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/2_Concepts_and_Techniques/scan/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/2_Concepts_and_Techniques/scan/.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/2_Concepts_and_Techniques/scan/.vscode/extensions.json
vendored
Normal file
7
Samples/2_Concepts_and_Techniques/scan/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
||||
{
|
||||
"recommendations": [
|
||||
"nvidia.nsight-vscode-edition",
|
||||
"ms-vscode.cpptools",
|
||||
"ms-vscode.makefile-tools"
|
||||
]
|
||||
}
|
||||
10
Samples/2_Concepts_and_Techniques/scan/.vscode/launch.json
vendored
Normal file
10
Samples/2_Concepts_and_Techniques/scan/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
||||
{
|
||||
"configurations": [
|
||||
{
|
||||
"name": "CUDA C++: Launch",
|
||||
"type": "cuda-gdb",
|
||||
"request": "launch",
|
||||
"program": "${workspaceFolder}/scan"
|
||||
}
|
||||
]
|
||||
}
|
||||
15
Samples/2_Concepts_and_Techniques/scan/.vscode/tasks.json
vendored
Normal file
15
Samples/2_Concepts_and_Techniques/scan/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
||||
{
|
||||
"version": "2.0.0",
|
||||
"tasks": [
|
||||
{
|
||||
"label": "sample",
|
||||
"type": "shell",
|
||||
"command": "make dbg=1",
|
||||
"problemMatcher": ["$nvcc"],
|
||||
"group": {
|
||||
"kind": "build",
|
||||
"isDefault": true
|
||||
}
|
||||
}
|
||||
]
|
||||
}
|
||||
346
Samples/2_Concepts_and_Techniques/scan/Makefile
Normal file
346
Samples/2_Concepts_and_Techniques/scan/Makefile
Normal file
@@ -0,0 +1,346 @@
|
||||
################################################################################
|
||||
# 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
|
||||
|
||||
ifeq ($(SAMPLE_ENABLED),0)
|
||||
EXEC ?= @echo "[@]"
|
||||
endif
|
||||
|
||||
################################################################################
|
||||
|
||||
# Target rules
|
||||
all: build
|
||||
|
||||
build: scan
|
||||
|
||||
check.deps:
|
||||
ifeq ($(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
|
||||
|
||||
main.o:main.cpp
|
||||
$(EXEC) $(NVCC) $(INCLUDES) $(ALL_CCFLAGS) $(GENCODE_FLAGS) -o $@ -c $<
|
||||
|
||||
scan.o:scan.cu
|
||||
$(EXEC) $(NVCC) $(INCLUDES) $(ALL_CCFLAGS) $(GENCODE_FLAGS) -o $@ -c $<
|
||||
|
||||
scan_gold.o:scan_gold.cpp
|
||||
$(EXEC) $(NVCC) $(INCLUDES) $(ALL_CCFLAGS) $(GENCODE_FLAGS) -o $@ -c $<
|
||||
|
||||
scan: main.o scan.o scan_gold.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
|
||||
$(EXEC) ./scan
|
||||
|
||||
testrun: build
|
||||
|
||||
clean:
|
||||
rm -f scan main.o scan.o scan_gold.o
|
||||
rm -rf ../../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)/scan
|
||||
|
||||
clobber: clean
|
||||
81
Samples/2_Concepts_and_Techniques/scan/NsightEclipse.xml
Normal file
81
Samples/2_Concepts_and_Techniques/scan/NsightEclipse.xml
Normal file
@@ -0,0 +1,81 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<!DOCTYPE entry SYSTEM "SamplesInfo.dtd">
|
||||
<entry>
|
||||
<name>scan</name>
|
||||
<cuda_api_list>
|
||||
<toolkit>cudaMalloc</toolkit>
|
||||
<toolkit>cudaFree</toolkit>
|
||||
<toolkit>cudaDeviceSynchronize</toolkit>
|
||||
