mirror of
https://github.com/NVIDIA/cuda-samples.git
synced 2026-10-23 21:28:17 +08:00
Add and update samples for cuda 11.1 support
This commit is contained in:
Mahesh Doijade
@@ -268,20 +268,24 @@ LIBRARIES :=
|
||||
|
||||
################################################################################
|
||||
|
||||
PTX_FILE := matrixMul_kernel${TARGET_SIZE}.ptx
|
||||
FATBIN_FILE := matrixMul_kernel${TARGET_SIZE}.fatbin
|
||||
|
||||
# Gencode arguments
|
||||
SMS ?=
|
||||
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64))
|
||||
SMS ?= 35 37 50 52 60 61 70 72 75 80 86
|
||||
else
|
||||
SMS ?= 35 37 50 52 60 61 70 75 80 86
|
||||
endif
|
||||
|
||||
ifeq ($(SMS),)
|
||||
$(info >>> WARNING - no SM architectures have been specified - waiving sample <<<)
|
||||
SAMPLE_ENABLED := 0
|
||||
endif
|
||||
|
||||
ifeq ($(GENCODE_FLAGS),)
|
||||
# Generate SASS code for each SM architecture listed in $(SMS)
|
||||
$(foreach sm,$(SMS),$(eval GENCODE_FLAGS += -gencode arch=compute_$(sm),code=sm_$(sm)))
|
||||
|
||||
ifeq ($(SMS),)
|
||||
# Generate PTX code from SM 35
|
||||
GENCODE_FLAGS += -gencode arch=compute_35,code=compute_35
|
||||
endif
|
||||
|
||||
# Generate PTX code from the highest SM architecture in $(SMS) to guarantee forward-compatibility
|
||||
HIGHEST_SM := $(lastword $(sort $(SMS)))
|
||||
ifneq ($(HIGHEST_SM),)
|
||||
@@ -355,7 +359,7 @@ endif
|
||||
# Target rules
|
||||
all: build
|
||||
|
||||
build: matrixMulDrv $(PTX_FILE)
|
||||
build: matrixMulDrv $(FATBIN_FILE)
|
||||
|
||||
check.deps:
|
||||
ifeq ($(SAMPLE_ENABLED),0)
|
||||
@@ -364,8 +368,8 @@ else
|
||||
@echo "Sample is ready - all dependencies have been met"
|
||||
endif
|
||||
|
||||
$(PTX_FILE): matrixMul_kernel.cu
|
||||
$(EXEC) $(NVCC) $(INCLUDES) $(ALL_CCFLAGS) $(GENCODE_FLAGS) -o $@ -ptx $<
|
||||
$(FATBIN_FILE): matrixMul_kernel.cu
|
||||
$(EXEC) $(NVCC) $(INCLUDES) $(ALL_CCFLAGS) $(GENCODE_FLAGS) -o $@ -fatbin $<
|
||||
$(EXEC) mkdir -p data
|
||||
$(EXEC) cp -f $@ ./data
|
||||
$(EXEC) mkdir -p ../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)
|
||||
@@ -383,8 +387,9 @@ run: build
|
||||
$(EXEC) ./matrixMulDrv
|
||||
|
||||
clean:
|
||||
rm -f matrixMulDrv matrixMulDrv.o data/$(PTX_FILE) $(PTX_FILE)
|
||||
rm -f matrixMulDrv matrixMulDrv.o data/$(FATBIN_FILE) $(FATBIN_FILE)
|
||||
rm -rf ../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)/matrixMulDrv
|
||||
rm -rf ../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)/$(PTX_FILE)
|
||||
|
||||
rm -rf ../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)/$(FATBIN_FILE)
|
||||
|
||||
clobber: clean
|
||||
|
||||
@@ -10,7 +10,7 @@ CUDA Driver API, Matrix Multiply
|
||||
|
||||
## Supported SM Architectures
|
||||
|
||||
[SM 3.5 ](https://developer.nvidia.com/cuda-gpus) [SM 3.7 ](https://developer.nvidia.com/cuda-gpus) [SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus)
|
||||
[SM 3.5 ](https://developer.nvidia.com/cuda-gpus) [SM 3.7 ](https://developer.nvidia.com/cuda-gpus) [SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus)
|
||||
|
||||
## Supported OSes
|
||||
|
||||
