mirror of
https://github.com/NVIDIA/cuda-samples.git
synced 2026-10-23 21:28:17 +08:00
add and update samples with CUDA 11.3 support
This commit is contained in:
Rutwik Choughule
357
Samples/simpleCUBLAS_LU/Makefile
Normal file
357
Samples/simpleCUBLAS_LU/Makefile
Normal file
@@ -0,0 +1,357 @@
|
||||
################################################################################
|
||||
# Copyright (c) 2019, NVIDIA CORPORATION. All rights reserved.
|
||||
#
|
||||
# Redistribution and use in source and binary forms, with or without
|
||||
# modification, are permitted provided that the following conditions
|
||||
# are met:
|
||||
# * Redistributions of source code must retain the above copyright
|
||||
# notice, this list of conditions and the following disclaimer.
|
||||
# * Redistributions in binary form must reproduce the above copyright
|
||||
# notice, this list of conditions and the following disclaimer in the
|
||||
# documentation and/or other materials provided with the distribution.
|
||||
# * Neither the name of NVIDIA CORPORATION nor the names of its
|
||||
# contributors may be used to endorse or promote products derived
|
||||
# from this software without specific prior written permission.
|
||||
#
|
||||
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
|
||||
# EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
||||
# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
|
||||
# PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
|
||||
# CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
|
||||
# EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
|
||||
# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
|
||||
# PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
|
||||
# OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
||||
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
||||
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
#
|
||||
################################################################################
|
||||
#
|
||||
# Makefile project only supported on Mac OS X and Linux Platforms)
|
||||
#
|
||||
################################################################################
|
||||
|
||||
# Location of the CUDA Toolkit
|
||||
CUDA_PATH ?= /usr/local/cuda
|
||||
|
||||
##############################
|
||||
# start deprecated interface #
|
||||
##############################
|
||||
ifeq ($(x86_64),1)
|
||||
$(info WARNING - x86_64 variable has been deprecated)
|
||||
$(info WARNING - please use TARGET_ARCH=x86_64 instead)
|
||||
TARGET_ARCH ?= x86_64
|
||||
endif
|
||||
ifeq ($(ARMv7),1)
|
||||
$(info WARNING - ARMv7 variable has been deprecated)
|
||||
$(info WARNING - please use TARGET_ARCH=armv7l instead)
|
||||
TARGET_ARCH ?= armv7l
|
||||
endif
|
||||
ifeq ($(aarch64),1)
|
||||
$(info WARNING - aarch64 variable has been deprecated)
|
||||
$(info WARNING - please use TARGET_ARCH=aarch64 instead)
|
||||
TARGET_ARCH ?= aarch64
|
||||
endif
|
||||
ifeq ($(ppc64le),1)
|
||||
$(info WARNING - ppc64le variable has been deprecated)
|
||||
$(info WARNING - please use TARGET_ARCH=ppc64le instead)
|
||||
TARGET_ARCH ?= ppc64le
|
||||
endif
|
||||
ifneq ($(GCC),)
|
||||
$(info WARNING - GCC variable has been deprecated)
|
||||
$(info WARNING - please use HOST_COMPILER=$(GCC) instead)
|
||||
HOST_COMPILER ?= $(GCC)
|
||||
endif
|
||||
ifneq ($(abi),)
|
||||
$(error ERROR - abi variable has been removed)
|
||||
endif
|
||||
############################
|
||||
# end deprecated interface #
|
||||
############################
|
||||
|
||||
# architecture
|
||||
HOST_ARCH := $(shell uname -m)
|
||||
TARGET_ARCH ?= $(HOST_ARCH)
|
||||
ifneq (,$(filter $(TARGET_ARCH),x86_64 aarch64 sbsa ppc64le armv7l))
|
||||
ifneq ($(TARGET_ARCH),$(HOST_ARCH))
|
||||
ifneq (,$(filter $(TARGET_ARCH),x86_64 aarch64 sbsa ppc64le))
|
||||
TARGET_SIZE := 64
|
||||
else ifneq (,$(filter $(TARGET_ARCH),armv7l))
|
||||
TARGET_SIZE := 32
|
||||
endif
|
||||
else
|
||||
TARGET_SIZE := $(shell getconf LONG_BIT)
|
||||
endif
|
||||
else
|
||||
$(error ERROR - unsupported value $(TARGET_ARCH) for TARGET_ARCH!)
|
||||
endif
|
||||
|
||||
# sbsa and aarch64 systems look similar. Need to differentiate them at host level for now.
|
||||
ifeq ($(HOST_ARCH),aarch64)
|
||||
ifeq ($(CUDA_PATH)/targets/sbsa-linux,$(shell ls -1d $(CUDA_PATH)/targets/sbsa-linux 2>/dev/null))
|
||||
HOST_ARCH := sbsa
|
||||
TARGET_ARCH := sbsa
|
||||
endif
|
||||
endif
|
||||
|
||||
ifneq ($(TARGET_ARCH),$(HOST_ARCH))
|
||||
ifeq (,$(filter $(HOST_ARCH)-$(TARGET_ARCH),aarch64-armv7l x86_64-armv7l x86_64-aarch64 x86_64-sbsa x86_64-ppc64le))
|
||||
$(error ERROR - cross compiling from $(HOST_ARCH) to $(TARGET_ARCH) is not supported!)
|
||||
endif
|
||||
endif
|
||||
|
||||
# When on native aarch64 system with userspace of 32-bit, change TARGET_ARCH to armv7l
|
||||
ifeq ($(HOST_ARCH)-$(TARGET_ARCH)-$(TARGET_SIZE),aarch64-aarch64-32)
|
||||
TARGET_ARCH = armv7l
|
||||
endif
|
||||
|
||||
# operating system
|
||||
HOST_OS := $(shell uname -s 2>/dev/null | tr "[:upper:]" "[:lower:]")
|
||||
TARGET_OS ?= $(HOST_OS)
|
||||
ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
|
||||
$(error ERROR - unsupported value $(TARGET_OS) for TARGET_OS!)