<toolkit>cudaMemcpy</toolkit>
|
||||
</cuda_api_list>
|
||||
<description><![CDATA[This example demonstrates an efficient CUDA implementation of parallel prefix sum, also known as "scan". Given an array of numbers, scan computes a new array in which each element is the sum of all the elements before it in the input array.]]></description>
|
||||
<devicecompilation>whole</devicecompilation>
|
||||
<includepaths>
|
||||
<path>./</path>
|
||||
<path>../</path>
|
||||
<path>../../../Common</path>
|
||||
</includepaths>
|
||||
<keyconcepts>
|
||||
<concept level="advanced">Data-Parallel Algorithms</concept>
|
||||
<concept level="advanced">Performance Strategies</concept>
|
||||
</keyconcepts>
|
||||
<keywords>
|
||||
<keyword>GPGPU</keyword>
|
||||
<keyword>CUDA</keyword>
|
||||
<keyword>scan</keyword>
|
||||
<keyword>parallel prefix sum</keyword>
|
||||
<keyword>Data-Parallel Algorithms</keyword>
|
||||
</keywords>
|
||||
<libraries>
|
||||
</libraries>
|
||||
<librarypaths>
|
||||
</librarypaths>
|
||||
<nsight_eclipse>true</nsight_eclipse>
|
||||
<primary_file>main.cpp</primary_file>
|
||||
<scopes>
|
||||
<scope>1:CUDA Advanced Topics</scope>
|
||||
<scope>1:Data-Parallel Algorithms</scope>
|
||||
<scope>1:Performance Strategies</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>CUDA Parallel Prefix Sum (Scan)</title>
|
||||
<type>exe</type>
|
||||
</entry>
|
||||
70
Samples/2_Concepts_and_Techniques/scan/README.md
Normal file
70
Samples/2_Concepts_and_Techniques/scan/README.md
Normal file
@@ -0,0 +1,70 @@
|
||||
# scan - CUDA Parallel Prefix Sum (Scan)
|
||||
|
||||
## Description
|
||||
|
||||
This example demonstrates an efficient CUDA implementation of parallel prefix sum, also known as "scan". Given an array of numbers, scan computes a new array in which each element is the sum of all the elements before it in the input array.
|
||||
|
||||
## Key Concepts
|
||||
|
||||
Data-Parallel Algorithms, Performance Strategies
|
||||
|
||||
## 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)
|
||||
cudaMalloc, cudaFree, cudaDeviceSynchronize, cudaMemcpy
|
||||
|
||||
## Prerequisites
|
||||
|
||||
Download and install the [CUDA Toolkit 11.6](https://developer.nvidia.com/cuda-downloads) for your corresponding platform.
|
||||
|
||||
## 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)
|
||||
|
||||
192
Samples/2_Concepts_and_Techniques/scan/main.cpp
Normal file
192
Samples/2_Concepts_and_Techniques/scan/main.cpp
Normal file
@@ -0,0 +1,192 @@
|
||||
/* 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.
|
||||
*/
|
||||
|
||||
#include <cuda_runtime.h>
|
||||
#include <helper_cuda.h>
|
||||
#include <helper_functions.h>
|
||||
|
||||
#include "scan_common.h"
|
||||
|
||||
int main(int argc, char **argv) {
|
||||
printf("%s Starting...\n\n", argv[0]);
|
||||
|
||||
// Use command-line specified CUDA device, otherwise use device with highest
|
||||
// Gflops/s
|
||||
findCudaDevice(argc, (const char **)argv);
|
||||
|
||||
uint *d_Input, *d_Output;
|
||||
uint *h_Input, *h_OutputCPU, *h_OutputGPU;
|
||||
StopWatchInterface *hTimer = NULL;
|
||||
const uint N = 13 * 1048576 / 2;
|
||||
|
||||
