAHybridMode`. Demonstrate usage of cuDLA in hybrid mode.
* Added `cuDLAStandaloneMode`. Demonstrate usage of cuDLA in standalone mode. * Added `cuDLAStandaloneMode`. Demonstrate usage of cuDLA in standalone mode.

View File

@@ -666,8 +666,6 @@ inline int _ConvertSMVer2Cores(int major, int minor) {
{0x80, 64}, {0x80, 64},
{0x86, 128}, {0x86, 128},
{0x87, 128}, {0x87, 128},
{0x89, 128},
{0x90, 128},
{-1, -1}}; {-1, -1}};
int index = 0; int index = 0;
@@ -714,9 +712,6 @@ inline const char* _ConvertSMVer2ArchName(int major, int minor) {
{0x75, "Turing"}, {0x75, "Turing"},
{0x80, "Ampere"}, {0x80, "Ampere"},
{0x86, "Ampere"}, {0x86, "Ampere"},
{0x87, "Ampere"},
{0x89, "Ada"},
{0x90, "Hopper"},
{-1, "Graphics Device"}}; {-1, "Graphics Device"}};
int index = 0; int index = 0;

View File

@@ -114,8 +114,6 @@ inline int _ConvertSMVer2CoresDRV(int major, int minor) {
{0x80, 64}, {0x80, 64},
{0x86, 128}, {0x86, 128},
{0x87, 128}, {0x87, 128},
{0x89, 128},
{0x90, 128},
{-1, -1}}; {-1, -1}};
int index = 0; int index = 0;

View File

@@ -421,7 +421,6 @@ inline char *sdkFindFilePath(const char *filename,
} }
// File not found // File not found
printf("\nerror: sdkFindFilePath: file <%s> not found!\n", filename);
return 0; return 0;
} }

View File

@@ -49,11 +49,6 @@
void compileFileToCUBIN(char *filename, int argc, char **argv, char **cubinResult, void compileFileToCUBIN(char *filename, int argc, char **argv, char **cubinResult,
size_t *cubinResultSize, int requiresCGheaders) { size_t *cubinResultSize, int requiresCGheaders) {
if (!filename) {
std::cerr << "\nerror: filename is empty for compileFileToCUBIN()!\n";
exit(1);
}
std::ifstream inputFile(filename, std::ifstream inputFile(filename,
std::ios::in | std::ios::binary | std::ios::ate); std::ios::in | std::ios::binary | std::ios::ate);
@@ -116,12 +111,7 @@ void compileFileToCUBIN(char *filename, int argc, char **argv, char **cubinResul
compileOptions = "--include-path="; compileOptions = "--include-path=";
char *strPath = sdkFindFilePath(HeaderNames, argv[0]); std::string path = sdkFindFilePath(HeaderNames, argv[0]);
if (!strPath) {
std::cerr << "\nerror: header file " << HeaderNames << " not found!\n";
exit(1);
}
std::string path = strPath;
if (!path.empty()) { if (!path.empty()) {
std::size_t found = path.find(HeaderNames); std::size_t found = path.find(HeaderNames);
path.erase(found); path.erase(found);
@@ -130,7 +120,6 @@ void compileFileToCUBIN(char *filename, int argc, char **argv, char **cubinResul
"\nCooperativeGroups headers not found, please install it in %s " "\nCooperativeGroups headers not found, please install it in %s "
"sample directory..\n Exiting..\n", "sample directory..\n Exiting..\n",
argv[0]); argv[0]);
exit(1);
} }
compileOptions += path.c_str(); compileOptions += path.c_str();
compileParams[numCompileOptions] = reinterpret_cast<char *>( compileParams[numCompileOptions] = reinterpret_cast<char *>(

View File

@@ -1,12 +1,15 @@
# CUDA Samples # CUDA Samples
Samples for CUDA Developers which demonstrates features in CUDA Toolkit. This version supports [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads). Samples for CUDA Developers which demonstrates features in CUDA Toolkit. This version supports [CUDA Toolkit 11.6](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 12.2 ### CUDA 11.6
* Added new folder structure for samples
* Added support of Visual Studio 2022 to all samples supported on [Windows](#windows-1).
* All CUDA samples are now only available on [GitHub](https://github.com/nvidia/cuda-samples). They are no longer available via CUDA toolkit.
### [older versions...](./CHANGELOG.md) ### [older versions...](./CHANGELOG.md)
@@ -14,7 +17,7 @@ This section describes the release notes for the CUDA Samples on GitHub only.
### Prerequisites ### Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](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
@@ -89,9 +92,6 @@ Samples that are specific to domain (Graphics, Finance, Image Processing).
### [6. Performance](./Samples/6_Performance/README.md) ### [6. Performance](./Samples/6_Performance/README.md)
Samples that demonstrate performance optimization. Samples that demonstrate performance optimization.
### [7. libNVVM](./Samples/7_libNVVM/README.md)
Samples that demonstrate the use of libNVVVM and NVVM IR.
## Dependencies ## Dependencies
Some CUDA Samples rely on third-party applications and/or libraries, or features provided by the CUDA Toolkit and Driver, to either build or execute. These dependencies are listed below. Some CUDA Samples rely on third-party applications and/or libraries, or features provided by the CUDA Toolkit and Driver, to either build or execute. These dependencies are listed below.
@@ -246,10 +246,6 @@ FP16 is a 16-bit floating-point format. One bit is used for the sign, five bits
NVCC support of [C++11 features](https://en.wikipedia.org/wiki/C++11). NVCC support of [C++11 features](https://en.wikipedia.org/wiki/C++11).
#### CMake
The libNVVM samples are built using [CMake](https://cmake.org/) 3.10 or later.
## Contributors Guide ## Contributors Guide
We welcome your input on issues and suggestions for samples. At this time we are not accepting contributions from the public, check back here as we evolve our contribution model. We welcome your input on issues and suggestions for samples. At this time we are not accepting contributions from the public, check back here as we evolve our contribution model.

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -322,9 +318,9 @@ endif
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 50 52 60 61 70 75 80 86 89 90 SMS ?= 35 37 50 52 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

View File

@@ -6,11 +6,11 @@
<toolkit>cudaStreamDestroy</toolkit> <toolkit>cudaStreamDestroy</toolkit>
<toolkit>cudaFree</toolkit> <toolkit>cudaFree</toolkit>
<toolkit>cudaMallocManaged</toolkit> <toolkit>cudaMallocManaged</toolkit>
<toolkit>cudaStreamCreate</toolkit>
<toolkit>cudaDeviceSynchronize</toolkit>
<toolkit>cudaStreamAttachMemAsync</toolkit> <toolkit>cudaStreamAttachMemAsync</toolkit>
<toolkit>cudaSetDevice</toolkit> <toolkit>cudaSetDevice</toolkit>
<toolkit>cudaDeviceSynchronize</toolkit>
<toolkit>cudaStreamSynchronize</toolkit> <toolkit>cudaStreamSynchronize</toolkit>
<toolkit>cudaStreamCreate</toolkit>
<toolkit>cudaGetDeviceProperties</toolkit> <toolkit>cudaGetDeviceProperties</toolkit>
</cuda_api_list> </cuda_api_list>
<description><![CDATA[This sample demonstrates the use of OpenMP and streams with Unified Memory on a single GPU.]]></description> <description><![CDATA[This sample demonstrates the use of OpenMP and streams with Unified Memory on a single GPU.]]></description>
@@ -57,6 +57,19 @@
<scope>1:CUDA Systems Integration</scope> <scope>1:CUDA Systems Integration</scope>
<scope>1:Unified Memory</scope> <scope>1:Unified Memory</scope>
</scopes> </scopes>
<sm-arch>sm35</sm-arch>
<sm-arch>sm37</sm-arch>
<sm-arch>sm50</sm-arch>
<sm-arch>sm52</sm-arch>
<sm-arch>sm53</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>
<sm-arch>sm87</sm-arch>
<supported_envs> <supported_envs>
<env> <env>
<arch>x86_64</arch> <arch>x86_64</arch>

