mirror of
https://github.com/NVIDIA/cuda-samples.git
synced 2026-10-23 21:28:17 +08:00
add and update samples for CUDA 11.6
This commit is contained in:
Rutwik Choughule
18
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/.vscode/c_cpp_properties.json
vendored
Normal file
@@ -0,0 +1,18 @@
|
||||
{
|
||||
"configurations": [
|
||||
{
|
||||
"name": "Linux",
|
||||
"includePath": [
|
||||
"${workspaceFolder}/**",
|
||||
"${workspaceFolder}/../../../Common"
|
||||
],
|
||||
"defines": [],
|
||||
"compilerPath": "/usr/local/cuda/bin/nvcc",
|
||||
"cStandard": "gnu17",
|
||||
"cppStandard": "gnu++14",
|
||||
"intelliSenseMode": "linux-gcc-x64",
|
||||
"configurationProvider": "ms-vscode.makefile-tools"
|
||||
}
|
||||
],
|
||||
"version": 4
|
||||
}
|
||||
7
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/.vscode/extensions.json
vendored
Normal file
7
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
||||
{
|
||||
"recommendations": [
|
||||
"nvidia.nsight-vscode-edition",
|
||||
"ms-vscode.cpptools",
|
||||
"ms-vscode.makefile-tools"
|
||||
]
|
||||
}
|
||||
10
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/.vscode/launch.json
vendored
Normal file
10
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
||||
{
|
||||
"configurations": [
|
||||
{
|
||||
"name": "CUDA C++: Launch",
|
||||
"type": "cuda-gdb",
|
||||
"request": "launch",
|
||||
"program": "${workspaceFolder}/simpleCUFFT_MGPU"
|
||||
}
|
||||
]
|
||||
}
|
||||
15
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/.vscode/tasks.json
vendored
Normal file
15
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
||||
{
|
||||
"version": "2.0.0",
|
||||
"tasks": [
|
||||
{
|
||||
"label": "sample",
|
||||
"type": "shell",
|
||||
"command": "make dbg=1",
|
||||
"problemMatcher": ["$nvcc"],
|
||||
"group": {
|
||||
"kind": "build",
|
||||
"isDefault": true
|
||||
}
|
||||
}
|
||||
]
|
||||
}
|
||||
342
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/Makefile
Normal file
342
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/Makefile
Normal file
@@ -0,0 +1,342 @@
|
||||
################################################################################
|
||||
# Copyright (c) 2022, 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 += -D_QNX_SOURCE
|
||||
NVCCFLAGS += --qpp-config 8.3.0,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
|
||||
|
||||
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 $(TARGET_ARCH),armv7l aarch64 sbsa))
|
||||
SMS ?= 53 61 70 72 75 80 86 87
|
||||
else
|
||||
SMS ?= 35 37 50 52 60 61 70 75 80 86
|
||||
endif
|
||||
|
||||
ifeq ($(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 --std=c++11
|
||||
|
||||
LIBRARIES += -lcufft
|
||||
|
||||
ifeq ($(SAMPLE_ENABLED),0)
|
||||
EXEC ?= @echo "[@]"
|
||||
endif
|
||||
|
||||
################################################################################
|
||||
|
||||
# Target rules
|
||||
all: build
|
||||
|
||||
build: simpleCUFFT_MGPU
|
||||
|
||||
check.deps:
|
||||
ifeq ($(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
|
||||
|
||||
simpleCUFFT_MGPU.o:simpleCUFFT_MGPU.cu
|
||||
$(EXEC) $(NVCC) $(INCLUDES) $(ALL_CCFLAGS) $(GENCODE_FLAGS) -o $@ -c $<
|
||||
|
||||
simpleCUFFT_MGPU: simpleCUFFT_MGPU.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
|
||||
$(EXEC) ./simpleCUFFT_MGPU
|
||||
|
||||
testrun: build
|
||||
|
||||
clean:
|
||||
rm -f simpleCUFFT_MGPU simpleCUFFT_MGPU.o
|
||||
rm -rf ../../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)/simpleCUFFT_MGPU
|
||||
|
||||
clobber: clean
|
||||
82
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/NsightEclipse.xml
Normal file
82
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/NsightEclipse.xml
Normal file
@@ -0,0 +1,82 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<!DOCTYPE entry SYSTEM "SamplesInfo.dtd">
|
||||
<entry>
|
||||
<name>simpleCUFFT_MGPU</name>
|
||||
<cuda_api_list>
|
||||
<toolkit>cudaGetDeviceCount</toolkit>
|
||||
<toolkit>cudaDeviceSynchronize</toolkit>
|
||||
<toolkit>cudaSetDevice</toolkit>