|
||||
endif
|
||||
|
||||
# host compiler
|
||||
ifeq ($(TARGET_OS),darwin)
|
||||
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
|
||||
HOST_COMPILER ?= clang++
|
||||
endif
|
||||
else ifneq ($(TARGET_ARCH),$(HOST_ARCH))
|
||||
ifeq ($(HOST_ARCH)-$(TARGET_ARCH),x86_64-armv7l)
|
||||
ifeq ($(TARGET_OS),linux)
|
||||
HOST_COMPILER ?= arm-linux-gnueabihf-g++
|
||||
else ifeq ($(TARGET_OS),qnx)
|
||||
ifeq ($(QNX_HOST),)
|
||||
$(error ERROR - QNX_HOST must be passed to the QNX host toolchain)
|
||||
endif
|
||||
ifeq ($(QNX_TARGET),)
|
||||
$(error ERROR - QNX_TARGET must be passed to the QNX target toolchain)
|
||||
endif
|
||||
export QNX_HOST
|
||||
export QNX_TARGET
|
||||
HOST_COMPILER ?= $(QNX_HOST)/usr/bin/arm-unknown-nto-qnx6.6.0eabi-g++
|
||||
else ifeq ($(TARGET_OS),android)
|
||||
HOST_COMPILER ?= arm-linux-androideabi-g++
|
||||
endif
|
||||
else ifeq ($(TARGET_ARCH),aarch64)
|
||||
ifeq ($(TARGET_OS), linux)
|
||||
HOST_COMPILER ?= aarch64-linux-gnu-g++
|
||||
else ifeq ($(TARGET_OS),qnx)
|
||||
ifeq ($(QNX_HOST),)
|
||||
$(error ERROR - QNX_HOST must be passed to the QNX host toolchain)
|
||||
endif
|
||||
ifeq ($(QNX_TARGET),)
|
||||
$(error ERROR - QNX_TARGET must be passed to the QNX target toolchain)
|
||||
endif
|
||||
export QNX_HOST
|
||||
export QNX_TARGET
|
||||
HOST_COMPILER ?= $(QNX_HOST)/usr/bin/q++
|
||||
else ifeq ($(TARGET_OS), android)
|
||||
HOST_COMPILER ?= aarch64-linux-android-clang++
|
||||
endif
|
||||
else ifeq ($(TARGET_ARCH),sbsa)
|
||||
HOST_COMPILER ?= aarch64-linux-gnu-g++
|
||||
else ifeq ($(TARGET_ARCH),ppc64le)
|
||||
HOST_COMPILER ?= powerpc64le-linux-gnu-g++
|
||||
endif
|
||||
endif
|
||||
HOST_COMPILER ?= g++
|
||||
NVCC := $(CUDA_PATH)/bin/nvcc -ccbin $(HOST_COMPILER)
|
||||
|
||||
# internal flags
|
||||
NVCCFLAGS := -m${TARGET_SIZE}
|
||||
CCFLAGS :=
|
||||
LDFLAGS :=
|
||||
|
||||
# build flags
|
||||
ifeq ($(TARGET_OS),darwin)
|
||||
LDFLAGS += -rpath $(CUDA_PATH)/lib
|
||||
CCFLAGS += -arch $(HOST_ARCH)
|
||||
else ifeq ($(HOST_ARCH)-$(TARGET_ARCH)-$(TARGET_OS),x86_64-armv7l-linux)
|
||||
LDFLAGS += --dynamic-linker=/lib/ld-linux-armhf.so.3
|
||||
CCFLAGS += -mfloat-abi=hard
|
||||
else ifeq ($(TARGET_OS),android)
|
||||
LDFLAGS += -pie
|
||||
CCFLAGS += -fpie -fpic -fexceptions
|
||||
endif
|
||||
|
||||
ifneq ($(TARGET_ARCH),$(HOST_ARCH))
|
||||
ifeq ($(TARGET_ARCH)-$(TARGET_OS),armv7l-linux)
|
||||
ifneq ($(TARGET_FS),)
|
||||
GCCVERSIONLTEQ46 := $(shell expr `$(HOST_COMPILER) -dumpversion` \<= 4.6)
|
||||
ifeq ($(GCCVERSIONLTEQ46),1)
|
||||
CCFLAGS += --sysroot=$(TARGET_FS)
|
||||
endif
|
||||
LDFLAGS += --sysroot=$(TARGET_FS)
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/lib
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/usr/lib
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/usr/lib/arm-linux-gnueabihf
|
||||
endif
|
||||
endif
|
||||
ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-linux)
|
||||
ifneq ($(TARGET_FS),)
|
||||
GCCVERSIONLTEQ46 := $(shell expr `$(HOST_COMPILER) -dumpversion` \<= 4.6)
|
||||
ifeq ($(GCCVERSIONLTEQ46),1)
|
||||
CCFLAGS += --sysroot=$(TARGET_FS)
|
||||
endif
|
||||
LDFLAGS += --sysroot=$(TARGET_FS)
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/lib -L$(TARGET_FS)/lib
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/lib/aarch64-linux-gnu -L$(TARGET_FS)/lib/aarch64-linux-gnu
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/usr/lib -L$(TARGET_FS)/usr/lib
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/usr/lib/aarch64-linux-gnu -L$(TARGET_FS)/usr/lib/aarch64-linux-gnu
|
||||
LDFLAGS += --unresolved-symbols=ignore-in-shared-libs
|
||||
CCFLAGS += -isystem=$(TARGET_FS)/usr/include -I$(TARGET_FS)/usr/include -I$(TARGET_FS)/usr/include/libdrm
|
||||
CCFLAGS += -isystem=$(TARGET_FS)/usr/include/aarch64-linux-gnu -I$(TARGET_FS)/usr/include/aarch64-linux-gnu
|
||||
endif
|
||||
endif
|
||||
ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-qnx)
|
||||
NVCCFLAGS += --qpp-config 5.4.0,gcc_ntoaarch64le
|
||||
CCFLAGS += -DWIN_INTERFACE_CUSTOM -I/usr/include/aarch64-qnx-gnu
|
||||
LDFLAGS += -lsocket
|
||||
LDFLAGS += -L/usr/lib/aarch64-qnx-gnu
|
||||
CCFLAGS += "-Wl\,-rpath-link\,/usr/lib/aarch64-qnx-gnu"
|
||||
ifdef TARGET_OVERRIDE
|
||||
LDFLAGS += -lslog2
|
||||
endif
|
||||
|
||||
ifneq ($(TARGET_FS),)
|
||||
LDFLAGS += -L$(TARGET_FS)/usr/lib
|
||||
CCFLAGS += "-Wl\,-rpath-link\,$(TARGET_FS)/usr/lib"
|
||||
LDFLAGS += -L$(TARGET_FS)/usr/libnvidia
|
||||
CCFLAGS += "-Wl\,-rpath-link\,$(TARGET_FS)/usr/libnvidia"
|
||||
CCFLAGS += -I$(TARGET_FS)/../include
|
||||
endif
|
||||
endif
|
||||
endif
|
||||
|
||||
ifdef TARGET_OVERRIDE # cuda toolkit targets override
|
||||
NVCCFLAGS += -target-dir $(TARGET_OVERRIDE)
|
||||
endif
|
||||
|
||||
# Install directory of different arch
|
||||
CUDA_INSTALL_TARGET_DIR :=
|
||||
ifeq ($(TARGET_ARCH)-$(TARGET_OS),armv7l-linux)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/armv7-linux-gnueabihf/
|
||||
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-linux)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/aarch64-linux/
|
||||
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),sbsa-linux)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/sbsa-linux/
|
||||
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),armv7l-android)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/armv7-linux-androideabi/