printf("Allocating and initializing host arrays...\n");
|
||||
sdkCreateTimer(&hTimer);
|
||||
h_Input = (uint *)malloc(N * sizeof(uint));
|
||||
h_OutputCPU = (uint *)malloc(N * sizeof(uint));
|
||||
h_OutputGPU = (uint *)malloc(N * sizeof(uint));
|
||||
srand(2009);
|
||||
|
||||
for (uint i = 0; i < N; i++) {
|
||||
h_Input[i] = rand();
|
||||
}
|
||||
|
||||
printf("Allocating and initializing CUDA arrays...\n");
|
||||
checkCudaErrors(cudaMalloc((void **)&d_Input, N * sizeof(uint)));
|
||||
checkCudaErrors(cudaMalloc((void **)&d_Output, N * sizeof(uint)));
|
||||
checkCudaErrors(
|
||||
cudaMemcpy(d_Input, h_Input, N * sizeof(uint), cudaMemcpyHostToDevice));
|
||||
|
||||
printf("Initializing CUDA-C scan...\n\n");
|
||||
initScan();
|
||||
|
||||
int globalFlag = 1;
|
||||
size_t szWorkgroup;
|
||||
const int iCycles = 100;
|
||||
printf(
|
||||
"*** Running GPU scan for short arrays (%d identical iterations)...\n\n",
|
||||
iCycles);
|
||||
|
||||
for (uint arrayLength = MIN_SHORT_ARRAY_SIZE;
|
||||
arrayLength <= MAX_SHORT_ARRAY_SIZE; arrayLength <<= 1) {
|
||||
printf("Running scan for %u elements (%u arrays)...\n", arrayLength,
|
||||
N / arrayLength);
|
||||
checkCudaErrors(cudaDeviceSynchronize());
|
||||
sdkResetTimer(&hTimer);
|
||||
sdkStartTimer(&hTimer);
|
||||
|
||||
for (int i = 0; i < iCycles; i++) {
|
||||
szWorkgroup =
|
||||
scanExclusiveShort(d_Output, d_Input, N / arrayLength, arrayLength);
|
||||
}
|
||||
|
||||
checkCudaErrors(cudaDeviceSynchronize());
|
||||
sdkStopTimer(&hTimer);
|
||||
double timerValue = 1.0e-3 * sdkGetTimerValue(&hTimer) / iCycles;
|
||||
|
||||
printf("Validating the results...\n");
|
||||
printf("...reading back GPU results\n");
|
||||
checkCudaErrors(cudaMemcpy(h_OutputGPU, d_Output, N * sizeof(uint),
|
||||
cudaMemcpyDeviceToHost));
|
||||
|
||||
printf(" ...scanExclusiveHost()\n");
|
||||
scanExclusiveHost(h_OutputCPU, h_Input, N / arrayLength, arrayLength);
|
||||
|
||||
// Compare GPU results with CPU results and accumulate error for this test
|
||||
printf(" ...comparing the results\n");
|
||||
int localFlag = 1;
|
||||
|
||||
for (uint i = 0; i < N; i++) {
|
||||
if (h_OutputCPU[i] != h_OutputGPU[i]) {
|
||||
localFlag = 0;
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
// Log message on individual test result, then accumulate to global flag
|
||||
printf(" ...Results %s\n\n",
|
||||
(localFlag == 1) ? "Match" : "DON'T Match !!!");
|
||||
globalFlag = globalFlag && localFlag;
|
||||
|
||||
// Data log
|
||||
if (arrayLength == MAX_SHORT_ARRAY_SIZE) {
|
||||
printf("\n");
|
||||
printf(
|
||||
"scan, Throughput = %.4f MElements/s, Time = %.5f s, Size = %u "
|
||||
"Elements, NumDevsUsed = %u, Workgroup = %u\n",
|
||||
(1.0e-6 * (double)arrayLength / timerValue), timerValue,
|
||||
(unsigned int)arrayLength, 1, (unsigned int)szWorkgroup);
|
||||
printf("\n");
|
||||
}
|
||||
}
|
||||
|
||||
printf(
|
||||
"***Running GPU scan for large arrays (%u identical iterations)...\n\n",
|
||||
iCycles);
|
||||
|
||||
for (uint arrayLength = MIN_LARGE_ARRAY_SIZE;
|
||||
arrayLength <= MAX_LARGE_ARRAY_SIZE; arrayLength <<= 1) {
|
||||
printf("Running scan for %u elements (%u arrays)...\n", arrayLength,