|
||||
<toolkit>cudaXtFree</toolkit>
|
||||
<toolkit>cudaGetDeviceProperties</toolkit>
|
||||
</cuda_api_list>
|
||||
<description><![CDATA[Example of using CUFFT. In this example, CUFFT is used to compute the 1D-convolution of some signal with some filter by transforming both into frequency domain, multiplying them together, and transforming the signal back to time domain on Multiple GPU.]]></description>
|
||||
<devicecompilation>whole</devicecompilation>
|
||||
<includepaths>
|
||||
<path>./</path>
|
||||
<path>../</path>
|
||||
<path>../../../Common</path>
|
||||
</includepaths>
|
||||
<keyconcepts>
|
||||
<concept level="basic">Image Processing</concept>
|
||||
<concept level="basic">CUFFT Library</concept>
|
||||
</keyconcepts>
|
||||
<keywords>
|
||||
<keyword>CUDA</keyword>
|
||||
<keyword>CUFFT</keyword>
|
||||
</keywords>
|
||||
<libraries>
|
||||
<library>cufft</library>
|
||||
</libraries>
|
||||
<librarypaths>
|
||||
</librarypaths>
|
||||
<nsight_eclipse>true</nsight_eclipse>
|
||||
<primary_file>simpleCUFFT_MGPU.cu</primary_file>
|
||||
<required_dependencies>
|
||||
<dependency>CUFFT</dependency>
|
||||
</required_dependencies>
|
||||
<scopes>
|
||||
<scope>1:CUDA Basic Topics</scope>
|
||||
<scope>2:Image Processing</scope>
|
||||
</scopes>
|
||||
<sm-arch>sm35</sm-arch>
|
||||
<sm-arch>sm37</sm-arch>
|
||||
<sm-arch>sm50</sm-arch>
|
||||
<sm-arch>sm52</sm-arch>
|
||||
<sm-arch>sm53</sm-arch>
|
||||
<sm-arch>sm60</sm-arch>
|
||||
<sm-arch>sm61</sm-arch>
|
||||
<sm-arch>sm70</sm-arch>
|
||||
<sm-arch>sm72</sm-arch>
|
||||
<sm-arch>sm75</sm-arch>
|
||||
<sm-arch>sm80</sm-arch>
|
||||
<sm-arch>sm86</sm-arch>
|
||||
<sm-arch>sm87</sm-arch>
|
||||
<supported_envs>
|
||||
<env>
|
||||
<arch>x86_64</arch>
|
||||
<platform>linux</platform>
|
||||
</env>
|
||||
<env>
|
||||
<platform>windows7</platform>
|
||||
</env>
|
||||
<env>
|
||||
<arch>x86_64</arch>
|
||||
<platform>macosx</platform>
|
||||
</env>
|
||||
<env>
|
||||
<arch>arm</arch>
|
||||
</env>
|
||||
<env>
|
||||
<arch>sbsa</arch>
|
||||
</env>
|
||||
<env>
|
||||
<arch>ppc64le</arch>
|
||||
<platform>linux</platform>
|
||||
</env>
|
||||
</supported_envs>
|
||||
<supported_sm_architectures>
|
||||
<include>all</include>
|
||||
</supported_sm_architectures>
|
||||
<title>Simple CUFFT_MGPU</title>
|
||||
<type>exe</type>
|
||||
</entry>
|
||||
74
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/README.md
Normal file
74
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/README.md
Normal file
@@ -0,0 +1,74 @@
|
||||
# simpleCUFFT_MGPU - Simple CUFFT_MGPU
|
||||
|
||||
## Description
|
||||
|
||||
Example of using CUFFT. In this example, CUFFT is used to compute the 1D-convolution of some signal with some filter by transforming both into frequency domain, multiplying them together, and transforming the signal back to time domain on Multiple GPU.
|
||||
|
||||
## Key Concepts
|
||||
|
||||
Image Processing, CUFFT Library
|
||||
|
||||
## Supported SM Architectures
|
||||
|
||||
[SM 3.5 ](https://developer.nvidia.com/cuda-gpus) [SM 3.7 ](https://developer.nvidia.com/cuda-gpus) [SM 5.0 ](https://developer.nvidia.com/cuda-gpus) [SM 5.2 ](https://developer.nvidia.com/cuda-gpus) [SM 5.3 ](https://developer.nvidia.com/cuda-gpus) [SM 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus) [SM 8.7 ](https://developer.nvidia.com/cuda-gpus)
|
||||
|
||||
## Supported OSes
|
||||
|
||||
Linux, Windows
|
||||
|
||||
## Supported CPU Architecture
|
||||
|
||||
x86_64, ppc64le, armv7l
|
||||
|
||||
## CUDA APIs involved
|
||||
|
||||
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
|
||||
cudaGetDeviceCount, cudaDeviceSynchronize, cudaSetDevice, cudaXtFree, cudaGetDeviceProperties
|
||||
|
||||
## Dependencies needed to build/run
|
||||
[CUFFT](../../README.md#cufft)
|
||||
|
||||
## Prerequisites
|
||||
|
||||
Download and install the [CUDA Toolkit 11.6](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||
Make sure the dependencies mentioned in [Dependencies]() section above are installed.