View File

@@ -10,6 +10,8 @@ CUDA Systems Integration, OpenMP, CUBLAS, Multithreading, Unified Memory, CUDA S
## Supported SM Architectures ## 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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
Linux, Windows Linux, Windows
@@ -21,14 +23,14 @@ x86_64, ppc64le, armv7l
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaStreamDestroy, cudaFree, cudaMallocManaged, cudaStreamAttachMemAsync, cudaSetDevice, cudaDeviceSynchronize, cudaStreamSynchronize, cudaStreamCreate, cudaGetDeviceProperties cudaStreamDestroy, cudaFree, cudaMallocManaged, cudaStreamCreate, cudaDeviceSynchronize, cudaStreamAttachMemAsync, cudaSetDevice, cudaStreamSynchronize, cudaGetDeviceProperties
## Dependencies needed to build/run ## Dependencies needed to build/run
[OpenMP](../../../README.md#openmp), [UVM](../../../README.md#uvm), [CUBLAS](../../../README.md#cublas) [OpenMP](../../README.md#openmp), [UVM](../../README.md#uvm), [CUBLAS](../../README.md#cublas)
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](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

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/UnifiedMemoryStreams.exe</OutputFile> <OutputFile>$(OutDir)/UnifiedMemoryStreams.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/UnifiedMemoryStreams.exe</OutputFile> <OutputFile>$(OutDir)/UnifiedMemoryStreams.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/UnifiedMemoryStreams.exe</OutputFile> <OutputFile>$(OutDir)/UnifiedMemoryStreams.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -283,9 +279,9 @@ LIBRARIES :=
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 50 52 60 61 70 75 80 86 89 90 SMS ?= 35 37 50 52 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

View File

@@ -3,21 +3,21 @@
<entry> <entry>
<name>asyncAPI</name> <name>asyncAPI</name>
<cuda_api_list> <cuda_api_list>
<toolkit>cudaProfilerStop</toolkit> <toolkit>cudaMemset</toolkit>
<toolkit>cudaMalloc</toolkit>
<toolkit>cudaMemcpyAsync</toolkit>
<toolkit>cudaFree</toolkit> <toolkit>cudaFree</toolkit>
<toolkit>cudaEventRecord</toolkit>
<toolkit>cudaMallocHost</toolkit> <toolkit>cudaMallocHost</toolkit>
<toolkit>cudaProfilerStart</toolkit> <toolkit>cudaProfilerStart</toolkit>
<toolkit>cudaDeviceSynchronize</toolkit>
<toolkit>cudaEventRecord</toolkit>
<toolkit>cudaFreeHost</toolkit>
<toolkit>cudaMemset</toolkit>
<toolkit>cudaEventDestroy</toolkit>
<toolkit>cudaEventQuery</toolkit>
<toolkit>cudaEventElapsedTime</toolkit>
<toolkit>cudaGetDeviceProperties</toolkit>
<toolkit>cudaEventCreate</toolkit> <toolkit>cudaEventCreate</toolkit>
<toolkit>cudaEventElapsedTime</toolkit>
<toolkit>cudaDeviceSynchronize</toolkit>
<toolkit>cudaFreeHost</toolkit>
<toolkit>cudaMalloc</toolkit>
<toolkit>cudaEventQuery</toolkit>
<toolkit>cudaProfilerStop</toolkit>
<toolkit>cudaEventDestroy</toolkit>
<toolkit>cudaMemcpyAsync</toolkit>
<toolkit>cudaGetDeviceProperties</toolkit>
</cuda_api_list> </cuda_api_list>
<description><![CDATA[This sample illustrates the usage of CUDA events for both GPU timing and overlapping CPU and GPU execution. Events are inserted into a stream of CUDA calls. Since CUDA stream calls are asynchronous, the CPU can perform computations while GPU is executing (including DMA memcopies between the host and device). CPU can query CUDA events to determine whether GPU has completed tasks.]]></description> <description><![CDATA[This sample illustrates the usage of CUDA events for both GPU timing and overlapping CPU and GPU execution. Events are inserted into a stream of CUDA calls. Since CUDA stream calls are asynchronous, the CPU can perform computations while GPU is executing (including DMA memcopies between the host and device). CPU can query CUDA events to determine whether GPU has completed tasks.]]></description>
<devicecompilation>whole</devicecompilation> <devicecompilation>whole</devicecompilation>
@@ -46,6 +46,8 @@
<scope>1:CUDA Basic Topics</scope> <scope>1:CUDA Basic Topics</scope>
<scope>1:Performance Strategies</scope> <scope>1:Performance Strategies</scope>
</scopes> </scopes>
<sm-arch>sm35</sm-arch>
<sm-arch>sm37</sm-arch>
<sm-arch>sm50</sm-arch> <sm-arch>sm50</sm-arch>
<sm-arch>sm52</sm-arch> <sm-arch>sm52</sm-arch>
<sm-arch>sm53</sm-arch> <sm-arch>sm53</sm-arch>
@@ -57,8 +59,6 @@
<sm-arch>sm80</sm-arch> <sm-arch>sm80</sm-arch>
<sm-arch>sm86</sm-arch> <sm-arch>sm86</sm-arch>
<sm-arch>sm87</sm-arch> <sm-arch>sm87</sm-arch>
<sm-arch>sm89</sm-arch>
<sm-arch>sm90</sm-arch>
<supported_envs> <supported_envs>
<env> <env>
<arch>x86_64</arch> <arch>x86_64</arch>

View File

@@ -10,7 +10,7 @@ Asynchronous Data Transfers, CUDA Streams and Events
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,11 +23,11 @@ x86_64, ppc64le, armv7l
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaProfilerStop, cudaMalloc, cudaMemcpyAsync, cudaFree, cudaMallocHost, cudaProfilerStart, cudaDeviceSynchronize, cudaEventRecord, cudaFreeHost, cudaMemset, cudaEventDestroy, cudaEventQuery, cudaEventElapsedTime, cudaGetDeviceProperties, cudaEventCreate cudaMemset, cudaFree, cudaEventRecord, cudaMallocHost, cudaProfilerStart, cudaEventCreate, cudaEventElapsedTime, cudaDeviceSynchronize, cudaFreeHost, cudaMalloc, cudaEventQuery, cudaProfilerStop, cudaEventDestroy, cudaMemcpyAsync, cudaGetDeviceProperties
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
## Build and Run ## Build and Run

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/asyncAPI.exe</OutputFile> <OutputFile>$(OutDir)/asyncAPI.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/asyncAPI.exe</OutputFile> <OutputFile>$(OutDir)/asyncAPI.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/asyncAPI.exe</OutputFile> <OutputFile>$(OutDir)/asyncAPI.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -303,23 +299,20 @@ ifeq ($(TARGET_OS),linux)
#$(warning $(GCCVERSION)) #$(warning $(GCCVERSION))
IS_MIN_VERSION := $(shell expr `echo $(GCCVERSION)` \>= 47000) IS_MIN_VERSION := $(shell expr `echo $(GCCVERSION)` \>= 47000)
ifneq ($(CUSTOM_HOST_COMPILER), 1)
ifeq ($(IS_MIN_VERSION), 1) ifeq ($(IS_MIN_VERSION), 1)
$(info >>> GCC Version is greater or equal to 4.7.0 <<<) $(info >>> GCC Version is greater or equal to 4.7.0 <<<)
else else
$(info >>> Waiving build. Minimum GCC version required is 4.7.0<<<) $(info >>> Waiving build. Minimum GCC version required is 4.7.0<<<)
SAMPLE_ENABLED := 0 SAMPLE_ENABLED := 0
endif endif
else
$(warning >>> Custom HOST_COMPILER set; skipping GCC version check. This may lead to unintended behavior. Please note the minimum equivalent GCC version is 4.7.0 <<<)
endif
endif endif
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 50 52 60 61 70 75 80 86 89 90 SMS ?= 35 37 50 52 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)
@@ -370,6 +363,7 @@ run: build
$(EXEC) ./c++11_cuda $(EXEC) ./c++11_cuda
testrun: build testrun: build
$(EXEC) ./c++11_cuda --dummy-test-param
clean: clean:
rm -f c++11_cuda c++11_cuda.o rm -f c++11_cuda c++11_cuda.o