@@ -27,7 +27,7 @@ cuModuleLoad, cuModuleLoadDataEx, cuModuleGetFunction, cuMemAlloc, cuMemFree, cu
|
||||
|
||||
## 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
|
||||
|
||||
|
||||
@@ -46,13 +46,15 @@
|
||||
|
||||
// includes, system
|
||||
#include <builtin_types.h>
|
||||
#include <cuda.h>
|
||||
#include <drvapi_error_string.h>
|
||||
#include <math.h>
|
||||
#include <stdio.h>
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
#include <iostream>
|
||||
#include <cstring>
|
||||
|
||||
// includes, project
|
||||
// includes, project, CUDA
|
||||
#include <cuda.h>
|
||||
#include <helper_cuda_drvapi.h>
|
||||
#include <helper_image.h>
|
||||
#include <helper_string.h>
|
||||
@@ -63,8 +65,6 @@
|
||||
#include <string>
|
||||
#include "matrixMul.h"
|
||||
|
||||
// includes, CUDA
|
||||
const bool use_64bit_memory_address = false;
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// declaration, forward
|
||||
@@ -74,15 +74,10 @@ void randomInit(float *, int);
|
||||
extern "C" void computeGold(float *, const float *, const float *, unsigned int,
|
||||
unsigned int, unsigned int);
|
||||
|
||||
static CUresult initCUDA(int argc, char **argv, CUfunction *pMatrixMul);
|
||||
static int initCUDA(int argc, char **argv, CUfunction *pMatrixMul);
|
||||
|
||||
// define input ptx file for different platforms
|
||||
#if defined(_WIN64) || defined(__LP64__)
|
||||
#define PTX_FILE "matrixMul_kernel64.ptx"
|
||||
#define CUBIN_FILE "matrixMul_kernel64.cubin"
|
||||
#else
|
||||
#define PTX_FILE "matrixMul_kernel32.ptx"
|
||||
#define CUBIN_FILE "matrixMul_kernel32.cubin"
|
||||
#ifndef FATBIN_FILE
|
||||
#define FATBIN_FILE "matrixMul_kernel64.fatbin"
|
||||
#endif
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
@@ -118,13 +113,7 @@ void runTest(int argc, char **argv) {
|
||||
CUfunction matrixMul = NULL;
|
||||
int block_size = 32;
|
||||
|
||||
CUresult error_id = initCUDA(argc, argv, &matrixMul);
|
||||
|
||||
if (error_id != CUDA_SUCCESS) {
|
||||
printf("initCUDA() returned %d\n-> %s\n", error_id,
|
||||
getCudaDrvErrorString(error_id));
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
initCUDA(argc, argv, &matrixMul);
|
||||
|
||||
// set seed for rand()
|
||||
srand(2006);
|
||||
@@ -142,18 +131,6 @@ void runTest(int argc, char **argv) {
|
||||
constantInit(h_A, size_A, 1.0f);
|
||||
constantInit(h_B, size_B, valB);
|
||||
|
||||
// First reserve about 4GB of memory, so we ensure that all memory allocated
|
||||
// afterwards is > 4GB
|
||||
CUdeviceptr d_Mem[4];
|
||||
|
||||
if (use_64bit_memory_address) {
|
||||
unsigned int mem_size = 1024 * 1024 * 1024;
|
||||
checkCudaErrors(cuMemAlloc(&d_Mem[0], mem_size));
|
||||
checkCudaErrors(cuMemAlloc(&d_Mem[1], mem_size));
|
||||
checkCudaErrors(cuMemAlloc(&d_Mem[2], mem_size));
|
||||
checkCudaErrors(cuMemAlloc(&d_Mem[3], mem_size));
|
||||
}
|
||||
|
||||
// allocate device memory
|
||||
CUdeviceptr d_A;
|
||||
checkCudaErrors(cuMemAlloc(&d_A, mem_size_A));
|
||||
@@ -190,8 +167,6 @@ void runTest(int argc, char **argv) {
|
||||
if (1) {
|
||||
// This is the new CUDA 4.0 API for Kernel Parameter passing and Kernel
|