|
||||
|
||||
## Build and Run
|
||||
|
||||
### Windows
|
||||
The Windows samples are built using the Visual Studio IDE. Solution files (.sln) are provided for each supported version of Visual Studio, using the format:
|
||||
```
|
||||
*_vs<version>.sln - for Visual Studio <version>
|
||||
```
|
||||
Each individual sample has its own set of solution files in its directory:
|
||||
|
||||
To build/examine all the samples at once, the complete solution files should be used. To build/examine a single sample, the individual sample solution files should be used.
|
||||
> **Note:** Some samples require that the Microsoft DirectX SDK (June 2010 or newer) be installed and that the VC++ directory paths are properly set up (**Tools > Options...**). Check DirectX Dependencies section for details."
|
||||
|
||||
### Linux
|
||||
The Linux samples are built using makefiles. To use the makefiles, change the current directory to the sample directory you wish to build, and run make:
|
||||
```
|
||||
$ cd <sample_dir>
|
||||
$ make
|
||||
```
|
||||
The samples makefiles can take advantage of certain options:
|
||||
* **TARGET_ARCH=<arch>** - cross-compile targeting a specific architecture. Allowed architectures are x86_64, ppc64le, armv7l.
|
||||
By default, TARGET_ARCH is set to HOST_ARCH. On a x86_64 machine, not setting TARGET_ARCH is the equivalent of setting TARGET_ARCH=x86_64.<br/>
|
||||
`$ make TARGET_ARCH=x86_64` <br/> `$ make TARGET_ARCH=ppc64le` <br/> `$ make TARGET_ARCH=armv7l` <br/>
|
||||
See [here](http://docs.nvidia.com/cuda/cuda-samples/index.html#cross-samples) for more details.
|
||||
* **dbg=1** - build with debug symbols
|
||||
```
|
||||
$ make dbg=1
|
||||
```
|
||||
* **SMS="A B ..."** - override the SM architectures for which the sample will be built, where `"A B ..."` is a space-delimited list of SM architectures. For example, to generate SASS for SM 50 and SM 60, use `SMS="50 60"`.
|
||||
```
|
||||
$ make SMS="50 60"
|
||||
```
|
||||
|
||||
* **HOST_COMPILER=<host_compiler>** - override the default g++ host compiler. See the [Linux Installation Guide](http://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#system-requirements) for a list of supported host compilers.
|
||||
```
|
||||
$ make HOST_COMPILER=g++
|
||||
```
|
||||
|
||||
## References (for more details)
|
||||
|
||||
400
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/simpleCUFFT_MGPU.cu
Normal file
400
Samples/4_CUDA_Libraries/simpleCUFFT_MGPU/simpleCUFFT_MGPU.cu
Normal file
@@ -0,0 +1,400 @@
|
||||
/* Copyright (c) 2022, 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.