View File

@@ -7,9 +7,9 @@
</cflags> </cflags>
<cuda_api_list> <cuda_api_list>
<toolkit>cudaMalloc</toolkit> <toolkit>cudaMalloc</toolkit>
<toolkit>cudaMemcpy</toolkit>
<toolkit>cudaMemset</toolkit> <toolkit>cudaMemset</toolkit>
<toolkit>cudaFree</toolkit> <toolkit>cudaFree</toolkit>
<toolkit>cudaMemcpy</toolkit>
</cuda_api_list> </cuda_api_list>
<description><![CDATA[This sample demonstrates C++11 feature support in CUDA. It scans a input text file and prints no. of occurrences of x, y, z, w characters. ]]></description> <description><![CDATA[This sample demonstrates C++11 feature support in CUDA. It scans a input text file and prints no. of occurrences of x, y, z, w characters. ]]></description>
<devicecompilation>whole</devicecompilation> <devicecompilation>whole</devicecompilation>
@@ -31,6 +31,9 @@
</librarypaths> </librarypaths>
<nsight_eclipse>true</nsight_eclipse> <nsight_eclipse>true</nsight_eclipse>
<primary_file>c++11_cuda.cu</primary_file> <primary_file>c++11_cuda.cu</primary_file>
<qatests>
<qatest>--dummy-test-param</qatest>
</qatests>
<required_dependencies> <required_dependencies>
<dependency>CPP11</dependency> <dependency>CPP11</dependency>
</required_dependencies> </required_dependencies>
@@ -38,6 +41,8 @@
<scope>1:CUDA Advanced Topics</scope> <scope>1:CUDA Advanced Topics</scope>
<scope>1:C++11 CUDA</scope> <scope>1:C++11 CUDA</scope>
</scopes> </scopes>
<sm-arch>sm35</sm-arch>
<sm-arch>sm37</sm-arch>
<sm-arch>sm50</sm-arch> <sm-arch>sm50</sm-arch>
<sm-arch>sm52</sm-arch> <sm-arch>sm52</sm-arch>
<sm-arch>sm53</sm-arch> <sm-arch>sm53</sm-arch>
@@ -49,8 +54,6 @@
<sm-arch>sm80</sm-arch> <sm-arch>sm80</sm-arch>
<sm-arch>sm86</sm-arch> <sm-arch>sm86</sm-arch>
<sm-arch>sm87</sm-arch> <sm-arch>sm87</sm-arch>
<sm-arch>sm89</sm-arch>
<sm-arch>sm90</sm-arch>
<supported_envs> <supported_envs>
<env> <env>
<arch>x86_64</arch> <arch>x86_64</arch>

View File

@@ -10,7 +10,7 @@ CPP11 CUDA
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,14 +23,14 @@ x86_64, ppc64le, armv7l
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaMalloc, cudaMemcpy, cudaMemset, cudaFree cudaMalloc, cudaMemset, cudaFree, cudaMemcpy
## Dependencies needed to build/run ## Dependencies needed to build/run
[CPP11](../../../README.md#cpp11) [CPP11](../../README.md#cpp11)
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](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

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/c++11_cuda.exe</OutputFile> <OutputFile>$(OutDir)/c++11_cuda.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/c++11_cuda.exe</OutputFile> <OutputFile>$(OutDir)/c++11_cuda.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/c++11_cuda.exe</OutputFile> <OutputFile>$(OutDir)/c++11_cuda.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -283,9 +279,9 @@ LIBRARIES :=
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 50 52 60 61 70 75 80 86 89 90 SMS ?= 35 37 50 52 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

View File

@@ -4,8 +4,8 @@
<name>clock</name> <name>clock</name>
<cuda_api_list> <cuda_api_list>
<toolkit>cudaMalloc</toolkit> <toolkit>cudaMalloc</toolkit>
<toolkit>cudaMemcpy</toolkit>
<toolkit>cudaFree</toolkit> <toolkit>cudaFree</toolkit>
<toolkit>cudaMemcpy</toolkit>
</cuda_api_list> </cuda_api_list>
<description><![CDATA[This example shows how to use the clock function to measure the performance of block of threads of a kernel accurately.]]></description> <description><![CDATA[This example shows how to use the clock function to measure the performance of block of threads of a kernel accurately.]]></description>
<devicecompilation>whole</devicecompilation> <devicecompilation>whole</devicecompilation>
@@ -34,6 +34,8 @@
<scope>1:CUDA Basic Topics</scope> <scope>1:CUDA Basic Topics</scope>
<scope>1:Performance Strategies</scope> <scope>1:Performance Strategies</scope>
</scopes> </scopes>
<sm-arch>sm35</sm-arch>
<sm-arch>sm37</sm-arch>
<sm-arch>sm50</sm-arch> <sm-arch>sm50</sm-arch>
<sm-arch>sm52</sm-arch> <sm-arch>sm52</sm-arch>
<sm-arch>sm53</sm-arch> <sm-arch>sm53</sm-arch>
@@ -45,8 +47,6 @@
<sm-arch>sm80</sm-arch> <sm-arch>sm80</sm-arch>
<sm-arch>sm86</sm-arch> <sm-arch>sm86</sm-arch>
<sm-arch>sm87</sm-arch> <sm-arch>sm87</sm-arch>
<sm-arch>sm89</sm-arch>
<sm-arch>sm90</sm-arch>
<supported_envs> <supported_envs>
<env> <env>
<arch>x86_64</arch> <arch>x86_64</arch>

View File

@@ -10,7 +10,7 @@ Performance Strategies
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,11 +23,11 @@ x86_64, ppc64le, armv7l
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaMalloc, cudaMemcpy, cudaFree cudaMalloc, cudaFree, cudaMemcpy
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
## Build and Run ## Build and Run

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/clock.exe</OutputFile> <OutputFile>$(OutDir)/clock.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/clock.exe</OutputFile> <OutputFile>$(OutDir)/clock.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/clock.exe</OutputFile> <OutputFile>$(OutDir)/clock.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++

View File

@@ -10,7 +10,7 @@ Performance Strategies, Runtime Compilation
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,17 +23,17 @@ x86_64, ppc64le, aarch64
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Driver API](http://docs.nvidia.com/cuda/cuda-driver-api/index.html) ### [CUDA Driver API](http://docs.nvidia.com/cuda/cuda-driver-api/index.html)
cuMemcpyDtoH, cuLaunchKernel, cuMemcpyHtoD, cuCtxSynchronize, cuMemAlloc, cuMemFree, cuModuleGetFunction cuModuleGetFunction, cuMemAlloc, cuLaunchKernel, cuCtxSynchronize, cuMemFree, cuMemcpyDtoH, cuMemcpyHtoD
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaBlockSize, cudaGridSize cudaBlockSize, cudaGridSize
## Dependencies needed to build/run ## Dependencies needed to build/run
[NVRTC](../../../README.md#nvrtc) [NVRTC](../../README.md#nvrtc)
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](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