||||
// Launching (simplier method)
|
||||
if (use_64bit_memory_address &&
|
||||
(totalGlobalMem > (uint64_t)4 * 1024 * 1024 * 1024L)) {
|
||||
size_t Matrix_Width_A = (size_t)WA;
|
||||
size_t Matrix_Width_B = (size_t)WB;
|
||||
void *args[5] = {&d_C, &d_A, &d_B, &Matrix_Width_A, &Matrix_Width_B};
|
||||
@@ -199,17 +174,6 @@ void runTest(int argc, char **argv) {
|
||||
checkCudaErrors(cuLaunchKernel(
|
||||
matrixMul, grid.x, grid.y, grid.z, block.x, block.y, block.z,
|
||||
2 * block_size * block_size * sizeof(float), NULL, args, NULL));
|
||||
|
||||
} else {
|
||||
int Matrix_Width_A = WA;
|
||||
int Matrix_Width_B = WB;
|
||||
void *args[5] = {&d_C, &d_A, &d_B, &Matrix_Width_A, &Matrix_Width_B};
|
||||
// new CUDA 4.0 Driver API Kernel launch call
|
||||
checkCudaErrors(cuLaunchKernel(
|
||||
matrixMul, grid.x, grid.y, grid.z, block.x, block.y, block.z,
|
||||
2 * block_size * block_size * sizeof(float), NULL, args, NULL));
|
||||
}
|
||||
|
||||
} else {
|
||||
// This is the new CUDA 4.0 API for Kernel Parameter passing and Kernel
|
||||
// Launching (advanced method)
|
||||
@@ -225,24 +189,13 @@ void runTest(int argc, char **argv) {
|
||||
*(reinterpret_cast<CUdeviceptr *>(&argBuffer[offset])) = d_B;
|
||||
offset += sizeof(d_B);
|
||||
|
||||
if (use_64bit_memory_address &&
|
||||
(totalGlobalMem > (uint64_t)4 * 1024 * 1024 * 1024L)) {
|
||||
size_t Matrix_Width_A = (size_t)WA;
|
||||
size_t Matrix_Width_B = (size_t)WB;
|
||||
|
||||
*(reinterpret_cast<CUdeviceptr *>(&argBuffer[offset])) = Matrix_Width_A;
|
||||
offset += sizeof(Matrix_Width_A);
|
||||
*(reinterpret_cast<CUdeviceptr *>(&argBuffer[offset])) = Matrix_Width_B;
|
||||
offset += sizeof(Matrix_Width_B);
|
||||
} else {
|
||||
int Matrix_Width_A = WA;
|
||||
int Matrix_Width_B = WB;
|
||||
|
||||
*(reinterpret_cast<int *>(&argBuffer[offset])) = Matrix_Width_A;
|
||||
offset += sizeof(Matrix_Width_A);
|
||||
*(reinterpret_cast<int *>(&argBuffer[offset])) = Matrix_Width_B;
|
||||
offset += sizeof(Matrix_Width_B);
|
||||
}
|
||||
*(reinterpret_cast<CUdeviceptr *>(&argBuffer[offset])) = Matrix_Width_A;
|
||||
offset += sizeof(Matrix_Width_A);
|
||||
*(reinterpret_cast<CUdeviceptr *>(&argBuffer[offset])) = Matrix_Width_B;
|
||||
offset += sizeof(Matrix_Width_B);
|
||||
|
||||
void *kernel_launch_config[5] = {CU_LAUNCH_PARAM_BUFFER_POINTER, argBuffer,
|
||||
CU_LAUNCH_PARAM_BUFFER_SIZE, &offset,
|
||||
@@ -281,13 +234,6 @@ void runTest(int argc, char **argv) {
|
||||
"Results may vary when GPU Boost is enabled.\n");
|
||||
|
||||
// clean up memory
|
||||
if (use_64bit_memory_address) {
|
||||
cuMemFree(d_Mem[0]);
|
||||
cuMemFree(d_Mem[1]);
|
||||
cuMemFree(d_Mem[2]);
|
||||
cuMemFree(d_Mem[3]);
|
||||
}
|
||||
|
||||
free(h_A);
|
||||
free(h_B);
|
||||
free(h_C);
|
||||
@@ -304,46 +250,11 @@ void randomInit(float *data, int size) {
|
||||
}
|
||||
}
|
||||
|
||||
bool inline findModulePath(const char *module_file, std::string &module_path,
|
||||
char **argv, std::string &ptx_source) {
|
||||
char *actual_path = sdkFindFilePath(module_file, argv[0]);
|
||||
|
||||
if (actual_path) {
|
||||
module_path = actual_path;