|
||||
*/
|
||||
|
||||
/* Example showing the use of CUFFT for fast 1D-convolution using FFT. */
|
||||
|
||||
// System includes
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
|
||||
#include <string.h>
|
||||
#include <math.h>
|
||||
|
||||
// CUDA runtime
|
||||
#include <cuda_runtime.h>
|
||||
|
||||
//CUFFT Header file
|
||||
#include <cufftXt.h>
|
||||
|
||||
// helper functions and utilities to work with CUDA
|
||||
#include <helper_functions.h>
|
||||
#include <helper_cuda.h>
|
||||
|
||||
// Complex data type
|
||||
typedef float2 Complex;
|
||||
|
||||
static __device__ __host__ inline Complex ComplexAdd(Complex, Complex);
|
||||
static __device__ __host__ inline Complex ComplexScale(Complex, float);
|
||||
static __device__ __host__ inline Complex ComplexMul(Complex, Complex);
|
||||
static __global__ void ComplexPointwiseMulAndScale(cufftComplex *,
|
||||
cufftComplex *, int, float);
|
||||
|
||||
// Kernel for GPU
|
||||
void multiplyCoefficient(cudaLibXtDesc *, cudaLibXtDesc *, int, float, int);
|
||||
|
||||
// Filtering functions
|
||||
void Convolve(const Complex *, int, const Complex *, int, Complex *);
|
||||
|
||||
// Padding functions
|
||||
int PadData(const Complex *, Complex **, int, const Complex *, Complex **, int);
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Data configuration
|
||||
// The filter size is assumed to be a number smaller than the signal size
|
||||
///////////////////////////////////////////////////////////////////////////////
|
||||
const int SIGNAL_SIZE = 1018;
|
||||
const int FILTER_KERNEL_SIZE = 11;
|
||||
const int GPU_COUNT = 2;
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Program main
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
int main(int argc, char **argv) {
|
||||
printf("\n[simpleCUFFT_MGPU] is starting...\n\n");
|
||||
|
||||
int GPU_N;
|
||||
checkCudaErrors(cudaGetDeviceCount(&GPU_N));
|
||||
|
||||
if (GPU_N < GPU_COUNT) {
|
||||
printf("No. of GPU on node %d\n", GPU_N);
|
||||
printf("Two GPUs are required to run simpleCUFFT_MGPU sample code\n");
|
||||
exit(EXIT_WAIVED);
|
||||
}
|
||||
|
||||
int *major_minor = (int *)malloc(sizeof(int) * GPU_N * 2);
|
||||
int found2IdenticalGPUs = 0;
|
||||
int nGPUs = 2;
|
||||
int *whichGPUs;
|
||||
whichGPUs = (int *)malloc(sizeof(int) * nGPUs);
|
||||
|
||||
for (int i = 0; i < GPU_N; i++) {
|
||||
cudaDeviceProp deviceProp;
|
||||
checkCudaErrors(cudaGetDeviceProperties(&deviceProp, i));
|
||||
major_minor[i * 2] = deviceProp.major;
|
||||
major_minor[i * 2 + 1] = deviceProp.minor;
|
||||
printf("GPU Device %d: \"%s\" with compute capability %d.%d\n", i,
|
||||
deviceProp.name, deviceProp.major, deviceProp.minor);
|
||||
}
|
||||
|
||||
for (int i = 0; i < GPU_N; i++) {
|
||||
for (int j = i + 1; j < GPU_N; j++) {
|
||||
if ((major_minor[i * 2] == major_minor[j * 2]) &&
|
||||
(major_minor[i * 2 + 1] == major_minor[j * 2 + 1])) {
|
||||
whichGPUs[0] = i;
|
||||
whichGPUs[1] = j;
|
||||
found2IdenticalGPUs = 1;
|
||||
break;
|
||||
}
|
||||
}
|
||||
if (found2IdenticalGPUs) {
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
free(major_minor);
|
||||
if (!found2IdenticalGPUs) {
|
||||
printf(
|
||||
"No Two GPUs with same architecture found\nWaiving simpleCUFFT_2d_MGPU "
|
||||
"sample\n");
|
||||
exit(EXIT_WAIVED);
|
||||
}
|
||||
|
||||
// Allocate host memory for the signal
|
||||
Complex *h_signal = (Complex *)malloc(sizeof(Complex) * SIGNAL_SIZE);
|
||||
|
||||
// Initialize the memory for the signal
|
||||
for (int i = 0; i < SIGNAL_SIZE; ++i) {
|
||||
h_signal[i].x = rand() / (float)RAND_MAX;
|
||||
h_signal[i].y = 0;
|
||||
}
|
||||
|
||||
// Allocate host memory for the filter
|
||||
Complex *h_filter_kernel =
|
||||
(Complex *)malloc(sizeof(Complex) * FILTER_KERNEL_SIZE);
|
||||
|
||||
// Initialize the memory for the filter
|
||||
for (int i = 0; i < FILTER_KERNEL_SIZE; ++i) {
|
||||
h_filter_kernel[i].x = rand() / (float)RAND_MAX;
|
||||
h_filter_kernel[i].y = 0;
|
||||
}
|
||||
|
||||
// Pad signal and filter kernel
|
||||
Complex *h_padded_signal;
|
||||
Complex *h_padded_filter_kernel;
|
||||
int new_size =
|
||||
PadData(h_signal, &h_padded_signal, SIGNAL_SIZE, h_filter_kernel,