|
||||
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-android)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/aarch64-linux-androideabi/
|
||||
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),armv7l-qnx)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/ARMv7-linux-QNX/
|
||||
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-qnx)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/aarch64-qnx/
|
||||
else ifeq ($(TARGET_ARCH),ppc64le)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/ppc64le-linux/
|
||||
endif
|
||||
|
||||
# Debug build flags
|
||||
ifeq ($(dbg),1)
|
||||
NVCCFLAGS += -g -G
|
||||
BUILD_TYPE := debug
|
||||
else
|
||||
BUILD_TYPE := release
|
||||
endif
|
||||
|
||||
ALL_CCFLAGS :=
|
||||
ALL_CCFLAGS += $(NVCCFLAGS)
|
||||
ALL_CCFLAGS += $(EXTRA_NVCCFLAGS)
|
||||
ALL_CCFLAGS += $(addprefix -Xcompiler ,$(CCFLAGS))
|
||||
ALL_CCFLAGS += $(addprefix -Xcompiler ,$(EXTRA_CCFLAGS))
|
||||
|
||||
SAMPLE_ENABLED := 1
|
||||
|
||||
# This sample is not supported on Mac OSX
|
||||
ifeq ($(TARGET_OS),darwin)
|
||||
$(info >>> WARNING - simpleCUBLAS_LU is not supported on Mac OSX - waiving sample <<<)
|
||||
SAMPLE_ENABLED := 0
|
||||
endif
|
||||
|
||||
# This sample is not supported on ARMv7
|
||||
ifeq ($(TARGET_ARCH),armv7l)
|
||||
$(info >>> WARNING - simpleCUBLAS_LU is not supported on ARMv7 - waiving sample <<<)
|
||||
SAMPLE_ENABLED := 0
|
||||
endif
|
||||
|
||||
# This sample is not supported on QNX
|
||||
ifeq ($(TARGET_OS),qnx)
|
||||
$(info >>> WARNING - simpleCUBLAS_LU is not supported on QNX - waiving sample <<<)
|
||||
SAMPLE_ENABLED := 0
|
||||
endif
|
||||
|
||||
ALL_LDFLAGS :=
|
||||
ALL_LDFLAGS += $(ALL_CCFLAGS)
|
||||
ALL_LDFLAGS += $(addprefix -Xlinker ,$(LDFLAGS))
|
||||
ALL_LDFLAGS += $(addprefix -Xlinker ,$(EXTRA_LDFLAGS))
|
||||
|
||||
# Common includes and paths for CUDA
|
||||
INCLUDES := -I../../Common
|
||||
LIBRARIES :=
|
||||
|
||||
################################################################################
|
||||
|
||||
# Gencode arguments
|
||||
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64))
|
||||
SMS ?= 35 37 50 52 60 61 70 72 75 80 86
|
||||
else
|
||||
SMS ?= 35 37 50 52 60 61 70 75 80 86
|
||||
endif
|
||||
|
||||
ifeq ($(SMS),)
|
||||
$(info >>> WARNING - no SM architectures have been specified - waiving sample <<<)
|
||||
SAMPLE_ENABLED := 0
|
||||
endif
|
||||
|
||||
ifeq ($(GENCODE_FLAGS),)
|
||||
# Generate SASS code for each SM architecture listed in $(SMS)
|
||||
$(foreach sm,$(SMS),$(eval GENCODE_FLAGS += -gencode arch=compute_$(sm),code=sm_$(sm)))
|
||||
|
||||
# Generate PTX code from the highest SM architecture in $(SMS) to guarantee forward-compatibility
|
||||
HIGHEST_SM := $(lastword $(sort $(SMS)))
|
||||
ifneq ($(HIGHEST_SM),)
|
||||
GENCODE_FLAGS += -gencode arch=compute_$(HIGHEST_SM),code=compute_$(HIGHEST_SM)
|
||||
endif
|
||||
endif
|
||||
|
||||
ALL_CCFLAGS += --threads 0
|
||||
|
||||
LIBRARIES += -lcublas
|
||||
|
||||
ifeq ($(SAMPLE_ENABLED),0)
|
||||
EXEC ?= @echo "[@]"
|
||||
endif
|
||||
|
||||
################################################################################
|
||||
|
||||
# Target rules
|
||||
all: build
|
||||
|
||||
build: simpleCUBLAS_LU
|
||||
|
||||
check.deps:
|
||||
ifeq ($(SAMPLE_ENABLED),0)
|
||||
@echo "Sample will be waived due to the above missing dependencies"
|
||||
else
|
||||
@echo "Sample is ready - all dependencies have been met"
|
||||
endif
|
||||
|
||||
simpleCUBLAS_LU.o:simpleCUBLAS_LU.cpp
|
||||
$(EXEC) $(NVCC) $(INCLUDES) $(ALL_CCFLAGS) $(GENCODE_FLAGS) -o $@ -c $<
|
||||
|
||||
simpleCUBLAS_LU: simpleCUBLAS_LU.o
|
||||
$(EXEC) $(NVCC) $(ALL_LDFLAGS) $(GENCODE_FLAGS) -o $@ $+ $(LIBRARIES)
|
||||
$(EXEC) mkdir -p ../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)
|
||||
$(EXEC) cp $@ ../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)
|
||||
|
||||
run: build
|
||||
$(EXEC) ./simpleCUBLAS_LU
|
||||
|
||||
clean:
|
||||
rm -f simpleCUBLAS_LU simpleCUBLAS_LU.o
|
||||
rm -rf ../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)/simpleCUBLAS_LU
|
||||
|
||||
clobber: clean
|
||||
68
Samples/simpleCUBLAS_LU/NsightEclipse.xml
Normal file
68
Samples/simpleCUBLAS_LU/NsightEclipse.xml
Normal file
@@ -0,0 +1,68 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<!DOCTYPE entry SYSTEM "SamplesInfo.dtd">
|
||||
<entry>
|
||||
<name>simpleCUBLAS_LU</name>
|
||||
<description><![CDATA[CUDA sample demonstrating cuBLAS API cublasDgetrfBatched() for lower-upper (LU) decomposition of a matrix.]]></description>
|
||||
<devicecompilation>whole</devicecompilation>
|
||||
<includepaths>
|
||||
<path>./</path>
|
||||
<path>../</path>
|
||||
<path>../../common/inc</path>
|
||||
</includepaths>
|
||||
<keyconcepts>
|
||||
<concept level="basic">CUBLAS Library</concept>
|
||||
<concept level="basic">LU decomposition</concept>
|
||||
</keyconcepts>
|
||||
<keywords>
|
||||
<keyword>CUDA</keyword>
|
||||
<keyword>CUBLAS</keyword>
|
||||
<keyword>Linear Algebra</keyword>
|
||||
<keyword>LU decomposition</keyword>
|
||||
</keywords>
|
||||
<libraries>
|
||||
<library>cublas</library>
|
||||
</libraries>
|
||||
<librarypaths>
|
||||
</librarypaths>
|
||||
<nsight_eclipse>true</nsight_eclipse>
|
||||
<primary_file>simpleCUBLAS_LU.cpp</primary_file>
|
||||
<required_dependencies>
|
||||
<dependency>CUBLAS</dependency>