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -283,9 +279,9 @@ LIBRARIES :=
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 50 52 60 61 70 75 80 86 89 90 SMS ?= 35 37 50 52 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

View File

@@ -3,22 +3,22 @@
<entry> <entry>
<name>concurrentKernels</name> <name>concurrentKernels</name>
<cuda_api_list> <cuda_api_list>
<toolkit>cudaStreamDestroy</toolkit>
<toolkit>cudaMalloc</toolkit>
<toolkit>cudaMemcpyAsync</toolkit>
<toolkit>cudaFree</toolkit>
<toolkit>cudaMallocHost</toolkit>
<toolkit>cudaEventCreateWithFlags</toolkit>
<toolkit>cudaEventSynchronize</toolkit>
<toolkit>cudaEventRecord</toolkit>
<toolkit>cudaFreeHost</toolkit>
<toolkit>cudaGetDevice</toolkit>
<toolkit>cudaStreamWaitEvent</toolkit> <toolkit>cudaStreamWaitEvent</toolkit>
<toolkit>cudaEventDestroy</toolkit> <toolkit>cudaStreamDestroy</toolkit>
<toolkit>cudaEventElapsedTime</toolkit> <toolkit>cudaFree</toolkit>
<toolkit>cudaEventRecord</toolkit>
<toolkit>cudaMallocHost</toolkit>
<toolkit>cudaStreamCreate</toolkit> <toolkit>cudaStreamCreate</toolkit>
<toolkit>cudaGetDeviceProperties</toolkit>
<toolkit>cudaEventCreate</toolkit> <toolkit>cudaEventCreate</toolkit>
<toolkit>cudaEventElapsedTime</toolkit>
<toolkit>cudaEventSynchronize</toolkit>
<toolkit>cudaFreeHost</toolkit>
<toolkit>cudaMalloc</toolkit>
<toolkit>cudaEventCreateWithFlags</toolkit>
<toolkit>cudaEventDestroy</toolkit>
<toolkit>cudaMemcpyAsync</toolkit>
<toolkit>cudaGetDeviceProperties</toolkit>
<toolkit>cudaGetDevice</toolkit>
</cuda_api_list> </cuda_api_list>
<description><![CDATA[This sample demonstrates the use of CUDA streams for concurrent execution of several kernels on GPU device. It also illustrates how to introduce dependencies between CUDA streams with the new cudaStreamWaitEvent function.]]></description> <description><![CDATA[This sample demonstrates the use of CUDA streams for concurrent execution of several kernels on GPU device. It also illustrates how to introduce dependencies between CUDA streams with the new cudaStreamWaitEvent function.]]></description>
<devicecompilation>whole</devicecompilation> <devicecompilation>whole</devicecompilation>
@@ -44,6 +44,8 @@
<scope>1:CUDA Advanced Topics</scope> <scope>1:CUDA Advanced Topics</scope>
<scope>1:Performance Strategies</scope> <scope>1:Performance Strategies</scope>
</scopes> </scopes>
<sm-arch>sm35</sm-arch>
<sm-arch>sm37</sm-arch>
<sm-arch>sm50</sm-arch> <sm-arch>sm50</sm-arch>
<sm-arch>sm52</sm-arch> <sm-arch>sm52</sm-arch>
<sm-arch>sm53</sm-arch> <sm-arch>sm53</sm-arch>
@@ -55,8 +57,6 @@
<sm-arch>sm80</sm-arch> <sm-arch>sm80</sm-arch>
<sm-arch>sm86</sm-arch> <sm-arch>sm86</sm-arch>
<sm-arch>sm87</sm-arch> <sm-arch>sm87</sm-arch>
<sm-arch>sm89</sm-arch>
<sm-arch>sm90</sm-arch>
<supported_envs> <supported_envs>
<env> <env>
<arch>x86_64</arch> <arch>x86_64</arch>

View File

@@ -10,7 +10,7 @@ Performance Strategies
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,11 +23,11 @@ x86_64, ppc64le, armv7l
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaStreamDestroy, cudaMalloc, cudaMemcpyAsync, cudaFree, cudaMallocHost, cudaEventCreateWithFlags, cudaEventSynchronize, cudaEventRecord, cudaFreeHost, cudaGetDevice, cudaStreamWaitEvent, cudaEventDestroy, cudaEventElapsedTime, cudaStreamCreate, cudaGetDeviceProperties, cudaEventCreate cudaStreamWaitEvent, cudaStreamDestroy, cudaFree, cudaEventRecord, cudaMallocHost, cudaStreamCreate, cudaEventCreate, cudaEventElapsedTime, cudaEventSynchronize, cudaFreeHost, cudaMalloc, cudaEventCreateWithFlags, cudaEventDestroy, cudaMemcpyAsync, cudaGetDeviceProperties, cudaGetDevice
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
## Build and Run ## Build and Run

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/concurrentKernels.exe</OutputFile> <OutputFile>$(OutDir)/concurrentKernels.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/concurrentKernels.exe</OutputFile> <OutputFile>$(OutDir)/concurrentKernels.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/concurrentKernels.exe</OutputFile> <OutputFile>$(OutDir)/concurrentKernels.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -283,9 +279,9 @@ LIBRARIES :=
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 50 52 60 61 70 75 80 86 89 90 SMS ?= 35 37 50 52 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

View File

@@ -4,8 +4,8 @@
<name>cppIntegration</name> <name>cppIntegration</name>
<cuda_api_list> <cuda_api_list>
<toolkit>cudaMalloc</toolkit> <toolkit>cudaMalloc</toolkit>
<toolkit>cudaMemcpy</toolkit>
<toolkit>cudaFree</toolkit> <toolkit>cudaFree</toolkit>
<toolkit>cudaMemcpy</toolkit>
</cuda_api_list> </cuda_api_list>
<description><![CDATA[This example demonstrates how to integrate CUDA into an existing C++ application, i.e. the CUDA entry point on host side is only a function which is called from C++ code and only the file containing this function is compiled with nvcc. It also demonstrates that vector types can be used from cpp.]]></description> <description><![CDATA[This example demonstrates how to integrate CUDA into an existing C++ application, i.e. the CUDA entry point on host side is only a function which is called from C++ code and only the file containing this function is compiled with nvcc. It also demonstrates that vector types can be used from cpp.]]></description>
<devicecompilation>whole</devicecompilation> <devicecompilation>whole</devicecompilation>
@@ -28,6 +28,8 @@
<scopes> <scopes>
<scope>1:CUDA Basic Topics</scope> <scope>1:CUDA Basic Topics</scope>
</scopes> </scopes>
<sm-arch>sm35</sm-arch>
<sm-arch>sm37</sm-arch>
<sm-arch>sm50</sm-arch> <sm-arch>sm50</sm-arch>
<sm-arch>sm52</sm-arch> <sm-arch>sm52</sm-arch>
<sm-arch>sm53</sm-arch> <sm-arch>sm53</sm-arch>
@@ -39,8 +41,6 @@
<sm-arch>sm80</sm-arch> <sm-arch>sm80</sm-arch>
<sm-arch>sm86</sm-arch> <sm-arch>sm86</sm-arch>
<sm-arch>sm87</sm-arch> <sm-arch>sm87</sm-arch>
<sm-arch>sm89</sm-arch>
<sm-arch>sm90</sm-arch>
<supported_envs> <supported_envs>
<env> <env>
<arch>x86_64</arch> <arch>x86_64</arch>

