Add and update samples for cuda 11.1 support

This commit is contained in:
Mahesh Doijade
2020-09-15 23:45:56 +05:30
parent e6ce58fef4
commit cd76533c3f
453 changed files with 5227 additions and 9325 deletions

View File

@@ -280,9 +280,9 @@ LIBRARIES :=
# Gencode arguments
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64))
SMS ?= 70 72 75 80
SMS ?= 70 72 75 80 86
else
SMS ?= 70 75 80
SMS ?= 70 75 80 86
endif
ifeq ($(SMS),)

View File

@@ -46,6 +46,7 @@ In addition to that, it demonstrates the use of the new CUDA function attribute
<sm-arch>sm72</sm-arch>
<sm-arch>sm75</sm-arch>
<sm-arch>sm80</sm-arch>
<sm-arch>sm86</sm-arch>
<supported_envs>
<env>
<arch>x86_64</arch>

View File

@@ -14,7 +14,7 @@ Matrix Multiply, WMMA, Tensor Cores
## 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 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus)
## Supported OSes
@@ -31,7 +31,7 @@ cudaMallocManaged, cudaDeviceSynchronize, cudaFuncSetAttribute, cudaEventCreate,
## Prerequisites
Download and install the [CUDA Toolkit 11.0](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
Download and install the [CUDA Toolkit 11.1](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
## Build and Run

View File

@@ -1,20 +0,0 @@

Microsoft Visual Studio Solution File, Format Version 12.00
# Visual Studio 2012
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "cudaTensorCoreGemm", "cudaTensorCoreGemm_vs2012.vcxproj", "{997E0757-EA74-4A4E-A0FC-47D8C8831A15}"
EndProject
Global
GlobalSection(SolutionConfigurationPlatforms) = preSolution
Debug|x64 = Debug|x64
Release|x64 = Release|x64
EndGlobalSection
GlobalSection(ProjectConfigurationPlatforms) = postSolution
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.ActiveCfg = Debug|x64
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.Build.0 = Debug|x64
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.ActiveCfg = Release|x64
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.Build.0 = Release|x64
EndGlobalSection
GlobalSection(SolutionProperties) = preSolution
HideSolutionNode = FALSE
EndGlobalSection
EndGlobal

View File

@@ -1,107 +0,0 @@
<?xml version="1.0" encoding="utf-8"?>
<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
<PropertyGroup>
<CUDAPropsPath Condition="'$(CUDAPropsPath)'==''">$(VCTargetsPath)\BuildCustomizations</CUDAPropsPath>
</PropertyGroup>
<ItemGroup Label="ProjectConfigurations">
<ProjectConfiguration Include="Debug|x64">
<Configuration>Debug</Configuration>
<Platform>x64</Platform>
</ProjectConfiguration>
<ProjectConfiguration Include="Release|x64">
<Configuration>Release</Configuration>
<Platform>x64</Platform>
</ProjectConfiguration>
</ItemGroup>
<PropertyGroup Label="Globals">
<ProjectGuid>{997E0757-EA74-4A4E-A0FC-47D8C8831A15}</ProjectGuid>
<RootNamespace>cudaTensorCoreGemm_vs2012</RootNamespace>
<ProjectName>cudaTensorCoreGemm</ProjectName>
<CudaToolkitCustomDir />
</PropertyGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
<PropertyGroup>
<ConfigurationType>Application</ConfigurationType>
<CharacterSet>MultiByte</CharacterSet>
<PlatformToolset>v110</PlatformToolset>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)'=='Debug'">
<UseDebugLibraries>true</UseDebugLibraries>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)'=='Release'">
<WholeProgramOptimization>true</WholeProgramOptimization>
</PropertyGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
<ImportGroup Label="ExtensionSettings">
<Import Project="$(CUDAPropsPath)\CUDA 11.0.props" />
</ImportGroup>
<ImportGroup Label="PropertySheets">
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
</ImportGroup>
<PropertyGroup Label="UserMacros" />
<PropertyGroup>
<IntDir>$(Platform)/$(Configuration)/</IntDir>
<IncludePath>$(IncludePath)</IncludePath>
<CodeAnalysisRuleSet>AllRules.ruleset</CodeAnalysisRuleSet>
<CodeAnalysisRules />
<CodeAnalysisRuleAssemblies />
</PropertyGroup>
<PropertyGroup Condition="'$(Platform)'=='x64'">
<OutDir>../../bin/win64/$(Configuration)/</OutDir>
</PropertyGroup>
<ItemDefinitionGroup>
<ClCompile>
<WarningLevel>Level3</WarningLevel>
<PreprocessorDefinitions>WIN32;_MBCS;%(PreprocessorDefinitions)</PreprocessorDefinitions>
<AdditionalIncludeDirectories>./;$(CudaToolkitDir)/include;../../Common;</AdditionalIncludeDirectories>
</ClCompile>
<Link>
<SubSystem>Console</SubSystem>
<AdditionalDependencies>cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
<OutputFile>$(OutDir)/cudaTensorCoreGemm.exe</OutputFile>
</Link>
<CudaCompile>
<CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;</CodeGeneration>
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
<Include>./;../../Common</Include>
<Defines>WIN32</Defines>
</CudaCompile>
</ItemDefinitionGroup>
<ItemDefinitionGroup Condition="'$(Configuration)'=='Debug'">
<ClCompile>
<Optimization>Disabled</Optimization>
<RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
</ClCompile>
<Link>
<GenerateDebugInformation>true</GenerateDebugInformation>
<LinkTimeCodeGeneration>Default</LinkTimeCodeGeneration>
</Link>
<CudaCompile>
<Runtime>MTd</Runtime>
<TargetMachinePlatform>64</TargetMachinePlatform>
</CudaCompile>
</ItemDefinitionGroup>
<ItemDefinitionGroup Condition="'$(Configuration)'=='Release'">
<ClCompile>
<Optimization>MaxSpeed</Optimization>
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
</ClCompile>
<Link>
<GenerateDebugInformation>false</GenerateDebugInformation>
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
</Link>
<CudaCompile>
<Runtime>MT</Runtime>
<TargetMachinePlatform>64</TargetMachinePlatform>
</CudaCompile>
</ItemDefinitionGroup>
<ItemGroup>
<CudaCompile Include="cudaTensorCoreGemm.cu" />
</ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 11.0.targets" />
</ImportGroup>
</Project>