|
||||
</required_dependencies>
|
||||
<scopes>
|
||||
<scope>1:CUDA Basic Topics</scope>
|
||||
<scope>3:Linear Algebra</scope>
|
||||
</scopes>
|
||||
<sm-arch>sm35</sm-arch>
|
||||
<sm-arch>sm37</sm-arch>
|
||||
<sm-arch>sm50</sm-arch>
|
||||
<sm-arch>sm52</sm-arch>
|
||||
<sm-arch>sm60</sm-arch>
|
||||
<sm-arch>sm61</sm-arch>
|
||||
<sm-arch>sm70</sm-arch>
|
||||
<sm-arch>sm72</sm-arch>
|
||||
<sm-arch>sm75</sm-arch>
|
||||
<sm-arch>sm80</sm-arch>
|
||||
<sm-arch>sm86</sm-arch>
|
||||
<supported_envs>
|
||||
<env>
|
||||
<arch>x86_64</arch>
|
||||
<platform>linux</platform>
|
||||
</env>
|
||||
<env>
|
||||
<platform>windows7</platform>
|
||||
</env>
|
||||
<env>
|
||||
<arch>aarch64</arch>
|
||||
</env>
|
||||
<env>
|
||||
<arch>ppc64le</arch>
|
||||
<platform>linux</platform>
|
||||
</env>
|
||||
</supported_envs>
|
||||
<supported_sm_architectures>
|
||||
<include>all</include>
|
||||
</supported_sm_architectures>
|
||||
<title>Simple CUBLAS LU</title>
|
||||
<type>exe</type>
|
||||
</entry>
|
||||
71
Samples/simpleCUBLAS_LU/README.md
Normal file
71
Samples/simpleCUBLAS_LU/README.md
Normal file
@@ -0,0 +1,71 @@
|
||||
# simpleCUBLAS_LU - Simple CUBLAS LU
|
||||
|
||||
## Description
|
||||
|
||||
CUDA sample demonstrating cuBLAS API cublasDgetrfBatched() for lower-upper (LU) decomposition of a matrix.
|
||||
|
||||
## Key Concepts
|
||||
|
||||
CUBLAS Library, LU decomposition
|
||||
|
||||
## Supported SM Architectures
|
||||
|
||||
[SM 3.5 ](https://developer.nvidia.com/cuda-gpus) [SM 3.7 ](https://developer.nvidia.com/cuda-gpus) [SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus)
|
||||
|
||||
## Supported OSes
|
||||
|
||||
Linux, Windows
|
||||
|
||||
## Supported CPU Architecture
|
||||
|
||||
x86_64, ppc64le, aarch64
|
||||
|
||||
## CUDA APIs involved
|
||||
|
||||
## Dependencies needed to build/run
|
||||
[CUBLAS](../../README.md#cublas)
|
||||
|
||||
## Prerequisites
|
||||
|
||||
Download and install the [CUDA Toolkit 11.3](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
||||
|
||||
## Build and Run
|
||||
|
||||
### Windows
|
||||
The Windows samples are built using the Visual Studio IDE. Solution files (.sln) are provided for each supported version of Visual Studio, using the format:
|
||||
```
|
||||
*_vs<version>.sln - for Visual Studio <version>
|
||||
```
|
||||
Each individual sample has its own set of solution files in its directory:
|
||||
|
||||
To build/examine all the samples at once, the complete solution files should be used. To build/examine a single sample, the individual sample solution files should be used.
|
||||
> **Note:** Some samples require that the Microsoft DirectX SDK (June 2010 or newer) be installed and that the VC++ directory paths are properly set up (**Tools > Options...**). Check DirectX Dependencies section for details."
|
||||
|
||||
### Linux
|
||||
The Linux samples are built using makefiles. To use the makefiles, change the current directory to the sample directory you wish to build, and run make:
|
||||
```
|
||||
$ cd <sample_dir>
|
||||
$ make
|
||||
```
|
||||
The samples makefiles can take advantage of certain options:
|
||||
* **TARGET_ARCH=<arch>** - cross-compile targeting a specific architecture. Allowed architectures are x86_64, ppc64le, aarch64.
|
||||
By default, TARGET_ARCH is set to HOST_ARCH. On a x86_64 machine, not setting TARGET_ARCH is the equivalent of setting TARGET_ARCH=x86_64.<br/>
|
||||
`$ make TARGET_ARCH=x86_64` <br/> `$ make TARGET_ARCH=ppc64le` <br/> `$ make TARGET_ARCH=aarch64` <br/>
|
||||
See [here](http://docs.nvidia.com/cuda/cuda-samples/index.html#cross-samples) for more details.
|
||||
* **dbg=1** - build with debug symbols
|
||||
```
|
||||
$ make dbg=1
|
||||
```
|
||||
* **SMS="A B ..."** - override the SM architectures for which the sample will be built, where `"A B ..."` is a space-delimited list of SM architectures. For example, to generate SASS for SM 50 and SM 60, use `SMS="50 60"`.
|
||||
```
|
||||
$ make SMS="50 60"
|
||||
```
|
||||
|
||||
* **HOST_COMPILER=<host_compiler>** - override the default g++ host compiler. See the [Linux Installation Guide](http://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#system-requirements) for a list of supported host compilers.
|
||||
```
|
||||
$ make HOST_COMPILER=g++
|
||||
```
|
||||
|
||||
## References (for more details)
|
||||
|
||||
417
Samples/simpleCUBLAS_LU/simpleCUBLAS_LU.cpp
Normal file
417
Samples/simpleCUBLAS_LU/simpleCUBLAS_LU.cpp
Normal file
@@ -0,0 +1,417 @@
|
||||
/* Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved.
|
||||
*
|
||||
* Redistribution and use in source and binary forms, with or without
|
||||
* modification, are permitted provided that the following conditions
|
||||
* are met:
|
||||
* * Redistributions of source code must retain the above copyright
|
||||
* notice, this list of conditions and the following disclaimer.
|
||||
* * Redistributions in binary form must reproduce the above copyright
|
||||
* notice, this list of conditions and the following disclaimer in the
|
||||
* documentation and/or other materials provided with the distribution.
|
||||
* * Neither the name of NVIDIA CORPORATION nor the names of its
|
||||
* contributors may be used to endorse or promote products derived
|
||||
* from this software without specific prior written permission.