|
||||
} else {
|
||||
printf("> findModulePath file not found: <%s> \n", module_file);
|
||||
return false;
|
||||
}
|
||||
|
||||
if (module_path.empty()) {
|
||||
printf("> findModulePath file not found: <%s> \n", module_file);
|
||||
return false;
|
||||
} else {
|
||||
printf("> findModulePath <%s>\n", module_path.c_str());
|
||||
|
||||
if (module_path.rfind(".ptx") != std::string::npos) {
|
||||
FILE *fp = fopen(module_path.c_str(), "rb");
|
||||
fseek(fp, 0, SEEK_END);
|
||||
int file_size = ftell(fp);
|
||||
char *buf = new char[file_size + 1];
|
||||
fseek(fp, 0, SEEK_SET);
|
||||
fread(buf, sizeof(char), file_size, fp);
|
||||
fclose(fp);
|
||||
buf[file_size] = '\0';
|
||||
ptx_source = buf;
|
||||
delete[] buf;
|
||||
}
|
||||
|
||||
return true;
|
||||
}
|
||||
}
|
||||
|
||||
static CUresult initCUDA(int argc, char **argv, CUfunction *pMatrixMul) {
|
||||
static int initCUDA(int argc, char **argv, CUfunction *pMatrixMul) {
|
||||
CUfunction cuFunction = 0;
|
||||
CUresult status;
|
||||
int major = 0, minor = 0;
|
||||
char deviceName[100];
|
||||
std::string module_path, ptx_source;
|
||||
|
||||
cuDevice = findCudaDeviceDRV(argc, (const char **)argv);
|
||||
|
||||
@@ -352,85 +263,39 @@ static CUresult initCUDA(int argc, char **argv, CUfunction *pMatrixMul) {
|
||||
&major, CU_DEVICE_ATTRIBUTE_COMPUTE_CAPABILITY_MAJOR, cuDevice));
|
||||
checkCudaErrors(cuDeviceGetAttribute(
|
||||
&minor, CU_DEVICE_ATTRIBUTE_COMPUTE_CAPABILITY_MINOR, cuDevice));
|
||||
checkCudaErrors(cuDeviceGetName(deviceName, 256, cuDevice));
|
||||
checkCudaErrors(cuDeviceGetName(deviceName, sizeof(deviceName), cuDevice));
|
||||
printf("> GPU Device has SM %d.%d compute capability\n", major, minor);
|
||||
|
||||
checkCudaErrors(cuDeviceTotalMem(&totalGlobalMem, cuDevice));
|
||||
printf(" Total amount of global memory: %llu bytes\n",
|
||||
(long long unsigned int)totalGlobalMem);
|
||||
printf(" 64-bit Memory Address: %s\n",
|
||||
(totalGlobalMem > (uint64_t)4 * 1024 * 1024 * 1024L) ? "YES" : "NO");
|
||||
|
||||
status = cuCtxCreate(&cuContext, 0, cuDevice);
|
||||
|
||||
if (CUDA_SUCCESS != status) {
|
||||
goto Error;
|
||||
}
|
||||
checkCudaErrors(cuCtxCreate(&cuContext, 0, cuDevice));
|
||||
|
||||
// first search for the module path before we load the results
|
||||
if (!findModulePath(PTX_FILE, module_path, argv, ptx_source)) {
|
||||
if (!findModulePath(CUBIN_FILE, module_path, argv, ptx_source)) {
|
||||
printf(
|
||||
"> findModulePath could not find <matrixMul_kernel> ptx or cubin\n");
|
||||
status = CUDA_ERROR_NOT_FOUND;
|
||||
goto Error;
|
||||
}
|
||||
} else {
|
||||
printf("> initCUDA loading module: <%s>\n", module_path.c_str());
|
||||
}
|
||||
std::string module_path;
|
||||
std::ostringstream fatbin;
|
||||
|
||||
if (module_path.rfind("ptx") != std::string::npos) {
|
||||
// in this branch we use compilation with parameters
|
||||
const unsigned int jitNumOptions = 3;
|
||||
CUjit_option *jitOptions = new CUjit_option[jitNumOptions];
|
||||
void **jitOptVals = new void *[jitNumOptions];
|
||||
|
||||
// set up size of compilation log buffer
|
||||