|
||||
N / arrayLength);
|
||||
checkCudaErrors(cudaDeviceSynchronize());
|
||||
sdkResetTimer(&hTimer);
|
||||
sdkStartTimer(&hTimer);
|
||||
|
||||
for (int i = 0; i < iCycles; i++) {
|
||||
szWorkgroup =
|
||||
scanExclusiveLarge(d_Output, d_Input, N / arrayLength, arrayLength);
|
||||
}
|
||||
|
||||
checkCudaErrors(cudaDeviceSynchronize());
|
||||
sdkStopTimer(&hTimer);
|
||||
double timerValue = 1.0e-3 * sdkGetTimerValue(&hTimer) / iCycles;
|
||||
|
||||
printf("Validating the results...\n");
|
||||
printf("...reading back GPU results\n");
|
||||
checkCudaErrors(cudaMemcpy(h_OutputGPU, d_Output, N * sizeof(uint),
|
||||
cudaMemcpyDeviceToHost));
|
||||
|
||||
printf("...scanExclusiveHost()\n");
|
||||
scanExclusiveHost(h_OutputCPU, h_Input, N / arrayLength, arrayLength);
|
||||
|
||||
// Compare GPU results with CPU results and accumulate error for this test
|
||||
printf(" ...comparing the results\n");
|
||||
int localFlag = 1;
|
||||
|
||||
for (uint i = 0; i < N; i++) {
|
||||
if (h_OutputCPU[i] != h_OutputGPU[i]) {
|
||||
localFlag = 0;
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
// Log message on individual test result, then accumulate to global flag
|
||||
printf(" ...Results %s\n\n",
|
||||
(localFlag == 1) ? "Match" : "DON'T Match !!!");
|
||||
globalFlag = globalFlag && localFlag;
|
||||
|
||||
// Data log
|
||||
if (arrayLength == MAX_LARGE_ARRAY_SIZE) {
|
||||
printf("\n");
|
||||
printf(
|
||||
"scan, Throughput = %.4f MElements/s, Time = %.5f s, Size = %u "
|
||||
"Elements, NumDevsUsed = %u, Workgroup = %u\n",
|
||||
(1.0e-6 * (double)arrayLength / timerValue), timerValue,
|
||||
(unsigned int)arrayLength, 1, (unsigned int)szWorkgroup);
|
||||
printf("\n");
|
||||
}
|
||||
}
|
||||
|
||||
printf("Shutting down...\n");
|
||||
closeScan();
|
||||
checkCudaErrors(cudaFree(d_Output));
|
||||
checkCudaErrors(cudaFree(d_Input));
|
||||
|
||||
sdkDeleteTimer(&hTimer);
|
||||
|
||||
// pass or fail (cumulative... all tests in the loop)
|
||||
exit(globalFlag ? EXIT_SUCCESS : EXIT_FAILURE);
|
||||
}
|
||||
267
Samples/2_Concepts_and_Techniques/scan/scan.cu
Normal file
267
Samples/2_Concepts_and_Techniques/scan/scan.cu
Normal file
@@ -0,0 +1,267 @@
|
||||
/* 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.
|
||||
*/
|
||||
|
||||
#include <assert.h>
|
||||
#include <cooperative_groups.h>
|
||||
|
||||
namespace cg = cooperative_groups;
|
||||
#include <helper_cuda.h>
|
||||
#include "scan_common.h"
|
||||
|
||||
// All three kernels run 512 threads per workgroup
|
||||
// Must be a power of two
|
||||
#define THREADBLOCK_SIZE 256
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Basic scan codelets
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Naive inclusive scan: O(N * log2(N)) operations
|
||||
// Allocate 2 * 'size' local memory, initialize the first half
|
||||
// with 'size' zeros avoiding if(pos >= offset) condition evaluation
|
||||
// and saving instructions
|
||||
inline __device__ uint scan1Inclusive(uint idata, volatile uint *s_Data,
|
||||
uint size, cg::thread_block cta) {
|
||||
uint pos = 2 * threadIdx.x - (threadIdx.x & (size - 1));
|
||||
s_Data[pos] = 0;