|
||||
&h_padded_filter_kernel, FILTER_KERNEL_SIZE);
|
||||
|
||||
// cufftCreate() - Create an empty plan
|
||||
cufftResult result;
|
||||
cufftHandle plan_input;
|
||||
checkCudaErrors(cufftCreate(&plan_input));
|
||||
|
||||
// cufftXtSetGPUs() - Define which GPUs to use
|
||||
result = cufftXtSetGPUs(plan_input, nGPUs, whichGPUs);
|
||||
|
||||
if (result == CUFFT_INVALID_DEVICE) {
|
||||
printf("This sample requires two GPUs on the same board.\n");
|
||||
printf("No such board was found. Waiving sample.\n");
|
||||
exit(EXIT_WAIVED);
|
||||
} else if (result != CUFFT_SUCCESS) {
|
||||
printf("cufftXtSetGPUs failed\n");
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
|
||||
// Print the device information to run the code
|
||||
printf("\nRunning on GPUs\n");
|
||||
for (int i = 0; i < nGPUs; i++) {
|
||||
cudaDeviceProp deviceProp;
|
||||
checkCudaErrors(cudaGetDeviceProperties(&deviceProp, whichGPUs[i]));
|
||||
printf("GPU Device %d: \"%s\" with compute capability %d.%d\n",
|
||||
whichGPUs[i], deviceProp.name, deviceProp.major, deviceProp.minor);
|
||||
}
|
||||
|
||||
size_t *worksize;
|
||||
worksize = (size_t *)malloc(sizeof(size_t) * nGPUs);
|
||||
|
||||
// cufftMakePlan1d() - Create the plan
|
||||
checkCudaErrors(
|
||||
cufftMakePlan1d(plan_input, new_size, CUFFT_C2C, 1, worksize));
|
||||
|
||||
// cufftXtMalloc() - Malloc data on multiple GPUs
|
||||
cudaLibXtDesc *d_signal;
|
||||
checkCudaErrors(cufftXtMalloc(plan_input, (cudaLibXtDesc **)&d_signal,
|
||||
CUFFT_XT_FORMAT_INPLACE));
|
||||
cudaLibXtDesc *d_out_signal;
|
||||
checkCudaErrors(cufftXtMalloc(plan_input, (cudaLibXtDesc **)&d_out_signal,
|
||||
CUFFT_XT_FORMAT_INPLACE));
|
||||
cudaLibXtDesc *d_filter_kernel;
|
||||
checkCudaErrors(cufftXtMalloc(plan_input, (cudaLibXtDesc **)&d_filter_kernel,
|
||||
CUFFT_XT_FORMAT_INPLACE));
|
||||
cudaLibXtDesc *d_out_filter_kernel;
|
||||
checkCudaErrors(cufftXtMalloc(plan_input,
|
||||
(cudaLibXtDesc **)&d_out_filter_kernel,
|
||||
CUFFT_XT_FORMAT_INPLACE));
|
||||
|
||||
// cufftXtMemcpy() - Copy data from host to multiple GPUs
|
||||
checkCudaErrors(cufftXtMemcpy(plan_input, d_signal, h_padded_signal,
|
||||
CUFFT_COPY_HOST_TO_DEVICE));
|
||||
checkCudaErrors(cufftXtMemcpy(plan_input, d_filter_kernel,
|
||||
h_padded_filter_kernel,
|
||||
CUFFT_COPY_HOST_TO_DEVICE));
|
||||
|
||||
// cufftXtExecDescriptorC2C() - Execute FFT on data on multiple GPUs
|
||||
checkCudaErrors(
|
||||
cufftXtExecDescriptorC2C(plan_input, d_signal, d_signal, CUFFT_FORWARD));
|
||||
checkCudaErrors(cufftXtExecDescriptorC2C(plan_input, d_filter_kernel,
|
||||
d_filter_kernel, CUFFT_FORWARD));
|
||||
|
||||
// cufftXtMemcpy() - Copy the data to natural order on GPUs
|
||||
checkCudaErrors(cufftXtMemcpy(plan_input, d_out_signal, d_signal,
|
||||
CUFFT_COPY_DEVICE_TO_DEVICE));
|
||||
checkCudaErrors(cufftXtMemcpy(plan_input, d_out_filter_kernel,
|
||||
d_filter_kernel, CUFFT_COPY_DEVICE_TO_DEVICE));
|
||||
|
||||
printf("\n\nValue of Library Descriptor\n");
|
||||
printf("Number of GPUs %d\n", d_out_signal->descriptor->nGPUs);
|
||||
printf("Device id %d %d\n", d_out_signal->descriptor->GPUs[0],
|
||||
d_out_signal->descriptor->GPUs[1]);
|
||||
printf("Data size on GPU %ld %ld\n",
|
||||
(long)(d_out_signal->descriptor->size[0] / sizeof(cufftComplex)),
|
||||
(long)(d_out_signal->descriptor->size[1] / sizeof(cufftComplex)));
|
||||
|
||||
// Multiply the coefficients together and normalize the result
|
||||
printf("Launching ComplexPointwiseMulAndScale<<< >>>\n");
|
||||
multiplyCoefficient(d_out_signal, d_out_filter_kernel, new_size,
|
||||
1.0f / new_size, nGPUs);
|
||||
|
||||
// cufftXtExecDescriptorC2C() - Execute inverse FFT on data on multiple GPUs
|
||||
printf("Transforming signal back cufftExecC2C\n");
|
||||
checkCudaErrors(cufftXtExecDescriptorC2C(plan_input, d_out_signal,
|
||||
d_out_signal, CUFFT_INVERSE));
|
||||
|
||||
// Create host pointer pointing to padded signal
|
||||
Complex *h_convolved_signal = h_padded_signal;
|
||||
|
||||
// Allocate host memory for the convolution result
|
||||
Complex *h_convolved_signal_ref =
|
||||
(Complex *)malloc(sizeof(Complex) * SIGNAL_SIZE);
|