View File

@@ -1,20 +0,0 @@

Microsoft Visual Studio Solution File, Format Version 13.00
# Visual Studio 2013
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "cudaTensorCoreGemm", "cudaTensorCoreGemm_vs2013.vcxproj", "{997E0757-EA74-4A4E-A0FC-47D8C8831A15}"
EndProject
Global
GlobalSection(SolutionConfigurationPlatforms) = preSolution
Debug|x64 = Debug|x64
Release|x64 = Release|x64
EndGlobalSection
GlobalSection(ProjectConfigurationPlatforms) = postSolution
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.ActiveCfg = Debug|x64
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.Build.0 = Debug|x64
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.ActiveCfg = Release|x64
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.Build.0 = Release|x64
EndGlobalSection
GlobalSection(SolutionProperties) = preSolution
HideSolutionNode = FALSE
EndGlobalSection
EndGlobal

View File

@@ -1,107 +0,0 @@
<?xml version="1.0" encoding="utf-8"?>
<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
<PropertyGroup>
<CUDAPropsPath Condition="'$(CUDAPropsPath)'==''">$(VCTargetsPath)\BuildCustomizations</CUDAPropsPath>
</PropertyGroup>
<ItemGroup Label="ProjectConfigurations">
<ProjectConfiguration Include="Debug|x64">
<Configuration>Debug</Configuration>
<Platform>x64</Platform>
</ProjectConfiguration>
<ProjectConfiguration Include="Release|x64">
<Configuration>Release</Configuration>
<Platform>x64</Platform>
</ProjectConfiguration>
</ItemGroup>
<PropertyGroup Label="Globals">
<ProjectGuid>{997E0757-EA74-4A4E-A0FC-47D8C8831A15}</ProjectGuid>
<RootNamespace>cudaTensorCoreGemm_vs2013</RootNamespace>
<ProjectName>cudaTensorCoreGemm</ProjectName>
<CudaToolkitCustomDir />
</PropertyGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
<PropertyGroup>
<ConfigurationType>Application</ConfigurationType>
<CharacterSet>MultiByte</CharacterSet>
<PlatformToolset>v120</PlatformToolset>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)'=='Debug'">
<UseDebugLibraries>true</UseDebugLibraries>
</PropertyGroup>
<PropertyGroup Condition="'$(Configuration)'=='Release'">
<WholeProgramOptimization>true</WholeProgramOptimization>
</PropertyGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
<ImportGroup Label="ExtensionSettings">
<Import Project="$(CUDAPropsPath)\CUDA 11.0.props" />
</ImportGroup>
<ImportGroup Label="PropertySheets">
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
</ImportGroup>
<PropertyGroup Label="UserMacros" />
<PropertyGroup>
<IntDir>$(Platform)/$(Configuration)/</IntDir>
<IncludePath>$(IncludePath)</IncludePath>
<CodeAnalysisRuleSet>AllRules.ruleset</CodeAnalysisRuleSet>
<CodeAnalysisRules />
<CodeAnalysisRuleAssemblies />
</PropertyGroup>
<PropertyGroup Condition="'$(Platform)'=='x64'">
<OutDir>../../bin/win64/$(Configuration)/</OutDir>
</PropertyGroup>
<ItemDefinitionGroup>
<ClCompile>
<WarningLevel>Level3</WarningLevel>
<PreprocessorDefinitions>WIN32;_MBCS;%(PreprocessorDefinitions)</PreprocessorDefinitions>
<AdditionalIncludeDirectories>./;$(CudaToolkitDir)/include;../../Common;</AdditionalIncludeDirectories>
</ClCompile>
<Link>
<SubSystem>Console</SubSystem>
<AdditionalDependencies>cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
<OutputFile>$(OutDir)/cudaTensorCoreGemm.exe</OutputFile>
</Link>
<CudaCompile>
<CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;</CodeGeneration>
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
<Include>./;../../Common</Include>
<Defines>WIN32</Defines>
</CudaCompile>
</ItemDefinitionGroup>
<ItemDefinitionGroup Condition="'$(Configuration)'=='Debug'">
<ClCompile>
<Optimization>Disabled</Optimization>
<RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
</ClCompile>
<Link>
<GenerateDebugInformation>true</GenerateDebugInformation>
<LinkTimeCodeGeneration>Default</LinkTimeCodeGeneration>
</Link>
<CudaCompile>
<Runtime>MTd</Runtime>
<TargetMachinePlatform>64</TargetMachinePlatform>
</CudaCompile>
</ItemDefinitionGroup>
<ItemDefinitionGroup Condition="'$(Configuration)'=='Release'">
<ClCompile>
<Optimization>MaxSpeed</Optimization>
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
</ClCompile>
<Link>
<GenerateDebugInformation>false</GenerateDebugInformation>
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
</Link>
<CudaCompile>
<Runtime>MT</Runtime>
<TargetMachinePlatform>64</TargetMachinePlatform>
</CudaCompile>
</ItemDefinitionGroup>
<ItemGroup>
<CudaCompile Include="cudaTensorCoreGemm.cu" />
</ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 11.0.targets" />
</ImportGroup>
</Project>