|
||||
pos += size;
|
||||
s_Data[pos] = idata;
|
||||
|
||||
for (uint offset = 1; offset < size; offset <<= 1) {
|
||||
cg::sync(cta);
|
||||
uint t = s_Data[pos] + s_Data[pos - offset];
|
||||
cg::sync(cta);
|
||||
s_Data[pos] = t;
|
||||
}
|
||||
|
||||
return s_Data[pos];
|
||||
}
|
||||
|
||||
inline __device__ uint scan1Exclusive(uint idata, volatile uint *s_Data,
|
||||
uint size, cg::thread_block cta) {
|
||||
return scan1Inclusive(idata, s_Data, size, cta) - idata;
|
||||
}
|
||||
|
||||
inline __device__ uint4 scan4Inclusive(uint4 idata4, volatile uint *s_Data,
|
||||
uint size, cg::thread_block cta) {
|
||||
// Level-0 inclusive scan
|
||||
idata4.y += idata4.x;
|
||||
idata4.z += idata4.y;
|
||||
idata4.w += idata4.z;
|
||||
|
||||
// Level-1 exclusive scan
|
||||
uint oval = scan1Exclusive(idata4.w, s_Data, size / 4, cta);
|
||||
|
||||
idata4.x += oval;
|
||||
idata4.y += oval;
|
||||
idata4.z += oval;
|
||||
idata4.w += oval;
|
||||
|
||||
return idata4;
|
||||
}
|
||||
|
||||
// Exclusive vector scan: the array to be scanned is stored
|
||||
// in local thread memory scope as uint4
|
||||
inline __device__ uint4 scan4Exclusive(uint4 idata4, volatile uint *s_Data,
|
||||
uint size, cg::thread_block cta) {
|
||||
uint4 odata4 = scan4Inclusive(idata4, s_Data, size, cta);
|
||||
odata4.x -= idata4.x;
|
||||
odata4.y -= idata4.y;
|
||||
odata4.z -= idata4.z;
|
||||
odata4.w -= idata4.w;
|
||||
return odata4;
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Scan kernels
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
__global__ void scanExclusiveShared(uint4 *d_Dst, uint4 *d_Src, uint size) {
|
||||
// Handle to thread block group
|
||||
cg::thread_block cta = cg::this_thread_block();
|
||||
__shared__ uint s_Data[2 * THREADBLOCK_SIZE];
|
||||
|
||||
uint pos = blockIdx.x * blockDim.x + threadIdx.x;
|
||||
|
||||
// Load data
|
||||
uint4 idata4 = d_Src[pos];
|
||||
|
||||
// Calculate exclusive scan
|
||||
uint4 odata4 = scan4Exclusive(idata4, s_Data, size, cta);
|
||||
|
||||
// Write back
|
||||
d_Dst[pos] = odata4;
|
||||
}
|
||||
|
||||
// Exclusive scan of top elements of bottom-level scans (4 * THREADBLOCK_SIZE)
|
||||
__global__ void scanExclusiveShared2(uint *d_Buf, uint *d_Dst, uint *d_Src,
|
||||
uint N, uint arrayLength) {
|
||||
// Handle to thread block group
|
||||
cg::thread_block cta = cg::this_thread_block();
|
||||
__shared__ uint s_Data[2 * THREADBLOCK_SIZE];
|
||||
|
||||
// Skip loads and stores for inactive threads of last threadblock (pos >= N)
|
||||
uint pos = blockIdx.x * blockDim.x + threadIdx.x;
|
||||
|
||||
// Load top elements
|
||||
// Convert results of bottom-level scan back to inclusive
|
||||
uint idata = 0;
|
||||
|
||||
if (pos < N)
|
||||
idata = d_Dst[(4 * THREADBLOCK_SIZE) - 1 + (4 * THREADBLOCK_SIZE) * pos] +
|
||||
d_Src[(4 * THREADBLOCK_SIZE) - 1 + (4 * THREADBLOCK_SIZE) * pos];
|
||||
|
||||
// Compute
|
||||
uint odata = scan1Exclusive(idata, s_Data, arrayLength, cta);
|
||||
|
||||
// Avoid out-of-bound access
|
||||
if (pos < N) {
|
||||
d_Buf[pos] = odata;
|
||||
}
|
||||
}
|
||||
|
||||
// Final step of large-array scan: combine basic inclusive scan with exclusive