||||
|
||||
// cufftXtMemcpy() - Copy data from multiple GPUs to host
|
||||
checkCudaErrors(cufftXtMemcpy(plan_input, h_convolved_signal, d_out_signal,
|
||||
CUFFT_COPY_DEVICE_TO_HOST));
|
||||
|
||||
// Convolve on the host
|
||||
Convolve(h_signal, SIGNAL_SIZE, h_filter_kernel, FILTER_KERNEL_SIZE,
|
||||
h_convolved_signal_ref);
|
||||
|
||||
// Compare CPU and GPU result
|
||||
bool bTestResult =
|
||||
sdkCompareL2fe((float *)h_convolved_signal_ref,
|
||||
(float *)h_convolved_signal, 2 * SIGNAL_SIZE, 1e-5f);
|
||||
printf("\nvalue of TestResult %d\n", bTestResult);
|
||||
|
||||
// Cleanup memory
|
||||
free(whichGPUs);
|
||||
free(worksize);
|
||||
free(h_signal);
|
||||
free(h_filter_kernel);
|
||||
free(h_padded_signal);
|
||||
free(h_padded_filter_kernel);
|
||||
free(h_convolved_signal_ref);
|
||||
|
||||
// cudaXtFree() - Free GPU memory
|
||||
checkCudaErrors(cufftXtFree(d_signal));
|
||||
checkCudaErrors(cufftXtFree(d_filter_kernel));
|
||||
checkCudaErrors(cufftXtFree(d_out_signal));
|
||||
checkCudaErrors(cufftXtFree(d_out_filter_kernel));
|
||||
|
||||
// cufftDestroy() - Destroy FFT plan
|
||||
checkCudaErrors(cufftDestroy(plan_input));
|
||||
|
||||
exit(bTestResult ? EXIT_SUCCESS : EXIT_FAILURE);
|
||||
}
|
||||
|
||||
///////////////////////////////////////////////////////////////////////////////////
|
||||
// Function for padding original data
|
||||
//////////////////////////////////////////////////////////////////////////////////
|
||||
int PadData(const Complex *signal, Complex **padded_signal, int signal_size,
|
||||
const Complex *filter_kernel, Complex **padded_filter_kernel,
|
||||
int filter_kernel_size) {
|
||||
int minRadius = filter_kernel_size / 2;
|
||||
int maxRadius = filter_kernel_size - minRadius;
|
||||
int new_size = signal_size + maxRadius;
|
||||
|
||||
// Pad signal
|
||||
Complex *new_data = (Complex *)malloc(sizeof(Complex) * new_size);
|
||||
memcpy(new_data + 0, signal, signal_size * sizeof(Complex));
|
||||
memset(new_data + signal_size, 0, (new_size - signal_size) * sizeof(Complex));
|
||||
*padded_signal = new_data;
|
||||
|
||||
// Pad filter
|
||||
new_data = (Complex *)malloc(sizeof(Complex) * new_size);
|
||||
memcpy(new_data + 0, filter_kernel + minRadius, maxRadius * sizeof(Complex));
|
||||
memset(new_data + maxRadius, 0,
|
||||
(new_size - filter_kernel_size) * sizeof(Complex));
|
||||
memcpy(new_data + new_size - minRadius, filter_kernel,
|
||||
minRadius * sizeof(Complex));
|
||||
*padded_filter_kernel = new_data;
|
||||
|
||||
return new_size;
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Filtering operations - Computing Convolution on the host
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
void Convolve(const Complex *signal, int signal_size,
|
||||
const Complex *filter_kernel, int filter_kernel_size,
|
||||
Complex *filtered_signal) {
|
||||
int minRadius = filter_kernel_size / 2;
|
||||
int maxRadius = filter_kernel_size - minRadius;
|
||||
|
||||
// Loop over output element indices
|
||||
for (int i = 0; i < signal_size; ++i) {
|
||||
filtered_signal[i].x = filtered_signal[i].y = 0;
|
||||
|
||||
// Loop over convolution indices
|
||||
for (int j = -maxRadius + 1; j <= minRadius; ++j) {
|
||||
int k = i + j;
|
||||
|
||||
if (k >= 0 && k < signal_size) {
|
||||
filtered_signal[i] =
|
||||
ComplexAdd(filtered_signal[i],
|
||||
ComplexMul(signal[k], filter_kernel[minRadius - j]));
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Launch Kernel on multiple GPU
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
void multiplyCoefficient(cudaLibXtDesc *d_signal,
|
||||
cudaLibXtDesc *d_filter_kernel, int new_size,
|
||||
float val, int nGPUs) {
|
||||
int device;
|
||||
// Launch the ComplexPointwiseMulAndScale<<< >>> kernel on multiple GPU
|
||||
for (int i = 0; i < nGPUs; i++) {
|
||||
device = d_signal->descriptor->GPUs[i];
|
||||
|
||||
// Set device
|
||||
checkCudaErrors(cudaSetDevice(device));
|
||||
|
||||
// Perform GPU computations
|
||||
ComplexPointwiseMulAndScale<<<32, 256>>>(
|
||||
(cufftComplex *)d_signal->descriptor->data[i],
|
||||
(cufftComplex *)d_filter_kernel->descriptor->data[i],
|
||||