|
||||
*
|
||||
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
|
||||
* EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
||||
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
|
||||
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
|
||||
* CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
|
||||
* EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
|
||||
* PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
|
||||
* PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
|
||||
* OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
||||
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
||||
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
*/
|
||||
|
||||
/*
|
||||
* This example demonstrates how to use the cuBLAS library API
|
||||
* for lower-upper (LU) decomposition of a matrix. LU decomposition
|
||||
* factors a matrix as the product of upper triangular matrix and
|
||||
* lower trianglular matrix.
|
||||
*
|
||||
* https://en.wikipedia.org/wiki/LU_decomposition
|
||||
*
|
||||
* This sample uses 10000 matrices of size 4x4 and performs
|
||||
* LU decomposition of them using batched decomposition API
|
||||
* of cuBLAS library. To test the correctness of upper and lower
|
||||
* matrices generated, they are multiplied and compared with the
|
||||
* original input matrix.
|
||||
*
|
||||
*/
|
||||
|
||||
#include <stdio.h>
|
||||
#include <stdlib.h>
|
||||
|
||||
// cuda libraries and helpers
|
||||
#include <cublas_v2.h>
|
||||
#include <cuda_runtime.h>
|
||||
#include <helper_cuda.h>
|
||||
|
||||
// configurable parameters
|
||||
// dimension of matrix
|
||||
#define N 4
|
||||
#define BATCH_SIZE 10000
|
||||
|
||||
// use double precision data type
|
||||
#define DOUBLE_PRECISION /* comment this to use single precision */
|
||||
#ifdef DOUBLE_PRECISION
|
||||
#define DATA_TYPE double
|
||||
#define MAX_ERROR 1e-15
|
||||
#else
|
||||
#define DATA_TYPE float
|
||||
#define MAX_ERROR 1e-6
|
||||
#endif /* DOUBLE_PRCISION */
|
||||
|
||||
// use pivot vector while decomposing
|
||||
#define PIVOT /* comment this to disable pivot use */
|
||||
|
||||
// helper functions
|
||||
|
||||
// wrapper around cublas<t>getrfBatched()
|
||||
cublasStatus_t cublasXgetrfBatched(cublasHandle_t handle, int n,
|
||||
DATA_TYPE* const A[], int lda, int* P,
|
||||
int* info, int batchSize) {
|
||||
#ifdef DOUBLE_PRECISION
|
||||
return cublasDgetrfBatched(handle, n, A, lda, P, info, batchSize);
|
||||
#else
|
||||
return cublasSgetrfBatched(handle, n, A, lda, P, info, batchSize);
|
||||
#endif
|
||||
}
|
||||
|
||||
// wrapper around malloc
|
||||
// clears the allocated memory to 0
|
||||
// terminates the program if malloc fails
|
||||
void* xmalloc(size_t size) {
|
||||
void* ptr = malloc(size);
|
||||
if (ptr == NULL) {
|
||||
printf("> ERROR: malloc for size %zu failed..\n", size);
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
memset(ptr, 0, size);
|
||||
return ptr;
|
||||
}
|
||||
|
||||
// initalize identity matrix
|
||||
void initIdentityMatrix(DATA_TYPE* mat) {
|
||||
// clear the matrix
|
||||
memset(mat, 0, N * N * sizeof(DATA_TYPE));
|
||||
|
||||
// set all diagonals to 1
|
||||
for (int i = 0; i < N; i++) {
|
||||
mat[(i * N) + i] = 1.0;
|
||||
}
|
||||
}
|
||||
|
||||
// initialize matrix with all elements as 0
|
||||
void initZeroMatrix(DATA_TYPE* mat) {
|
||||
memset(mat, 0, N * N * sizeof(DATA_TYPE));
|
||||
}
|
||||
|
||||
// fill random value in column-major matrix
|
||||
void initRandomMatrix(DATA_TYPE* mat) {
|
||||
for (int i = 0; i < N; i++) {
|
||||
for (int j = 0; j < N; j++) {
|
||||
mat[(j * N) + i] =
|
||||
(DATA_TYPE)1.0 + ((DATA_TYPE)rand() / (DATA_TYPE)RAND_MAX);
|
||||
}
|
||||
}
|
||||
|
||||
// diagonal dominant matrix to insure it is invertible matrix
|
||||
for (int i = 0; i < N; i++) {
|
||||
mat[(i * N) + i] += (DATA_TYPE)N;
|
||||
}
|
||||
}
|
||||
|
||||
// print column-major matrix
|
||||
void printMatrix(DATA_TYPE* mat) {
|
||||
for (int i = 0; i < N; i++) {
|
||||
for (int j = 0; j < N; j++) {
|
||||
printf("%20.16f ", mat[(j * N) + i]);
|
||||
}
|
||||
printf("\n");
|
||||
}
|
||||
printf("\n");
|
||||
}
|
||||
|
||||
// matrix mulitplication
|
||||
void matrixMultiply(DATA_TYPE* res, DATA_TYPE* mat1, DATA_TYPE* mat2) {
|
||||
initZeroMatrix(res);
|
||||
|
||||
for (int i = 0; i < N; i++) {
|
||||
for (int j = 0; j < N; j++) {
|
||||
for (int k = 0; k < N; k++) {
|
||||
res[(j * N) + i] += mat1[(k * N) + i] * mat2[(j * N) + k];
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// check matrix equality
|
||||
bool checkRelativeError(DATA_TYPE* mat1, DATA_TYPE* mat2, DATA_TYPE maxError) {
|
||||
DATA_TYPE err = (DATA_TYPE)0.0;
|
||||
DATA_TYPE refNorm = (DATA_TYPE)0.0;
|
||||
DATA_TYPE relError = (DATA_TYPE)0.0;
|
||||
DATA_TYPE relMaxError = (DATA_TYPE)0.0;
|
||||
|
||||
for (int i = 0; i < N * N; i++) {
|
||||
refNorm = abs(mat1[i]);
|
||||
err = abs(mat1[i] - mat2[i]);
|
||||
|
||||
if (refNorm != 0.0 && err > 0.0) {
|
||||
relError = err / refNorm;
|
||||
relMaxError = MAX(relMaxError, relError);
|
||||
}
|
||||
|
||||
if (relMaxError > maxError) return false;
|
||||
}
|
||||
return true;
|
||||
}
|
||||
|
||||
// decode lower and upper matrix from single matrix
|
||||
// returned by getrfBatched()
|
||||
void getLUdecoded(DATA_TYPE* mat, DATA_TYPE* L, DATA_TYPE* U) {
|
||||
// init L as identity matrix
|
||||
initIdentityMatrix(L);
|
||||
|
||||
// copy lower triangular values from mat to L (skip diagonal)
|
||||
for (int i = 0; i < N; i++) {
|
||||
for (int j = 0; j < i; j++) {
|
||||
L[(j * N) + i] = mat[(j * N) + i];
|
||||
}
|
||||
}
|
||||
|
||||
// init U as all zero