View File

@@ -10,7 +10,7 @@ CPP-CUDA Integration
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,11 +23,11 @@ x86_64, ppc64le, armv7l
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaMalloc, cudaMemcpy, cudaFree cudaMalloc, cudaFree, cudaMemcpy
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
## Build and Run ## Build and Run

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/cppIntegration.exe</OutputFile> <OutputFile>$(OutDir)/cppIntegration.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/cppIntegration.exe</OutputFile> <OutputFile>$(OutDir)/cppIntegration.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -105,6 +105,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/cppIntegration.exe</OutputFile> <OutputFile>$(OutDir)/cppIntegration.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -105,6 +105,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -283,9 +279,9 @@ LIBRARIES :=
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 50 52 60 61 70 75 80 86 89 90 SMS ?= 35 37 50 52 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

View File

@@ -3,17 +3,17 @@
<entry> <entry>
<name>cppOverload</name> <name>cppOverload</name>
<cuda_api_list> <cuda_api_list>
<toolkit>cudaMemcpy</toolkit>
<toolkit>cudaFuncSetCacheConfig</toolkit>
<toolkit>cudaFree</toolkit> <toolkit>cudaFree</toolkit>
<toolkit>cudaFuncSetCacheConfig</toolkit>
<toolkit>cudaMallocHost</toolkit> <toolkit>cudaMallocHost</toolkit>
<toolkit>cudaSetDevice</toolkit> <toolkit>cudaFuncGetAttributes</toolkit>
<toolkit>cudaGetDeviceProperties</toolkit> <toolkit>cudaGetDeviceCount</toolkit>
<toolkit>cudaDeviceSynchronize</toolkit> <toolkit>cudaDeviceSynchronize</toolkit>
<toolkit>cudaFreeHost</toolkit> <toolkit>cudaFreeHost</toolkit>
<toolkit>cudaMalloc</toolkit> <toolkit>cudaMalloc</toolkit>
<toolkit>cudaFuncGetAttributes</toolkit> <toolkit>cudaSetDevice</toolkit>
<toolkit>cudaGetDeviceCount</toolkit> <toolkit>cudaMemcpy</toolkit>
<toolkit>cudaGetDeviceProperties</toolkit>
</cuda_api_list> </cuda_api_list>
<description><![CDATA[This sample demonstrates how to use C++ function overloading on the GPU.]]></description> <description><![CDATA[This sample demonstrates how to use C++ function overloading on the GPU.]]></description>
<devicecompilation>whole</devicecompilation> <devicecompilation>whole</devicecompilation>
@@ -39,6 +39,8 @@
<scope>1:CUDA Basic Topics</scope> <scope>1:CUDA Basic Topics</scope>
<scope>1:Performance Strategies</scope> <scope>1:Performance Strategies</scope>
</scopes> </scopes>
<sm-arch>sm35</sm-arch>
<sm-arch>sm37</sm-arch>
<sm-arch>sm50</sm-arch> <sm-arch>sm50</sm-arch>
<sm-arch>sm52</sm-arch> <sm-arch>sm52</sm-arch>
<sm-arch>sm53</sm-arch> <sm-arch>sm53</sm-arch>
@@ -50,8 +52,6 @@
<sm-arch>sm80</sm-arch> <sm-arch>sm80</sm-arch>
<sm-arch>sm86</sm-arch> <sm-arch>sm86</sm-arch>
<sm-arch>sm87</sm-arch> <sm-arch>sm87</sm-arch>
<sm-arch>sm89</sm-arch>
<sm-arch>sm90</sm-arch>
<supported_envs> <supported_envs>
<env> <env>
<arch>x86_64</arch> <arch>x86_64</arch>

View File

@@ -10,7 +10,7 @@ C++ Function Overloading, CUDA Streams and Events
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,11 +23,11 @@ x86_64, ppc64le, armv7l
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaMemcpy, cudaFuncSetCacheConfig, cudaFree, cudaMallocHost, cudaSetDevice, cudaGetDeviceProperties, cudaDeviceSynchronize, cudaFreeHost, cudaMalloc, cudaFuncGetAttributes, cudaGetDeviceCount cudaFree, cudaFuncSetCacheConfig, cudaMallocHost, cudaFuncGetAttributes, cudaGetDeviceCount, cudaDeviceSynchronize, cudaFreeHost, cudaMalloc, cudaSetDevice, cudaMemcpy, cudaGetDeviceProperties
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
## Build and Run ## Build and Run

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/cppOverload.exe</OutputFile> <OutputFile>$(OutDir)/cppOverload.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/cppOverload.exe</OutputFile> <OutputFile>$(OutDir)/cppOverload.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/cppOverload.exe</OutputFile> <OutputFile>$(OutDir)/cppOverload.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -325,9 +321,9 @@ endif
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 50 52 60 61 70 75 80 86 89 90 SMS ?= 35 37 50 52 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

View File

@@ -10,7 +10,7 @@ CUDA Systems Integration, OpenMP, Multithreading
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,14 +23,14 @@ x86_64, ppc64le, armv7l
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaMemcpy, cudaGetErrorString, cudaFree, cudaGetLastError, cudaSetDevice, cudaGetDeviceCount, cudaGetDevice, cudaMemset, cudaMalloc, cudaGetDeviceProperties cudaMemset, cudaFree, cudaGetDeviceCount, cudaSetDevice, cudaMalloc, cudaGetLastError, cudaMemcpy, cudaGetErrorString, cudaGetDeviceProperties, cudaGetDevice
## Dependencies needed to build/run ## Dependencies needed to build/run
[OpenMP](../../../README.md#openmp) [OpenMP](../../README.md#openmp)
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](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

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/cudaOpenMP.exe</OutputFile> <OutputFile>$(OutDir)/cudaOpenMP.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/cudaOpenMP.exe</OutputFile> <OutputFile>$(OutDir)/cudaOpenMP.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/cudaOpenMP.exe</OutputFile> <OutputFile>$(OutDir)/cudaOpenMP.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -289,9 +285,9 @@ LIBRARIES :=
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 60 61 70 75 80 86 89 90 SMS ?= 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

View File

@@ -3,11 +3,11 @@
<entry> <entry>
<name>fp16ScalarProduct</name> <name>fp16ScalarProduct</name>
<cuda_api_list> <cuda_api_list>
<toolkit>cudaMemcpy</toolkit>
<toolkit>cudaFree</toolkit> <toolkit>cudaFree</toolkit>
<toolkit>cudaMallocHost</toolkit> <toolkit>cudaMallocHost</toolkit>
<toolkit>cudaFreeHost</toolkit> <toolkit>cudaFreeHost</toolkit>
<toolkit>cudaMalloc</toolkit> <toolkit>cudaMalloc</toolkit>
<toolkit>cudaMemcpy</toolkit>
<toolkit>cudaGetDeviceProperties</toolkit> <toolkit>cudaGetDeviceProperties</toolkit>
</cuda_api_list> </cuda_api_list>
<description><![CDATA[Calculates scalar product of two vectors of FP16 numbers.]]></description> <description><![CDATA[Calculates scalar product of two vectors of FP16 numbers.]]></description>
@@ -44,8 +44,6 @@
<sm-arch>sm80</sm-arch> <sm-arch>sm80</sm-arch>
<sm-arch>sm86</sm-arch> <sm-arch>sm86</sm-arch>
<sm-arch>sm87</sm-arch> <sm-arch>sm87</sm-arch>
<sm-arch>sm89</sm-arch>
<sm-arch>sm90</sm-arch>
<supported_envs> <supported_envs>
<env> <env>
<arch>arm</arch> <arch>arm</arch>