int(d_signal->descriptor->size[i] / sizeof(cufftComplex)), val);
|
||||
}
|
||||
|
||||
// Wait for device to finish all operation
|
||||
for (int i = 0; i < nGPUs; i++) {
|
||||
device = d_signal->descriptor->GPUs[i];
|
||||
checkCudaErrors(cudaSetDevice(device));
|
||||
cudaDeviceSynchronize();
|
||||
// Check if kernel execution generated and error
|
||||
getLastCudaError("Kernel execution failed [ ComplexPointwiseMulAndScale ]");
|
||||
}
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Complex operations
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
|
||||
// Complex addition
|
||||
static __device__ __host__ inline Complex ComplexAdd(Complex a, Complex b) {
|
||||
Complex c;
|
||||
c.x = a.x + b.x;
|
||||
c.y = a.y + b.y;
|
||||
return c;
|
||||
}
|
||||
|
||||
// Complex scale
|
||||
static __device__ __host__ inline Complex ComplexScale(Complex a, float s) {
|
||||
Complex c;
|
||||
c.x = s * a.x;
|
||||
c.y = s * a.y;
|
||||
return c;
|
||||
}
|
||||
|
||||
// Complex multiplication
|
||||
static __device__ __host__ inline Complex ComplexMul(Complex a, Complex b) {
|
||||
Complex c;
|
||||
c.x = a.x * b.x - a.y * b.y;
|
||||
c.y = a.x * b.y + a.y * b.x;
|
||||
return c;
|
||||
}
|
||||
// Complex pointwise multiplication
|
||||
static __global__ void ComplexPointwiseMulAndScale(cufftComplex *a,
|
||||
cufftComplex *b, int size,
|
||||
float scale) {
|
||||
const int numThreads = blockDim.x * gridDim.x;
|
||||
const int threadID = blockIdx.x * blockDim.x + threadIdx.x;
|
||||
for (int i = threadID; i < size; i += numThreads) {
|
||||
a[i] = ComplexScale(ComplexMul(a[i], b[i]), scale);
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,20 @@
|
||||
|
||||
Microsoft Visual Studio Solution File, Format Version 12.00
|
||||
# Visual Studio 2017
|
||||
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "simpleCUFFT_MGPU", "simpleCUFFT_MGPU_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
|
||||
@@ -0,0 +1,112 @@
|
||||
<?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>simpleCUFFT_MGPU_vs2017</RootNamespace>
|
||||
<ProjectName>simpleCUFFT_MGPU</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 11.6.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;$(CudaToolkitIncludeDir);</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cufft.lib;cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/simpleCUFFT_MGPU.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
|
||||
<Include>./;../../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(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="'$(Configuration)'=='Release'">
|
||||
<ClCompile>
|
||||
<Optimization>MaxSpeed</Optimization>
|
||||
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MT</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<CudaCompile Include="simpleCUFFT_MGPU.cu" />
|
||||
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
@@ -0,0 +1,20 @@
|
||||
|
||||
Microsoft Visual Studio Solution File, Format Version 12.00
|
||||
# Visual Studio 2019
|
||||
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "simpleCUFFT_MGPU", "simpleCUFFT_MGPU_vs2019.vcxproj", "{997E0757-EA74-4A4E-A0FC-47D8C8831A15}"
|
||||
EndProject
|
||||
Global
|
||||
GlobalSection(SolutionConfigurationPlatforms) = preSolution
|
||||
Debug|x64 = Debug|x64
|
||||
Release|x64 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(ProjectConfigurationPlatforms) = postSolution
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.ActiveCfg = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.Build.0 = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.ActiveCfg = Release|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.Build.0 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(SolutionProperties) = preSolution
|
||||
HideSolutionNode = FALSE
|
||||
EndGlobalSection
|
||||
EndGlobal
|
||||
@@ -0,0 +1,108 @@
|
||||
<?xml version="1.0" encoding="utf-8"?>
|
||||
<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
|
||||
<PropertyGroup>
|
||||
<CUDAPropsPath Condition="'$(CUDAPropsPath)'==''">$(VCTargetsPath)\BuildCustomizations</CUDAPropsPath>
|
||||
</PropertyGroup>
|
||||
<ItemGroup Label="ProjectConfigurations">
|
||||
<ProjectConfiguration Include="Debug|x64">