|
||||
initZeroMatrix(U);
|
||||
|
||||
// copy upper triangular values from mat to U
|
||||
for (int i = 0; i < N; i++) {
|
||||
for (int j = i; j < N; j++) {
|
||||
U[(j * N) + i] = mat[(j * N) + i];
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// generate permutation matrix from pivot vector
|
||||
void getPmatFromPivot(DATA_TYPE* Pmat, int* P) {
|
||||
int pivot[N];
|
||||
|
||||
// pivot vector in base-1
|
||||
// convert it to base-0
|
||||
for (int i = 0; i < N; i++) {
|
||||
P[i]--;
|
||||
}
|
||||
|
||||
// generate permutation vector from pivot
|
||||
// initialize pivot with identity sequence
|
||||
for (int k = 0; k < N; k++) {
|
||||
pivot[k] = k;
|
||||
}
|
||||
|
||||
// swap the indices according to pivot vector
|
||||
for (int k = 0; k < N; k++) {
|
||||
int q = P[k];
|
||||
|
||||
// swap pivot(k) and pivot(q)
|
||||
int s = pivot[k];
|
||||
int t = pivot[q];
|
||||
pivot[k] = t;
|
||||
pivot[q] = s;
|
||||
}
|
||||
|
||||
// generate permutation matrix from pivot vector
|
||||
initZeroMatrix(Pmat);
|
||||
for (int i = 0; i < N; i++) {
|
||||
int j = pivot[i];
|
||||
Pmat[(j * N) + i] = (DATA_TYPE)1.0;
|
||||
}
|
||||
}
|
||||
|
||||
int main(int argc, char** argv) {
|
||||
// cuBLAS variables
|
||||
cublasStatus_t status;
|
||||
cublasHandle_t handle;
|
||||
|
||||
// host variables
|
||||
size_t matSize = N * N * sizeof(DATA_TYPE);
|
||||
|
||||
DATA_TYPE* h_AarrayInput;
|
||||
DATA_TYPE* h_AarrayOutput;
|
||||
DATA_TYPE* h_ptr_array[BATCH_SIZE];
|
||||
|
||||
int* h_pivotArray;
|
||||
int* h_infoArray;
|
||||
|
||||
// device variables
|
||||
DATA_TYPE* d_Aarray;
|
||||
DATA_TYPE** d_ptr_array;
|
||||
|
||||
int* d_pivotArray;
|
||||
int* d_infoArray;
|
||||
|
||||
int err_count = 0;
|
||||
|
||||
// seed the rand() function with time
|
||||
srand(12345);
|
||||
|
||||
// find cuda device
|
||||
printf("> initializing..\n");
|
||||
int dev = findCudaDevice(argc, (const char**)argv);
|
||||
if (dev == -1) {
|
||||
return (EXIT_FAILURE);
|
||||
}
|
||||
|
||||
// initialize cuBLAS
|
||||
status = cublasCreate(&handle);
|
||||
if (status != CUBLAS_STATUS_SUCCESS) {
|
||||
printf("> ERROR: cuBLAS initialization failed..\n");
|
||||
return (EXIT_FAILURE);
|
||||
}
|
||||
|
||||
#ifdef DOUBLE_PRECISION
|
||||
printf("> using DOUBLE precision..\n");
|
||||
#else
|
||||
printf("> using SINGLE precision..\n");
|
||||
#endif
|
||||
|
||||
#ifdef PIVOT
|
||||
printf("> pivot ENABLED..\n");
|
||||
#else
|
||||
printf("> pivot DISABLED..\n");
|
||||
#endif
|
||||
|
||||
// allocate memory for host variables
|
||||
h_AarrayInput = (DATA_TYPE*)xmalloc(BATCH_SIZE * matSize);
|
||||
h_AarrayOutput = (DATA_TYPE*)xmalloc(BATCH_SIZE * matSize);
|
||||
|
||||
h_pivotArray = (int*)xmalloc(N * BATCH_SIZE * sizeof(int));
|
||||
h_infoArray = (int*)xmalloc(BATCH_SIZE * sizeof(int));
|
||||
|
||||
// allocate memory for device variables
|
||||
checkCudaErrors(cudaMalloc((void**)&d_Aarray, BATCH_SIZE * matSize));
|
||||
checkCudaErrors(
|
||||
cudaMalloc((void**)&d_pivotArray, N * BATCH_SIZE * sizeof(int)));
|
||||
checkCudaErrors(cudaMalloc((void**)&d_infoArray, BATCH_SIZE * sizeof(int)));
|
||||
checkCudaErrors(
|
||||
cudaMalloc((void**)&d_ptr_array, BATCH_SIZE * sizeof(DATA_TYPE*)));
|
||||
|
||||
// fill matrix with random data
|
||||
printf("> generating random matrices..\n");
|
||||
for (int i = 0; i < BATCH_SIZE; i++) {
|
||||
initRandomMatrix(h_AarrayInput + (i * N * N));
|
||||
}
|
||||
|
||||
// copy data to device from host
|
||||
printf("> copying data from host memory to GPU memory..\n");
|
||||
checkCudaErrors(cudaMemcpy(d_Aarray, h_AarrayInput, BATCH_SIZE * matSize,
|
||||
cudaMemcpyHostToDevice));
|
||||
|
||||
// create pointer array for matrices
|
||||
for (int i = 0; i < BATCH_SIZE; i++) h_ptr_array[i] = d_Aarray + (i * N * N);
|
||||
|
||||
// copy pointer array to device memory
|
||||
checkCudaErrors(cudaMemcpy(d_ptr_array, h_ptr_array,
|
||||
BATCH_SIZE * sizeof(DATA_TYPE*),
|
||||
cudaMemcpyHostToDevice));
|
||||
|
||||
// perform LU decomposition
|
||||
printf("> performing LU decomposition..\n");
|
||||
#ifdef PIVOT
|
||||
status = cublasXgetrfBatched(handle, N, d_ptr_array, N, d_pivotArray,
|
||||
d_infoArray, BATCH_SIZE);
|
||||
#else
|
||||
status = cublasXgetrfBatched(handle, N, d_ptr_array, N, NULL, d_infoArray,
|
||||
BATCH_SIZE);
|
||||
#endif /* PIVOT */
|
||||
if (status != CUBLAS_STATUS_SUCCESS) {
|
||||
printf("> ERROR: cublasDgetrfBatched() failed with error %s..\n",
|
||||
_cudaGetErrorEnum(status));
|
||||
return (EXIT_FAILURE);
|
||||
}
|
||||
|
||||
// copy data to host from device
|
||||
printf("> copying data from GPU memory to host memory..\n");
|
||||
checkCudaErrors(cudaMemcpy(h_AarrayOutput, d_Aarray, BATCH_SIZE * matSize,
|
||||
cudaMemcpyDeviceToHost));
|
||||
checkCudaErrors(cudaMemcpy(h_infoArray, d_infoArray, BATCH_SIZE * sizeof(int),
|
||||
cudaMemcpyDeviceToHost));
|
||||
#ifdef PIVOT
|
||||
checkCudaErrors(cudaMemcpy(h_pivotArray, d_pivotArray,
|
||||
N * BATCH_SIZE * sizeof(int),
|
||||
cudaMemcpyDeviceToHost));
|
||||
#endif /* PIVOT */
|
||||
|
||||
// verify the result
|
||||
printf("> verifying the result..\n");
|
||||
for (int i = 0; i < BATCH_SIZE; i++) {
|
||||
if (h_infoArray[i] == 0) {
|
||||
DATA_TYPE* A = h_AarrayInput + (i * N * N);
|
||||
DATA_TYPE* LU = h_AarrayOutput + (i * N * N);
|
||||
DATA_TYPE L[N * N];
|
||||
DATA_TYPE U[N * N];
|
||||
getLUdecoded(LU, L, U);
|
||||
|
||||
// test P * A = L * U
|
||||
int* P = h_pivotArray + (i * N);
|
||||
DATA_TYPE Pmat[N * N];
|
||||
#ifdef PIVOT
|
||||
getPmatFromPivot(Pmat, P);
|
||||
#else
|
||||
initIdentityMatrix(Pmat);
|
||||
#endif /* PIVOT */