View File

@@ -33,7 +33,7 @@
</PropertyGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
<ImportGroup Label="ExtensionSettings">
<Import Project="$(CUDAPropsPath)\CUDA 11.0.props" />
<Import Project="$(CUDAPropsPath)\CUDA 11.1.props" />
</ImportGroup>
<ImportGroup Label="PropertySheets">
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
@@ -62,7 +62,7 @@
<OutputFile>$(OutDir)/cudaTensorCoreGemm.exe</OutputFile>
</Link>
<CudaCompile>
<CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;</CodeGeneration>
<CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
<Include>./;../../Common</Include>
<Defines>WIN32</Defines>
@@ -102,6 +102,6 @@
</ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 11.0.targets" />
<Import Project="$(CUDAPropsPath)\CUDA 11.1.targets" />
</ImportGroup>
</Project>

View File

@@ -38,7 +38,7 @@
</PropertyGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
<ImportGroup Label="ExtensionSettings">
<Import Project="$(CUDAPropsPath)\CUDA 11.0.props" />
<Import Project="$(CUDAPropsPath)\CUDA 11.1.props" />
</ImportGroup>
<ImportGroup Label="PropertySheets">
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
@@ -67,7 +67,7 @@
<OutputFile>$(OutDir)/cudaTensorCoreGemm.exe</OutputFile>
</Link>
<CudaCompile>
<CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;</CodeGeneration>
<CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
<Include>./;../../Common</Include>
<Defines>WIN32</Defines>
@@ -107,6 +107,6 @@
</ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 11.0.targets" />
<Import Project="$(CUDAPropsPath)\CUDA 11.1.targets" />
</ImportGroup>
</Project>

View File

@@ -34,7 +34,7 @@
</PropertyGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
<ImportGroup Label="ExtensionSettings">
<Import Project="$(CUDAPropsPath)\CUDA 11.0.props" />
<Import Project="$(CUDAPropsPath)\CUDA 11.1.props" />
</ImportGroup>
<ImportGroup Label="PropertySheets">
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
@@ -63,7 +63,7 @@
<OutputFile>$(OutDir)/cudaTensorCoreGemm.exe</OutputFile>
</Link>
<CudaCompile>
<CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;</CodeGeneration>
<CodeGeneration>compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
<Include>./;../../Common</Include>
<Defines>WIN32</Defines>
@@ -103,6 +103,6 @@
</ItemGroup>
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
<ImportGroup Label="ExtensionTargets">
<Import Project="$(CUDAPropsPath)\CUDA 11.0.targets" />
<Import Project="$(CUDAPropsPath)\CUDA 11.1.targets" />
</ImportGroup>
</Project>