jitOptions[0] = CU_JIT_INFO_LOG_BUFFER_SIZE_BYTES;
|
||||
int jitLogBufferSize = 1024;
|
||||
jitOptVals[0] = reinterpret_cast<void *>(jitLogBufferSize);
|
||||
|
||||
// set up pointer to the compilation log buffer
|
||||
jitOptions[1] = CU_JIT_INFO_LOG_BUFFER;
|
||||
char *jitLogBuffer = new char[jitLogBufferSize];
|
||||
jitOptVals[1] = jitLogBuffer;
|
||||
|
||||
// set up pointer to set the Maximum # of registers for a particular kernel
|
||||
jitOptions[2] = CU_JIT_MAX_REGISTERS;
|
||||
int jitRegCount = 32;
|
||||
jitOptVals[2] = reinterpret_cast<void *>(jitRegCount);
|
||||
|
||||
status =
|
||||
cuModuleLoadDataEx(&cuModule, ptx_source.c_str(), jitNumOptions,
|
||||
jitOptions, reinterpret_cast<void **>(jitOptVals));
|
||||
|
||||
printf("> PTX JIT log:\n%s\n", jitLogBuffer);
|
||||
} else {
|
||||
status = cuModuleLoad(&cuModule, module_path.c_str());
|
||||
}
|
||||
|
||||
if (CUDA_SUCCESS != status) {
|
||||
goto Error;
|
||||
}
|
||||
|
||||
#if USE_64BIT_MEMORY_ADDRESS
|
||||
|
||||
if (totalGlobalMem > (uint64_t)4 * 1024 * 1024 * 1024L) {
|
||||
status = cuModuleGetFunction(&cuFunction, cuModule, "matrixMul_bs32_64bit");
|
||||
} else
|
||||
#endif
|
||||
if (!findFatbinPath(FATBIN_FILE, module_path, argv, fatbin))
|
||||
{
|
||||
status = cuModuleGetFunction(&cuFunction, cuModule, "matrixMul_bs32_32bit");
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
else
|
||||
{
|
||||
printf("> initCUDA loading module: <%s>\n", module_path.c_str());
|
||||
}
|
||||
|
||||
if (CUDA_SUCCESS != status) {
|
||||
goto Error;
|
||||
if (!fatbin.str().size())
|
||||
{
|
||||
printf("fatbin file empty. exiting..\n");
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
|
||||
// Create module from binary file (FATBIN)
|
||||
checkCudaErrors(cuModuleLoadData(&cuModule, fatbin.str().c_str()));
|
||||
checkCudaErrors(cuModuleGetFunction(&cuFunction, cuModule, "matrixMul_bs32_64bit"));
|
||||
|
||||
*pMatrixMul = cuFunction;
|
||||
|
||||
return CUDA_SUCCESS;
|
||||
Error:
|
||||
cuCtxDestroy(cuContext);
|
||||
return status;
|
||||
return 0;
|
||||
}
|
||||
|
||||
@@ -1,20 +0,0 @@
|
||||
|
||||
Microsoft Visual Studio Solution File, Format Version 12.00
|
||||
# Visual Studio 2012
|
||||
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "matrixMulDrv", "matrixMulDrv_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
|
||||
@@ -1,111 +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>matrixMulDrv_vs2012</RootNamespace>
|
||||
<ProjectName>matrixMulDrv</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>./;$(CudaToolkitDir)/include;../../Common;</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cuda.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/matrixMulDrv.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,compute_35;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
|
||||
<Include>./;../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(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="'$(Configuration)'=='Release'">
|
||||
<ClCompile>
|
||||
<Optimization>MaxSpeed</Optimization>
|
||||