View File

@@ -10,7 +10,7 @@ CUDA Runtime API
## Supported SM Architectures ## Supported SM Architectures
[SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,14 +23,14 @@ x86_64, ppc64le, armv7l
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaMemcpy, cudaFree, cudaMallocHost, cudaFreeHost, cudaMalloc, cudaGetDeviceProperties cudaFree, cudaMallocHost, cudaFreeHost, cudaMalloc, cudaMemcpy, cudaGetDeviceProperties
## Dependencies needed to build/run ## Dependencies needed to build/run
[FP16](../../../README.md#fp16) [FP16](../../README.md#fp16)
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](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

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/fp16ScalarProduct.exe</OutputFile> <OutputFile>$(OutDir)/fp16ScalarProduct.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <CodeGeneration>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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/fp16ScalarProduct.exe</OutputFile> <OutputFile>$(OutDir)/fp16ScalarProduct.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <CodeGeneration>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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/fp16ScalarProduct.exe</OutputFile> <OutputFile>$(OutDir)/fp16ScalarProduct.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <CodeGeneration>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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -283,9 +279,9 @@ LIBRARIES :=
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 50 52 60 61 70 75 80 86 89 90 SMS ?= 35 37 50 52 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

View File

@@ -3,20 +3,20 @@
<entry> <entry>
<name>matrixMul</name> <name>matrixMul</name>
<cuda_api_list> <cuda_api_list>
<toolkit>cudaStreamCreateWithFlags</toolkit>
<toolkit>cudaProfilerStop</toolkit>
<toolkit>cudaMalloc</toolkit>
<toolkit>cudaFree</toolkit> <toolkit>cudaFree</toolkit>
<toolkit>cudaEventRecord</toolkit>
<toolkit>cudaMallocHost</toolkit> <toolkit>cudaMallocHost</toolkit>
<toolkit>cudaProfilerStart</toolkit> <toolkit>cudaProfilerStart</toolkit>
<toolkit>cudaEventSynchronize</toolkit>
<toolkit>cudaEventRecord</toolkit>
<toolkit>cudaFreeHost</toolkit>
<toolkit>cudaStreamSynchronize</toolkit>
<toolkit>cudaEventDestroy</toolkit>
<toolkit>cudaEventElapsedTime</toolkit>
<toolkit>cudaMemcpyAsync</toolkit>
<toolkit>cudaEventCreate</toolkit> <toolkit>cudaEventCreate</toolkit>
<toolkit>cudaEventElapsedTime</toolkit>
<toolkit>cudaEventSynchronize</toolkit>
<toolkit>cudaFreeHost</toolkit>
<toolkit>cudaMalloc</toolkit>
<toolkit>cudaProfilerStop</toolkit>
<toolkit>cudaStreamCreateWithFlags</toolkit>
<toolkit>cudaEventDestroy</toolkit>
<toolkit>cudaStreamSynchronize</toolkit>
<toolkit>cudaMemcpyAsync</toolkit>
</cuda_api_list> </cuda_api_list>
<description><![CDATA[This sample implements matrix multiplication and is exactly the same as Chapter 6 of the programming guide. It has been written for clarity of exposition to illustrate various CUDA programming principles, not with the goal of providing the most performant generic kernel for matrix multiplication. To illustrate GPU performance for matrix multiply, this sample also shows how to use the new CUDA 4.0 interface for CUBLAS to demonstrate high-performance performance for matrix multiplication.]]></description> <description><![CDATA[This sample implements matrix multiplication and is exactly the same as Chapter 6 of the programming guide. It has been written for clarity of exposition to illustrate various CUDA programming principles, not with the goal of providing the most performant generic kernel for matrix multiplication. To illustrate GPU performance for matrix multiply, this sample also shows how to use the new CUDA 4.0 interface for CUBLAS to demonstrate high-performance performance for matrix multiplication.]]></description>
<devicecompilation>whole</devicecompilation> <devicecompilation>whole</devicecompilation>
@@ -43,6 +43,8 @@
<scope>1:CUDA Basic Topics</scope> <scope>1:CUDA Basic Topics</scope>
<scope>3:Linear Algebra</scope> <scope>3:Linear Algebra</scope>
</scopes> </scopes>
<sm-arch>sm35</sm-arch>
<sm-arch>sm37</sm-arch>
<sm-arch>sm50</sm-arch> <sm-arch>sm50</sm-arch>
<sm-arch>sm52</sm-arch> <sm-arch>sm52</sm-arch>
<sm-arch>sm53</sm-arch> <sm-arch>sm53</sm-arch>
@@ -54,8 +56,6 @@
<sm-arch>sm80</sm-arch> <sm-arch>sm80</sm-arch>
<sm-arch>sm86</sm-arch> <sm-arch>sm86</sm-arch>
<sm-arch>sm87</sm-arch> <sm-arch>sm87</sm-arch>
<sm-arch>sm89</sm-arch>
<sm-arch>sm90</sm-arch>
<supported_envs> <supported_envs>
<env> <env>
<arch>x86_64</arch> <arch>x86_64</arch>

View File

@@ -10,7 +10,7 @@ CUDA Runtime API, Linear Algebra
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,11 +23,11 @@ x86_64, ppc64le, armv7l, aarch64
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaStreamCreateWithFlags, cudaProfilerStop, cudaMalloc, cudaFree, cudaMallocHost, cudaProfilerStart, cudaEventSynchronize, cudaEventRecord, cudaFreeHost, cudaStreamSynchronize, cudaEventDestroy, cudaEventElapsedTime, cudaMemcpyAsync, cudaEventCreate cudaFree, cudaEventRecord, cudaMallocHost, cudaProfilerStart, cudaEventCreate, cudaEventElapsedTime, cudaEventSynchronize, cudaFreeHost, cudaMalloc, cudaProfilerStop, cudaStreamCreateWithFlags, cudaEventDestroy, cudaStreamSynchronize, cudaMemcpyAsync
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
## Build and Run ## Build and Run

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/matrixMul.exe</OutputFile> <OutputFile>$(OutDir)/matrixMul.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/matrixMul.exe</OutputFile> <OutputFile>$(OutDir)/matrixMul.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/matrixMul.exe</OutputFile> <OutputFile>$(OutDir)/matrixMul.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -287,9 +283,9 @@ FATBIN_FILE := matrixMul_kernel${TARGET_SIZE}.fatbin
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 50 52 60 61 70 75 80 86 89 90 SMS ?= 35 37 50 52 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

View File

@@ -10,7 +10,7 @@ CUDA Driver API, Matrix Multiply
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,11 +23,11 @@ x86_64, ppc64le, armv7l, aarch64
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Driver API](http://docs.nvidia.com/cuda/cuda-driver-api/index.html) ### [CUDA Driver API](http://docs.nvidia.com/cuda/cuda-driver-api/index.html)
cuMemcpyDtoH, cuLaunchKernel, cuMemcpyHtoD, cuDeviceGetName, cuDeviceTotalMem, cuDeviceGetAttribute, cuModuleLoadData, cuOccupancyMaxPotentialBlockSize, cuMemAlloc, cuMemFree, cuCtxDestroy, cuModuleGetFunction, cuCtxCreate cuModuleGetFunction, cuMemcpyHtoD, cuModuleLoadData, cuCtxCreate, cuLaunchKernel, cuDeviceGetName, cuMemAlloc, cuOccupancyMaxPotentialBlockSize, cuDeviceTotalMem, cuMemFree, cuMemcpyDtoH, cuCtxDestroy, cuDeviceGetAttribute
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
## Build and Run ## Build and Run