|
||||
// scan of top elements of input arrays
|
||||
__global__ void uniformUpdate(uint4 *d_Data, uint *d_Buffer) {
|
||||
// Handle to thread block group
|
||||
cg::thread_block cta = cg::this_thread_block();
|
||||
__shared__ uint buf;
|
||||
uint pos = blockIdx.x * blockDim.x + threadIdx.x;
|
||||
|
||||
if (threadIdx.x == 0) {
|
||||
buf = d_Buffer[blockIdx.x];
|
||||
}
|
||||
|
||||
cg::sync(cta);
|
||||
|
||||
uint4 data4 = d_Data[pos];
|
||||
data4.x += buf;
|
||||
data4.y += buf;
|
||||
data4.z += buf;
|
||||
data4.w += buf;
|
||||
d_Data[pos] = data4;
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Interface function
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Derived as 32768 (max power-of-two gridDim.x) * 4 * THREADBLOCK_SIZE
|
||||
// Due to scanExclusiveShared<<<>>>() 1D block addressing
|
||||
extern "C" const uint MAX_BATCH_ELEMENTS = 64 * 1048576;
|
||||
extern "C" const uint MIN_SHORT_ARRAY_SIZE = 4;
|
||||
extern "C" const uint MAX_SHORT_ARRAY_SIZE = 4 * THREADBLOCK_SIZE;
|
||||
extern "C" const uint MIN_LARGE_ARRAY_SIZE = 8 * THREADBLOCK_SIZE;
|
||||
extern "C" const uint MAX_LARGE_ARRAY_SIZE =
|
||||
4 * THREADBLOCK_SIZE * THREADBLOCK_SIZE;
|
||||
|
||||
// Internal exclusive scan buffer
|
||||
static uint *d_Buf;
|
||||
|
||||
extern "C" void initScan(void) {
|
||||
checkCudaErrors(
|
||||
cudaMalloc((void **)&d_Buf,
|
||||
(MAX_BATCH_ELEMENTS / (4 * THREADBLOCK_SIZE)) * sizeof(uint)));
|
||||
}
|
||||
|
||||
extern "C" void closeScan(void) { checkCudaErrors(cudaFree(d_Buf)); }
|
||||
|
||||
static uint factorRadix2(uint &log2L, uint L) {
|
||||
if (!L) {
|
||||
log2L = 0;
|
||||
return 0;
|
||||
} else {
|
||||
for (log2L = 0; (L & 1) == 0; L >>= 1, log2L++)
|
||||
;
|
||||
|
||||
return L;
|
||||
}
|
||||
}
|
||||
|
||||
static uint iDivUp(uint dividend, uint divisor) {
|
||||
return ((dividend % divisor) == 0) ? (dividend / divisor)
|
||||
: (dividend / divisor + 1);
|
||||
}
|
||||
|
||||
extern "C" size_t scanExclusiveShort(uint *d_Dst, uint *d_Src, uint batchSize,
|
||||
uint arrayLength) {
|
||||
// Check power-of-two factorization
|
||||
uint log2L;
|
||||
uint factorizationRemainder = factorRadix2(log2L, arrayLength);
|
||||
assert(factorizationRemainder == 1);
|
||||
|
||||
// Check supported size range
|
||||
assert((arrayLength >= MIN_SHORT_ARRAY_SIZE) &&
|
||||
(arrayLength <= MAX_SHORT_ARRAY_SIZE));
|
||||
|
||||
// Check total batch size limit
|
||||
assert((batchSize * arrayLength) <= MAX_BATCH_ELEMENTS);
|
||||
|
||||
// Check all threadblocks to be fully packed with data
|
||||
assert((batchSize * arrayLength) % (4 * THREADBLOCK_SIZE) == 0);
|
||||
|
||||
scanExclusiveShared<<<(batchSize * arrayLength) / (4 * THREADBLOCK_SIZE),
|
||||
THREADBLOCK_SIZE>>>((uint4 *)d_Dst, (uint4 *)d_Src,
|
||||
arrayLength);
|
||||
getLastCudaError("scanExclusiveShared() execution FAILED\n");
|
||||
|
||||
return THREADBLOCK_SIZE;
|
||||
}
|
||||
|
||||
extern "C" size_t scanExclusiveLarge(uint *d_Dst, uint *d_Src, uint batchSize,
|
||||
uint arrayLength) {
|
||||
// Check power-of-two factorization
|
||||
uint log2L;
|
||||