|
||||
|
||||
// perform matrix multiplication
|
||||
DATA_TYPE PxA[N * N];
|
||||
DATA_TYPE LxU[N * N];
|
||||
matrixMultiply(PxA, Pmat, A);
|
||||
matrixMultiply(LxU, L, U);
|
||||
|
||||
// check for equality of matrices
|
||||
if (!checkRelativeError(PxA, LxU, (DATA_TYPE)MAX_ERROR)) {
|
||||
printf("> ERROR: accuracy check failed for matrix number %05d..\n",
|
||||
i + 1);
|
||||
err_count++;
|
||||
}
|
||||
|
||||
} else if (h_infoArray[i] > 0) {
|
||||
printf(
|
||||
"> execution for matrix %05d is successful, but U is singular and "
|
||||
"U(%d,%d) = 0..\n",
|
||||
i + 1, h_infoArray[i] - 1, h_infoArray[i] - 1);
|
||||
} else // (h_infoArray[i] < 0)
|
||||
{
|
||||
printf("> ERROR: matrix %05d have an illegal value at index %d = %lf..\n",
|
||||
i + 1, -h_infoArray[i],
|
||||
*(h_AarrayInput + (i * N * N) + (-h_infoArray[i])));
|
||||
}
|
||||
}
|
||||
|
||||
// free device variables
|
||||
checkCudaErrors(cudaFree(d_ptr_array));
|
||||
checkCudaErrors(cudaFree(d_infoArray));
|
||||
checkCudaErrors(cudaFree(d_pivotArray));
|
||||
checkCudaErrors(cudaFree(d_Aarray));
|
||||
|
||||
// free host variables
|
||||
if (h_infoArray) free(h_infoArray);
|
||||
if (h_pivotArray) free(h_pivotArray);
|
||||
if (h_AarrayOutput) free(h_AarrayOutput);
|
||||
if (h_AarrayInput) free(h_AarrayInput);
|
||||
|
||||
// destroy cuBLAS handle
|
||||
status = cublasDestroy(handle);
|
||||
if (status != CUBLAS_STATUS_SUCCESS) {
|
||||
printf("> ERROR: cuBLAS uninitialization failed..\n");
|
||||
return (EXIT_FAILURE);
|
||||
}
|
||||
|
||||
if (err_count > 0) {
|
||||
printf("> TEST FAILED for %d matrices, with precision: %g\n", err_count,
|
||||
MAX_ERROR);
|
||||
return (EXIT_FAILURE);
|
||||
}
|
||||
|
||||
printf("> TEST SUCCESSFUL, with precision: %g\n", MAX_ERROR);
|
||||
return (EXIT_SUCCESS);
|
||||
}
|
||||
20
Samples/simpleCUBLAS_LU/simpleCUBLAS_LU_vs2017.sln
Normal file
20
Samples/simpleCUBLAS_LU/simpleCUBLAS_LU_vs2017.sln
Normal file
@@ -0,0 +1,20 @@
|
||||
|
||||
Microsoft Visual Studio Solution File, Format Version 12.00
|
||||
# Visual Studio 2017
|
||||
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "simpleCUBLAS_LU", "simpleCUBLAS_LU_vs2017.vcxproj", "{997E0757-EA74-4A4E-A0FC-47D8C8831A15}"
|
||||
EndProject
|
||||
Global
|
||||
GlobalSection(SolutionConfigurationPlatforms) = preSolution
|
||||
Debug|x64 = Debug|x64
|
||||
Release|x64 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(ProjectConfigurationPlatforms) = postSolution
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.ActiveCfg = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.Build.0 = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.ActiveCfg = Release|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.Build.0 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(SolutionProperties) = preSolution
|
||||
HideSolutionNode = FALSE
|
||||
EndGlobalSection
|
||||
EndGlobal
|
||||
113
Samples/simpleCUBLAS_LU/simpleCUBLAS_LU_vs2017.vcxproj
Normal file
113
Samples/simpleCUBLAS_LU/simpleCUBLAS_LU_vs2017.vcxproj
Normal file
@@ -0,0 +1,113 @@
|
||||
<?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>simpleCUBLAS_LU_vs2017</RootNamespace>
|
||||
<ProjectName>simpleCUBLAS_LU</ProjectName>
|
||||
<CudaToolkitCustomDir />
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(WindowsTargetPlatformVersion)'==''">
|
||||
<LatestTargetPlatformVersion>$([Microsoft.Build.Utilities.ToolLocationHelper]::GetLatestSDKTargetPlatformVersion('Windows', '10.0'))</LatestTargetPlatformVersion>
|
||||
<WindowsTargetPlatformVersion Condition="'$(WindowsTargetPlatformVersion)' == ''">$(LatestTargetPlatformVersion)</WindowsTargetPlatformVersion>
|
||||
<TargetPlatformVersion>$(WindowsTargetPlatformVersion)</TargetPlatformVersion>
|
||||
</PropertyGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
|
||||
<PropertyGroup>
|
||||
<ConfigurationType>Application</ConfigurationType>
|
||||
<CharacterSet>MultiByte</CharacterSet>
|
||||
<PlatformToolset>v141</PlatformToolset>
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(Configuration)'=='Debug'">
|
||||
<UseDebugLibraries>true</UseDebugLibraries>
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(Configuration)'=='Release'">
|
||||
<WholeProgramOptimization>true</WholeProgramOptimization>
|
||||
</PropertyGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||
<ImportGroup Label="ExtensionSettings">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.3.props" />
|
||||
</ImportGroup>
|
||||
<ImportGroup Label="PropertySheets">
|
||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||
</ImportGroup>
|
||||
<PropertyGroup Label="UserMacros" />
|
||||
<PropertyGroup>
|
||||
<IntDir>$(Platform)/$(Configuration)/</IntDir>
|
||||
<IncludePath>$(IncludePath)</IncludePath>
|
||||
<CodeAnalysisRuleSet>AllRules.ruleset</CodeAnalysisRuleSet>
|
||||
<CodeAnalysisRules />
|
||||
<CodeAnalysisRuleAssemblies />
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(Platform)'=='x64'">
|
||||
<OutDir>../../bin/win64/$(Configuration)/</OutDir>
|
||||
</PropertyGroup>
|
||||
<ItemDefinitionGroup>
|
||||
<ClCompile>
|
||||
<WarningLevel>Level3</WarningLevel>
|
||||
<PreprocessorDefinitions>WIN32;_MBCS;%(PreprocessorDefinitions)</PreprocessorDefinitions>
|
||||
<AdditionalIncludeDirectories>./;$(CudaToolkitDir)/include;../../Common;$(CudaToolkitIncludeDir);</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cublas.lib;cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/simpleCUBLAS_LU.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
|
||||
<Include>./;../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