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/matrixMulDrv.exe</OutputFile> <OutputFile>$(OutDir)/matrixMulDrv.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/matrixMulDrv.exe</OutputFile> <OutputFile>$(OutDir)/matrixMulDrv.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/matrixMulDrv.exe</OutputFile> <OutputFile>$(OutDir)/matrixMulDrv.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -291,8 +287,8 @@ ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
# Generate PTX code from SM 53 # Generate PTX code from SM 53
GENCODE_FLAGS += -gencode arch=compute_53,code=compute_53 GENCODE_FLAGS += -gencode arch=compute_53,code=compute_53
else else
# Generate PTX code from SM 50 # Generate PTX code from SM 35
GENCODE_FLAGS += -gencode arch=compute_50,code=compute_50 GENCODE_FLAGS += -gencode arch=compute_35,code=compute_35
endif endif
endif endif

View File

@@ -10,7 +10,7 @@ CUDA Driver API, CUDA Dynamically Linked Library
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,11 +23,11 @@ x86_64, ppc64le, armv7l
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Driver API](http://docs.nvidia.com/cuda/cuda-driver-api/index.html) ### [CUDA Driver API](http://docs.nvidia.com/cuda/cuda-driver-api/index.html)
cuMemcpyDtoH, cuDeviceGetName, cuParamSeti, cuModuleLoadDataEx, cuModuleGetFunction, cuLaunchGrid, cuFuncSetSharedSize, cuMemFree, cuParamSetSize, cuParamSetv, cuInit, cuMemcpyHtoD, cuLaunchKernel, cuDeviceGet, cuFuncSetBlockShape, cuCtxDestroy, cuDeviceGetCount, cuDeviceComputeCapability, cuCtxSynchronize, cuMemAlloc, cuCtxCreate cuParamSetv, cuMemFree, cuInit, cuModuleGetFunction, cuCtxDestroy, cuCtxCreate, cuDeviceGetName, cuCtxSynchronize, cuParamSeti, cuModuleLoadDataEx, cuDeviceGet, cuFuncSetSharedSize, cuMemAlloc, cuDeviceComputeCapability, cuFuncSetBlockShape, cuMemcpyHtoD, cuParamSetSize, cuLaunchGrid, cuDeviceGetCount, cuLaunchKernel, cuMemcpyDtoH
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
## Build and Run ## Build and Run

View File

@@ -95,7 +95,6 @@ inline int _ConvertSMVer2CoresDRV(int major, int minor) {
{0x80, 64}, {0x80, 64},
{0x86, 128}, {0x86, 128},
{0x87, 128}, {0x87, 128},
{0x90, 128},
{-1, -1}}; {-1, -1}};
int index = 0; int index = 0;

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/matrixMulDynlinkJIT.exe</OutputFile> <OutputFile>$(OutDir)/matrixMulDynlinkJIT.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>compute_50,compute_50;</CodeGeneration> <CodeGeneration>compute_35,compute_35;</CodeGeneration>
<AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -116,6 +116,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/matrixMulDynlinkJIT.exe</OutputFile> <OutputFile>$(OutDir)/matrixMulDynlinkJIT.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>compute_50,compute_50;</CodeGeneration> <CodeGeneration>compute_35,compute_35;</CodeGeneration>
<AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -112,6 +112,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/matrixMulDynlinkJIT.exe</OutputFile> <OutputFile>$(OutDir)/matrixMulDynlinkJIT.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>compute_50,compute_50;</CodeGeneration> <CodeGeneration>compute_35,compute_35;</CodeGeneration>
<AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -112,6 +112,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++

View File

@@ -10,7 +10,7 @@ CUDA Runtime API, Linear Algebra, Runtime Compilation
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,14 +23,14 @@ x86_64, ppc64le, aarch64
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Driver API](http://docs.nvidia.com/cuda/cuda-driver-api/index.html) ### [CUDA Driver API](http://docs.nvidia.com/cuda/cuda-driver-api/index.html)
cuMemcpyDtoH, cuLaunchKernel, cuMemcpyHtoD, cuCtxSynchronize, cuMemAlloc, cuMemFree, cuModuleGetFunction cuModuleGetFunction, cuMemAlloc, cuLaunchKernel, cuCtxSynchronize, cuMemFree, cuMemcpyDtoH, cuMemcpyHtoD
## Dependencies needed to build/run ## Dependencies needed to build/run
[NVRTC](../../../README.md#nvrtc) [NVRTC](../../README.md#nvrtc)
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](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

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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 @@ xcopy /y /e /s "$(CudaToolkitDir)include\cooperative_groups" .\cooperative_group
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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 @@ xcopy /y /e /s "$(CudaToolkitDir)include\cooperative_groups" .\cooperative_group
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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 @@ xcopy /y /e /s "$(CudaToolkitDir)include\cooperative_groups" .\cooperative_group
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -283,9 +279,9 @@ LIBRARIES :=
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 50 52 60 61 70 75 80 86 89 90 SMS ?= 35 37 50 52 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

View File

@@ -4,9 +4,9 @@
<name>mergeSort</name> <name>mergeSort</name>
<cuda_api_list> <cuda_api_list>
<toolkit>cudaMalloc</toolkit> <toolkit>cudaMalloc</toolkit>
<toolkit>cudaFree</toolkit>
<toolkit>cudaDeviceSynchronize</toolkit> <toolkit>cudaDeviceSynchronize</toolkit>
<toolkit>cudaMemcpy</toolkit> <toolkit>cudaMemcpy</toolkit>
<toolkit>cudaFree</toolkit>
</cuda_api_list> </cuda_api_list>
<description><![CDATA[This sample implements a merge sort (also known as Batcher's sort), algorithms belonging to the class of sorting networks. While generally subefficient on large sequences compared to algorithms with better asymptotic algorithmic complexity (i.e. merge sort or radix sort), may be the algorithms of choice for sorting batches of short- to mid-sized (key, value) array pairs. Refer to the excellent tutorial by H. W. Lang http://www.iti.fh-flensburg.de/lang/algorithmen/sortieren/networks/indexen.htm]]></description> <description><![CDATA[This sample implements a merge sort (also known as Batcher's sort), algorithms belonging to the class of sorting networks. While generally subefficient on large sequences compared to algorithms with better asymptotic algorithmic complexity (i.e. merge sort or radix sort), may be the algorithms of choice for sorting batches of short- to mid-sized (key, value) array pairs. Refer to the excellent tutorial by H. W. Lang http://www.iti.fh-flensburg.de/lang/algorithmen/sortieren/networks/indexen.htm]]></description>
<devicecompilation>whole</devicecompilation> <devicecompilation>whole</devicecompilation>
@@ -33,6 +33,8 @@
<scope>1:CUDA Advanced Topics</scope> <scope>1:CUDA Advanced Topics</scope>
<scope>1:Data-Parallel Algorithms</scope> <scope>1:Data-Parallel Algorithms</scope>
</scopes> </scopes>
<sm-arch>sm35</sm-arch>
<sm-arch>sm37</sm-arch>
<sm-arch>sm50</sm-arch> <sm-arch>sm50</sm-arch>
<sm-arch>sm52</sm-arch> <sm-arch>sm52</sm-arch>
<sm-arch>sm53</sm-arch> <sm-arch>sm53</sm-arch>
@@ -44,8 +46,6 @@
<sm-arch>sm80</sm-arch> <sm-arch>sm80</sm-arch>
<sm-arch>sm86</sm-arch> <sm-arch>sm86</sm-arch>
<sm-arch>sm87</sm-arch> <sm-arch>sm87</sm-arch>
<sm-arch>sm89</sm-arch>
<sm-arch>sm90</sm-arch>
<supported_envs> <supported_envs>
<env> <env>
<arch>x86_64</arch> <arch>x86_64</arch>