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MT</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<ClCompile Include="matrixMulDrv.cpp" />
|
||||
<CudaCompile Include="matrixMul_kernel.cu">
|
||||
<CompileOut Condition="'$(Platform)'=='x64'">data/%(Filename)64.ptx</CompileOut>
|
||||
<NvccCompilation>ptx</NvccCompilation>
|
||||
</CudaCompile>
|
||||
<ClInclude Include="matrixMul.h" />
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.0.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
@@ -1,20 +0,0 @@
|
||||
|
||||
Microsoft Visual Studio Solution File, Format Version 13.00
|
||||
# Visual Studio 2013
|
||||
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "matrixMulDrv", "matrixMulDrv_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
|
||||
@@ -1,111 +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>matrixMulDrv_vs2013</RootNamespace>
|
||||
<ProjectName>matrixMulDrv</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>./;$(CudaToolkitDir)/include;../../Common;</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cuda.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/matrixMulDrv.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,compute_35;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
|
||||
<Include>./;../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(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="'$(Configuration)'=='Release'">
|
||||
<ClCompile>
|
||||
<Optimization>MaxSpeed</Optimization>
|
||||
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MT</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<ClCompile Include="matrixMulDrv.cpp" />
|
||||
<CudaCompile Include="matrixMul_kernel.cu">
|
||||
<CompileOut Condition="'$(Platform)'=='x64'">data/%(Filename)64.ptx</CompileOut>
|
||||
<NvccCompilation>ptx</NvccCompilation>
|
||||
</CudaCompile>
|
||||
<ClInclude Include="matrixMul.h" />
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.0.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
@@ -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)/matrixMulDrv.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,compute_35;</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" %(AdditionalOptions)</AdditionalOptions>
|
||||
<Include>./;../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
@@ -99,13 +99,13 @@
|
||||
<ItemGroup>
|
||||
<ClCompile Include="matrixMulDrv.cpp" />
|
||||
<CudaCompile Include="matrixMul_kernel.cu">
|
||||
<CompileOut Condition="'$(Platform)'=='x64'">data/%(Filename)64.ptx</CompileOut>
|
||||
<NvccCompilation>ptx</NvccCompilation>
|
||||
<CompileOut Condition="'$(Platform)'=='x64'">data/%(Filename)64.fatbin</CompileOut>
|
||||
<NvccCompilation>fatbin</NvccCompilation>
|
||||
</CudaCompile>
|
||||
<ClInclude Include="matrixMul.h" />
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.0.targets" />
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.1.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
|
||||
@@ -38,7 +38,7 @@
|
||||
</PropertyGroup>
|
||||
<Import Project="$(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>$(OutDir)/matrixMulDrv.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,compute_35;</CodeGeneration>