uint factorizationRemainder = factorRadix2(log2L, arrayLength);
|
||||
assert(factorizationRemainder == 1);
|
||||
|
||||
// Check supported size range
|
||||
assert((arrayLength >= MIN_LARGE_ARRAY_SIZE) &&
|
||||
(arrayLength <= MAX_LARGE_ARRAY_SIZE));
|
||||
|
||||
// Check total batch size limit
|
||||
assert((batchSize * arrayLength) <= MAX_BATCH_ELEMENTS);
|
||||
|
||||
scanExclusiveShared<<<(batchSize * arrayLength) / (4 * THREADBLOCK_SIZE),
|
||||
THREADBLOCK_SIZE>>>((uint4 *)d_Dst, (uint4 *)d_Src,
|
||||
4 * THREADBLOCK_SIZE);
|
||||
getLastCudaError("scanExclusiveShared() execution FAILED\n");
|
||||
|
||||
// Not all threadblocks need to be packed with input data:
|
||||
// inactive threads of highest threadblock just don't do global reads and
|
||||
// writes
|
||||
const uint blockCount2 = iDivUp(
|
||||
(batchSize * arrayLength) / (4 * THREADBLOCK_SIZE), THREADBLOCK_SIZE);
|
||||
scanExclusiveShared2<<<blockCount2, THREADBLOCK_SIZE>>>(
|
||||
(uint *)d_Buf, (uint *)d_Dst, (uint *)d_Src,
|
||||
(batchSize * arrayLength) / (4 * THREADBLOCK_SIZE),
|
||||
arrayLength / (4 * THREADBLOCK_SIZE));
|
||||
getLastCudaError("scanExclusiveShared2() execution FAILED\n");
|
||||
|
||||
uniformUpdate<<<(batchSize * arrayLength) / (4 * THREADBLOCK_SIZE),
|
||||
THREADBLOCK_SIZE>>>((uint4 *)d_Dst, (uint *)d_Buf);
|
||||
getLastCudaError("uniformUpdate() execution FAILED\n");
|
||||
|
||||
return THREADBLOCK_SIZE;
|
||||
}
|
||||
65
Samples/2_Concepts_and_Techniques/scan/scan_common.h
Normal file
65
Samples/2_Concepts_and_Techniques/scan/scan_common.h
Normal file
@@ -0,0 +1,65 @@
|
||||
/* 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.
|
||||
*/
|
||||
|
||||
#ifndef SCAN_COMMON_H
|
||||
#define SCAN_COMMON_H
|
||||
|
||||
#include <stdlib.h>
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Shortcut typename
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
typedef unsigned int uint;
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Implementation limits
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
extern "C" const uint MAX_BATCH_ELEMENTS;
|
||||
extern "C" const uint MIN_SHORT_ARRAY_SIZE;
|
||||
extern "C" const uint MAX_SHORT_ARRAY_SIZE;
|
||||
extern "C" const uint MIN_LARGE_ARRAY_SIZE;
|
||||
extern "C" const uint MAX_LARGE_ARRAY_SIZE;
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// CUDA scan
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
extern "C" void initScan(void);
|
||||
extern "C" void closeScan(void);
|
||||
|
||||
extern "C" size_t scanExclusiveShort(uint *d_Dst, uint *d_Src, uint batchSize,
|
||||
uint arrayLength);
|
||||
|
||||
extern "C" size_t scanExclusiveLarge(uint *d_Dst, uint *d_Src, uint batchSize,
|
||||
uint arrayLength);
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Reference CPU scan
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
extern "C" void scanExclusiveHost(uint *dst, uint *src, uint batchSize,
|
||||
uint arrayLength);
|
||||
|
||||
#endif
|
||||
37
Samples/2_Concepts_and_Techniques/scan/scan_gold.cpp
Normal file
37
Samples/2_Concepts_and_Techniques/scan/scan_gold.cpp
Normal file
@@ -0,0 +1,37 @@
|
||||
/* 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.