View File

@@ -10,7 +10,7 @@ Data-Parallel Algorithms
## Supported SM Architectures ## Supported SM Architectures
[SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [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 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,11 +23,11 @@ x86_64, ppc64le, armv7l
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaMalloc, cudaDeviceSynchronize, cudaMemcpy, cudaFree cudaMalloc, cudaFree, cudaDeviceSynchronize, cudaMemcpy
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
## Build and Run ## Build and Run

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/mergeSort.exe</OutputFile> <OutputFile>$(OutDir)/mergeSort.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/mergeSort.exe</OutputFile> <OutputFile>$(OutDir)/mergeSort.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/mergeSort.exe</OutputFile> <OutputFile>$(OutDir)/mergeSort.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>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;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <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" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -309,23 +305,20 @@ ifeq ($(TARGET_OS),linux)
#$(warning $(GCCVERSION)) #$(warning $(GCCVERSION))
IS_MIN_VERSION := $(shell expr `echo $(GCCVERSION)` \>= 51000) IS_MIN_VERSION := $(shell expr `echo $(GCCVERSION)` \>= 51000)
ifneq ($(CUSTOM_HOST_COMPILER), 1)
ifeq ($(IS_MIN_VERSION), 1) ifeq ($(IS_MIN_VERSION), 1)
$(info >>> GCC Version is greater or equal to 5.1.0 <<<) $(info >>> GCC Version is greater or equal to 5.1.0 <<<)
else else
$(info >>> Waiving build. Minimum GCC version required is 5.1.0<<<) $(info >>> Waiving build. Minimum GCC version required is 5.1.0<<<)
SAMPLE_ENABLED := 0 SAMPLE_ENABLED := 0
endif endif
else
$(warning >>> Custom HOST_COMPILER set; skipping GCC version check. This may lead to unintended behavior. Please note the minimum equivalent GCC version is 5.1.0 <<<)
endif
endif endif
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 70 72 75 80 86 87 90 SMS ?= 70 72 75 80 86 87
else else
SMS ?= 70 75 80 86 89 90 SMS ?= 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

View File

@@ -6,17 +6,17 @@
<flag>--std=c++11</flag> <flag>--std=c++11</flag>
</cflags> </cflags>
<cuda_api_list> <cuda_api_list>
<toolkit>cudaStreamCreateWithFlags</toolkit>
<toolkit>cudaFree</toolkit> <toolkit>cudaFree</toolkit>
<toolkit>cudaDeviceGetAttribute</toolkit>
<toolkit>cudaMallocHost</toolkit> <toolkit>cudaMallocHost</toolkit>
<toolkit>cudaFreeHost</toolkit>
<toolkit>cudaStreamSynchronize</toolkit>
<toolkit>cudaLaunchCooperativeKernel</toolkit>
<toolkit>cudaMalloc</toolkit>
<toolkit>cudaOccupancyMaxActiveBlocksPerMultiprocessor</toolkit> <toolkit>cudaOccupancyMaxActiveBlocksPerMultiprocessor</toolkit>
<toolkit>cudaMemcpyAsync</toolkit>
<toolkit>cudaOccupancyMaxPotentialBlockSize</toolkit> <toolkit>cudaOccupancyMaxPotentialBlockSize</toolkit>
<toolkit>cudaDeviceGetAttribute</toolkit>
<toolkit>cudaFreeHost</toolkit>
<toolkit>cudaMalloc</toolkit>
<toolkit>cudaStreamCreateWithFlags</toolkit>
<toolkit>cudaLaunchCooperativeKernel</toolkit>
<toolkit>cudaStreamSynchronize</toolkit>
<toolkit>cudaMemcpyAsync</toolkit>
</cuda_api_list> </cuda_api_list>
<description><![CDATA[A simple demonstration of arrive wait barriers.]]></description> <description><![CDATA[A simple demonstration of arrive wait barriers.]]></description>
<devicecompilation>whole</devicecompilation> <devicecompilation>whole</devicecompilation>
@@ -53,8 +53,6 @@
<sm-arch>sm80</sm-arch> <sm-arch>sm80</sm-arch>
<sm-arch>sm86</sm-arch> <sm-arch>sm86</sm-arch>
<sm-arch>sm87</sm-arch> <sm-arch>sm87</sm-arch>
<sm-arch>sm89</sm-arch>
<sm-arch>sm90</sm-arch>
<supported_envs> <supported_envs>
<env> <env>
<arch>x86_64</arch> <arch>x86_64</arch>

View File

@@ -10,7 +10,7 @@ Arrive Wait Barrier
## Supported SM Architectures ## Supported SM Architectures
[SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus) [SM 8.9 ](https://developer.nvidia.com/cuda-gpus) [SM 9.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes ## Supported OSes
@@ -23,14 +23,14 @@ x86_64, ppc64le, armv7l, aarch64
## CUDA APIs involved ## CUDA APIs involved
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html) ### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
cudaStreamCreateWithFlags, cudaFree, cudaDeviceGetAttribute, cudaMallocHost, cudaFreeHost, cudaStreamSynchronize, cudaLaunchCooperativeKernel, cudaMalloc, cudaOccupancyMaxActiveBlocksPerMultiprocessor, cudaMemcpyAsync, cudaOccupancyMaxPotentialBlockSize cudaFree, cudaMallocHost, cudaOccupancyMaxActiveBlocksPerMultiprocessor, cudaOccupancyMaxPotentialBlockSize, cudaDeviceGetAttribute, cudaFreeHost, cudaMalloc, cudaStreamCreateWithFlags, cudaLaunchCooperativeKernel, cudaStreamSynchronize, cudaMemcpyAsync
## Dependencies needed to build/run ## Dependencies needed to build/run
[CPP11](../../../README.md#cpp11), [MBCG](../../../README.md#mbcg) [CPP11](../../README.md#cpp11), [MBCG](../../README.md#mbcg)
## Prerequisites ## Prerequisites
Download and install the [CUDA Toolkit 12.2](https://developer.nvidia.com/cuda-downloads) for your corresponding platform. Download and install the [CUDA Toolkit 11.6](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

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/simpleAWBarrier.exe</OutputFile> <OutputFile>$(OutDir)/simpleAWBarrier.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
<AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -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="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/simpleAWBarrier.exe</OutputFile> <OutputFile>$(OutDir)/simpleAWBarrier.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
<AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -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="$(CUDAPropsPath)\CUDA 12.3.props" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.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" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/simpleAWBarrier.exe</OutputFile> <OutputFile>$(OutDir)/simpleAWBarrier.exe</OutputFile>
</Link> </Link>
<CudaCompile> <CudaCompile>
<CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;compute_89,sm_89;compute_90,sm_90;</CodeGeneration> <CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
<AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions> <AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
<Include>./;../../../Common</Include> <Include>./;../../../Common</Include>
<Defines>WIN32</Defines> <Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup> </ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" /> <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets"> <ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 12.3.targets" /> <Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
</ImportGroup> </ImportGroup>
</Project> </Project>

View File

@@ -113,10 +113,6 @@ ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
endif endif
# host compiler # host compiler
ifdef HOST_COMPILER
CUSTOM_HOST_COMPILER = 1
endif
ifeq ($(TARGET_OS),darwin) ifeq ($(TARGET_OS),darwin)
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1) ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
HOST_COMPILER ?= clang++ HOST_COMPILER ?= clang++
@@ -289,9 +285,9 @@ LIBRARIES :=
# Gencode arguments # Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa)) ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64 sbsa))
SMS ?= 53 61 70 72 75 80 86 87 90 SMS ?= 53 61 70 72 75 80 86 87
else else
SMS ?= 50 52 60 61 70 75 80 86 89 90 SMS ?= 35 37 50 52 60 61 70 75 80 86
endif endif
ifeq ($(SMS),) ifeq ($(SMS),)

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