|
||||
<Configuration>Debug</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
<ProjectConfiguration Include="Release|x64">
|
||||
<Configuration>Release</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
</ItemGroup>
|
||||
<PropertyGroup Label="Globals">
|
||||
<ProjectGuid>{997E0757-EA74-4A4E-A0FC-47D8C8831A15}</ProjectGuid>
|
||||
<RootNamespace>simpleCUFFT_MGPU_vs2019</RootNamespace>
|
||||
<ProjectName>simpleCUFFT_MGPU</ProjectName>
|
||||
<CudaToolkitCustomDir />
|
||||
</PropertyGroup>
|
||||
<Import Project="$(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 11.6.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;$(CudaToolkitIncludeDir);</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cufft.lib;cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/simpleCUFFT_MGPU.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
|
||||
<Include>./;../../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(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="'$(Configuration)'=='Release'">
|
||||
<ClCompile>
|
||||
<Optimization>MaxSpeed</Optimization>
|
||||
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MT</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<CudaCompile Include="simpleCUFFT_MGPU.cu" />
|
||||
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
@@ -0,0 +1,20 @@
|
||||
|
||||
Microsoft Visual Studio Solution File, Format Version 12.00
|
||||
# Visual Studio 2022
|
||||
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "simpleCUFFT_MGPU", "simpleCUFFT_MGPU_vs2022.vcxproj", "{997E0757-EA74-4A4E-A0FC-47D8C8831A15}"
|
||||
EndProject
|
||||
Global
|
||||
GlobalSection(SolutionConfigurationPlatforms) = preSolution
|
||||
Debug|x64 = Debug|x64
|
||||
Release|x64 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(ProjectConfigurationPlatforms) = postSolution
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.ActiveCfg = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.Build.0 = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.ActiveCfg = Release|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.Build.0 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(SolutionProperties) = preSolution
|
||||
HideSolutionNode = FALSE
|
||||
EndGlobalSection
|
||||
EndGlobal
|
||||
@@ -0,0 +1,108 @@
|
||||
<?xml version="1.0" encoding="utf-8"?>
|
||||
<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
|
||||
<PropertyGroup>
|
||||
<CUDAPropsPath Condition="'$(CUDAPropsPath)'==''">$(VCTargetsPath)\BuildCustomizations</CUDAPropsPath>
|
||||
</PropertyGroup>
|
||||
<ItemGroup Label="ProjectConfigurations">
|
||||
<ProjectConfiguration Include="Debug|x64">
|
||||
<Configuration>Debug</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
<ProjectConfiguration Include="Release|x64">
|
||||
<Configuration>Release</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
</ItemGroup>
|
||||
<PropertyGroup Label="Globals">
|
||||
<ProjectGuid>{997E0757-EA74-4A4E-A0FC-47D8C8831A15}</ProjectGuid>
|
||||
<RootNamespace>simpleCUFFT_MGPU_vs2022</RootNamespace>
|
||||
<ProjectName>simpleCUFFT_MGPU</ProjectName>
|
||||
<CudaToolkitCustomDir />
|
||||
</PropertyGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
|
||||
<PropertyGroup>
|
||||
<ConfigurationType>Application</ConfigurationType>
|
||||
<CharacterSet>MultiByte</CharacterSet>
|
||||
<PlatformToolset>v143</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 11.6.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;$(CudaToolkitIncludeDir);</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cufft.lib;cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/simpleCUFFT_MGPU.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
|
||||
<Include>./;../../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(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="'$(Configuration)'=='Release'">
|
||||
<ClCompile>
|
||||
<Optimization>MaxSpeed</Optimization>
|
||||
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MT</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<CudaCompile Include="simpleCUFFT_MGPU.cu" />
|
||||
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
Reference in New Issue
Block a user