|
||||
*/
|
||||
|
||||
#include "scan_common.h"
|
||||
|
||||
extern "C" void scanExclusiveHost(uint *dst, uint *src, uint batchSize,
|
||||
uint arrayLength) {
|
||||
for (uint i = 0; i < batchSize; i++, src += arrayLength, dst += arrayLength) {
|
||||
dst[0] = 0;
|
||||
|
||||
for (uint j = 1; j < arrayLength; j++) dst[j] = src[j - 1] + dst[j - 1];
|
||||
}
|
||||
}
|
||||
20
Samples/2_Concepts_and_Techniques/scan/scan_vs2017.sln
Normal file
20
Samples/2_Concepts_and_Techniques/scan/scan_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}") = "scan", "scan_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
|
||||
114
Samples/2_Concepts_and_Techniques/scan/scan_vs2017.vcxproj
Normal file
114
Samples/2_Concepts_and_Techniques/scan/scan_vs2017.vcxproj
Normal file
@@ -0,0 +1,114 @@
|
||||
<?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>scan_vs2017</RootNamespace>
|
||||
<ProjectName>scan</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;</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/scan.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>
|
||||
<ClCompile Include="main.cpp" />
|
||||
<CudaCompile Include="scan.cu" />
|
||||
<ClCompile Include="scan_gold.cpp" />
|
||||
<ClInclude Include="scan_common.h" />
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
20
Samples/2_Concepts_and_Techniques/scan/scan_vs2019.sln
Normal file
20
Samples/2_Concepts_and_Techniques/scan/scan_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}") = "scan", "scan_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
|
||||
110
Samples/2_Concepts_and_Techniques/scan/scan_vs2019.vcxproj
Normal file
110
Samples/2_Concepts_and_Techniques/scan/scan_vs2019.vcxproj
Normal file
@@ -0,0 +1,110 @@
|
||||
<?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>scan_vs2019</RootNamespace>
|
||||
<ProjectName>scan</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>./;$(CudaToolkitDir)/include;../../../Common;</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/scan.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" --threads 0 </AdditionalOptions>
|
||||
<Include>./;../../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(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="'$(Configuration)'=='Release'">
|
||||
<ClCompile>
|
||||
<Optimization>MaxSpeed</Optimization>
|
||||
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MT</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<ClCompile Include="main.cpp" />
|
||||
<CudaCompile Include="scan.cu" />
|
||||
<ClCompile Include="scan_gold.cpp" />
|
||||
<ClInclude Include="scan_common.h" />
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.6.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
20
Samples/2_Concepts_and_Techniques/scan/scan_vs2022.sln
Normal file
20
Samples/2_Concepts_and_Techniques/scan/scan_vs2022.sln
Normal file
@@ -0,0 +1,20 @@
|
||||
|
||||
Microsoft Visual Studio Solution File, Format Version 12.00
|
||||
# Visual Studio 2022
|
||||
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "scan", "scan_vs2022.vcxproj", "{997E0757-EA74-4A4E-A0FC-47D8C8831A15}"
|
||||
EndProject
|
||||
Global
|
||||
GlobalSection(SolutionConfigurationPlatforms) = preSolution
|
||||
Debug|x64 = Debug|x64
|
||||
Release|x64 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(ProjectConfigurationPlatforms) = postSolution
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.ActiveCfg = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Debug|x64.Build.0 = Debug|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.ActiveCfg = Release|x64
|
||||
{997E0757-EA74-4A4E-A0FC-47D8C8831A15}.Release|x64.Build.0 = Release|x64
|
||||
EndGlobalSection
|
||||
GlobalSection(SolutionProperties) = preSolution
|
||||
HideSolutionNode = FALSE
|
||||
EndGlobalSection
|
||||
EndGlobal
|
||||
110
Samples/2_Concepts_and_Techniques/scan/scan_vs2022.vcxproj
Normal file
110
Samples/2_Concepts_and_Techniques/scan/scan_vs2022.vcxproj
Normal file
@@ -0,0 +1,110 @@
|
||||
<?xml version="1.0" encoding="utf-8"?>
|
||||
<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
|
||||
<PropertyGroup>
|
||||
<CUDAPropsPath Condition="'$(CUDAPropsPath)'==''">$(VCTargetsPath)\BuildCustomizations</CUDAPropsPath>
|
||||
</PropertyGroup>
|
||||
<ItemGroup Label="ProjectConfigurations">
|
||||
<ProjectConfiguration Include="Debug|x64">
|
||||
<Configuration>Debug</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
<ProjectConfiguration Include="Release|x64">
|
||||
<Configuration>Release</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
</ItemGroup>
|
||||
<PropertyGroup Label="Globals">
|
||||
<ProjectGuid>{997E0757-EA74-4A4E-A0FC-47D8C8831A15}</ProjectGuid>
|
||||
<RootNamespace>scan_vs2022</RootNamespace>
|
||||
<ProjectName>scan</ProjectName>
|
||||
<CudaToolkitCustomDir />
|
||||
</PropertyGroup>
|
||||
<Import Project="$(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;</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cudart_static.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/scan.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>
|
||||
<ClCompile Include="main.cpp" />
|
||||
<CudaCompile Include="scan.cu" />
|
||||
<ClCompile Include="scan_gold.cpp" />
|
||||
<ClInclude Include="scan_common.h" />
|
||||
</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