<AdditionalOptions>--threads 0</AdditionalOptions>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Configuration)'=='Debug'">
|
||||
<ClCompile>
|
||||
<Optimization>Disabled</Optimization>
|
||||
<RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>true</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>Default</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MTd</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Configuration)'=='Release'">
|
||||
<ClCompile>
|
||||
<Optimization>MaxSpeed</Optimization>
|
||||
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MT</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<ClCompile Include="simpleCUBLAS_LU.cpp" />
|
||||
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.3.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
20
Samples/simpleCUBLAS_LU/simpleCUBLAS_LU_vs2019.sln
Normal file
20
Samples/simpleCUBLAS_LU/simpleCUBLAS_LU_vs2019.sln
Normal file
@@ -0,0 +1,20 @@
|
||||
|
||||
Microsoft Visual Studio Solution File, Format Version 12.00
|
||||
# Visual Studio 2019
|
||||
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "simpleCUBLAS_LU", "simpleCUBLAS_LU_vs2019.vcxproj", "{997E0757-EA74-4A4E-A0FC-47D8C8831A15}"
|
||||
EndProject
|
||||
Global
|
||||
GlobalSection(SolutionConfigurationPlatforms) = preSolution
|
||||
Debug|x64 = Debug|x64
|
||||
Release|x64 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(ProjectConfigurationPlatforms) = postSolution
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.ActiveCfg = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.Build.0 = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.ActiveCfg = Release|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.Build.0 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(SolutionProperties) = preSolution
|
||||
HideSolutionNode = FALSE
|
||||
EndGlobalSection
|
||||
EndGlobal
|
||||
109
Samples/simpleCUBLAS_LU/simpleCUBLAS_LU_vs2019.vcxproj
Normal file
109
Samples/simpleCUBLAS_LU/simpleCUBLAS_LU_vs2019.vcxproj
Normal file
@@ -0,0 +1,109 @@
|
||||
<?xml version="1.0" encoding="utf-8"?>
|
||||
<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
|
||||
<PropertyGroup>
|
||||
<CUDAPropsPath Condition="'$(CUDAPropsPath)'==''">$(VCTargetsPath)\BuildCustomizations</CUDAPropsPath>
|
||||
</PropertyGroup>
|
||||
<ItemGroup Label="ProjectConfigurations">
|
||||
<ProjectConfiguration Include="Debug|x64">
|
||||
<Configuration>Debug</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
<ProjectConfiguration Include="Release|x64">
|
||||
<Configuration>Release</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
</ItemGroup>
|
||||
<PropertyGroup Label="Globals">
|
||||
<ProjectGuid>{997E0757-EA74-4A4E-A0FC-47D8C8831A15}</ProjectGuid>
|
||||
<RootNamespace>simpleCUBLAS_LU_vs2019</RootNamespace>
|
||||
<ProjectName>simpleCUBLAS_LU</ProjectName>
|
||||
<CudaToolkitCustomDir />
|
||||
</PropertyGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
|
||||
<PropertyGroup>
|
||||
<ConfigurationType>Application</ConfigurationType>
|
||||
<CharacterSet>MultiByte</CharacterSet>
|
||||
<PlatformToolset>v142</PlatformToolset>
|
||||
<WindowsTargetPlatformVersion>10.0</WindowsTargetPlatformVersion>
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(Configuration)'=='Debug'">
|
||||
<UseDebugLibraries>true</UseDebugLibraries>
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(Configuration)'=='Release'">
|
||||
<WholeProgramOptimization>true</WholeProgramOptimization>
|
||||
</PropertyGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
|
||||
<ImportGroup Label="ExtensionSettings">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.3.props" />
|
||||
</ImportGroup>
|
||||
<ImportGroup Label="PropertySheets">
|
||||
<Import Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" />
|
||||
</ImportGroup>
|
||||
<PropertyGroup Label="UserMacros" />
|
||||
<PropertyGroup>
|
||||
<IntDir>$(Platform)/$(Configuration)/</IntDir>
|
||||
<IncludePath>$(IncludePath)</IncludePath>
|
||||
<CodeAnalysisRuleSet>AllRules.ruleset</CodeAnalysisRuleSet>
|
||||
<CodeAnalysisRules />
|
||||
<CodeAnalysisRuleAssemblies />
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(Platform)'=='x64'">
|
||||
<OutDir>../../bin/win64/$(Configuration)/</OutDir>
|
||||
</PropertyGroup>
|
||||
<ItemDefinitionGroup>
|
||||
<ClCompile>
|
||||
<WarningLevel>Level3</WarningLevel>
|
||||
<PreprocessorDefinitions>WIN32;_MBCS;%(PreprocessorDefinitions)</PreprocessorDefinitions>
|
||||
<AdditionalIncludeDirectories>./;$(CudaToolkitDir)/include;../../Common;$(CudaToolkitIncludeDir);</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cublas.lib;cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/simpleCUBLAS_LU.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
|
||||
<Include>./;../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
<AdditionalOptions>--threads 0</AdditionalOptions>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Configuration)'=='Debug'">
|
||||
<ClCompile>
|
||||
<Optimization>Disabled</Optimization>
|
||||
<RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>true</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>Default</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MTd</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Configuration)'=='Release'">
|
||||
<ClCompile>
|
||||
<Optimization>MaxSpeed</Optimization>
|
||||
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MT</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<ClCompile Include="simpleCUBLAS_LU.cpp" />
|
||||
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.3.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
Reference in New Issue
Block a user