|
||||
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
|
||||
<Include>./;../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
@@ -104,13 +104,13 @@
|
||||
<ItemGroup>
|
||||
<ClCompile Include="matrixMulDrv.cpp" />
|
||||
<CudaCompile Include="matrixMul_kernel.cu">
|
||||
<CompileOut Condition="'$(Platform)'=='x64'">data/%(Filename)64.ptx</CompileOut>
|
||||
<NvccCompilation>ptx</NvccCompilation>
|
||||
<CompileOut Condition="'$(Platform)'=='x64'">data/%(Filename)64.fatbin</CompileOut>
|
||||
<NvccCompilation>fatbin</NvccCompilation>
|
||||
</CudaCompile>
|
||||
<ClInclude Include="matrixMul.h" />
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.0.targets" />
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.1.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
|
||||
@@ -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)/matrixMulDrv.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,compute_35;</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" %(AdditionalOptions)</AdditionalOptions>
|
||||
<Include>./;../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
@@ -100,13 +100,13 @@
|
||||
<ItemGroup>
|
||||
<ClCompile Include="matrixMulDrv.cpp" />
|
||||
<CudaCompile Include="matrixMul_kernel.cu">
|
||||
<CompileOut Condition="'$(Platform)'=='x64'">data/%(Filename)64.ptx</CompileOut>
|
||||
<NvccCompilation>ptx</NvccCompilation>
|
||||
<CompileOut Condition="'$(Platform)'=='x64'">data/%(Filename)64.fatbin</CompileOut>
|
||||
<NvccCompilation>fatbin</NvccCompilation>
|
||||
</CudaCompile>
|
||||
<ClInclude Include="matrixMul.h" />
|
||||
</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>
|
||||
|
||||
@@ -34,16 +34,8 @@
|
||||
|
||||
#include <stdio.h>
|
||||
|
||||
#define CHECK_BANK_CONFLICTS 0
|
||||
#if CHECK_BANK_CONFLICTS
|
||||
#define AS(i, j) \
|
||||
cutilBankChecker((reinterpret_cast<float *>(&As[0][0])), (block_size * i + j))
|
||||
#define BS(i, j) \
|
||||
cutilBankChecker((reinterpret_cast<float *>(&Bs[0][0])), (block_size * i + j))
|
||||
#else
|
||||
#define AS(i, j) As[i][j]
|
||||
#define BS(i, j) Bs[i][j]
|
||||
#endif
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
//! Matrix multiplication on the device: C = A * B
|
||||
@@ -119,18 +111,10 @@ __device__ void matrixMul(float *C, float *A, float *B, size_type wA,
|
||||
}
|
||||
|
||||
// C wrappers around our template kernel
|
||||
extern "C" __global__ void matrixMul_bs16_32bit(float *C, float *A, float *B,
|
||||
int wA, int wB) {
|
||||
matrixMul<16, int>(C, A, B, wA, wB);
|
||||
}
|
||||
extern "C" __global__ void matrixMul_bs16_64bit(float *C, float *A, float *B,
|
||||
size_t wA, size_t wB) {
|
||||
matrixMul<16, size_t>(C, A, B, wA, wB);
|
||||
}
|
||||
extern "C" __global__ void matrixMul_bs32_32bit(float *C, float *A, float *B,
|
||||
int wA, int wB) {
|
||||
matrixMul<32, int>(C, A, B, wA, wB);
|
||||
}
|
||||
extern "C" __global__ void matrixMul_bs32_64bit(float *C, float *A, float *B,
|
||||
size_t wA, size_t wB) {
|
||||
matrixMul<32, size_t>(C, A, B, wA, wB);
|
||||
|
||||
Reference in New Issue
Block a user