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/5_Domain_Specific/recursiveGaussian/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/5_Domain_Specific/recursiveGaussian/.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/5_Domain_Specific/recursiveGaussian/.vscode/extensions.json
vendored
Normal file
7
Samples/5_Domain_Specific/recursiveGaussian/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
||||
{
|
||||
"recommendations": [
|
||||
"nvidia.nsight-vscode-edition",
|
||||
"ms-vscode.cpptools",
|
||||
"ms-vscode.makefile-tools"
|
||||
]
|
||||
}
|
||||
10
Samples/5_Domain_Specific/recursiveGaussian/.vscode/launch.json
vendored
Normal file
10
Samples/5_Domain_Specific/recursiveGaussian/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
||||
{
|
||||
"configurations": [
|
||||
{
|
||||
"name": "CUDA C++: Launch",
|
||||
"type": "cuda-gdb",
|
||||
"request": "launch",
|
||||
"program": "${workspaceFolder}/recursiveGaussian"
|
||||
}
|
||||
]
|
||||
}
|
||||
15
Samples/5_Domain_Specific/recursiveGaussian/.vscode/tasks.json
vendored
Normal file
15
Samples/5_Domain_Specific/recursiveGaussian/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
||||
{
|
||||
"version": "2.0.0",
|
||||
"tasks": [
|
||||
{
|
||||
"label": "sample",
|
||||
"type": "shell",
|
||||
"command": "make dbg=1",
|
||||
"problemMatcher": ["$nvcc"],
|
||||
"group": {
|
||||
"kind": "build",
|
||||
"isDefault": true
|
||||
}
|
||||
}
|
||||
]
|
||||
}
|
||||
367
Samples/5_Domain_Specific/recursiveGaussian/Makefile
Normal file
367
Samples/5_Domain_Specific/recursiveGaussian/Makefile
Normal file
@@ -0,0 +1,367 @@
|
||||
################################################################################
|
||||
# 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
|
||||
|
||||
# This sample is not supported on QNX
|
||||
ifeq ($(TARGET_OS),qnx)
|
||||
$(info >>> WARNING - recursiveGaussian is not supported on QNX - waiving sample <<<)
|
||||
SAMPLE_ENABLED := 0
|
||||
endif
|
||||
|
||||
ALL_LDFLAGS :=
|
||||
ALL_LDFLAGS += $(ALL_CCFLAGS)
|
||||
ALL_LDFLAGS += $(addprefix -Xlinker ,$(LDFLAGS))
|
||||
ALL_LDFLAGS += $(addprefix -Xlinker ,$(EXTRA_LDFLAGS))
|
||||
|
||||
# Common includes and paths for CUDA
|
||||
INCLUDES := -I../../../Common
|
||||
LIBRARIES :=
|
||||
|
||||
################################################################################
|
||||
|
||||
# Makefile include to help find GL Libraries
|
||||
include ./findgllib.mk
|
||||
|
||||
# OpenGL specific libraries
|
||||
ifeq ($(TARGET_OS),darwin)
|
||||
# Mac OSX specific libraries and paths to include
|
||||
LIBRARIES += -L/System/Library/Frameworks/OpenGL.framework/Libraries
|
||||
LIBRARIES += -lGL -lGLU
|
||||
ALL_LDFLAGS += -Xlinker -framework -Xlinker GLUT
|
||||
else
|
||||
LIBRARIES += $(GLLINK)
|
||||
LIBRARIES += -lGL -lGLU -lglut
|
||||
endif
|
||||
|
||||
# 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: recursiveGaussian
|
||||
|
||||
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
|
||||
|
||||
recursiveGaussian.o:recursiveGaussian.cpp
|
||||
$(EXEC) $(NVCC) $(INCLUDES) $(ALL_CCFLAGS) $(GENCODE_FLAGS) -o $@ -c $<
|
||||
|
||||
recursiveGaussian_cuda.o:recursiveGaussian_cuda.cu
|
||||
$(EXEC) $(NVCC) $(INCLUDES) $(ALL_CCFLAGS) $(GENCODE_FLAGS) -o $@ -c $<
|
||||
|
||||
recursiveGaussian: recursiveGaussian.o recursiveGaussian_cuda.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) ./recursiveGaussian
|
||||
|
||||
testrun: build
|
||||
$(EXEC) ./recursiveGaussian -sigma=10 -file=ref_10.ppm
|
||||
$(EXEC) ./recursiveGaussian -sigma=14 -file=ref_14.ppm
|
||||
$(EXEC) ./recursiveGaussian -sigma=18 -file=ref_18.ppm
|
||||
$(EXEC) ./recursiveGaussian -sigma=22 -file=ref_22.ppm
|
||||
|
||||
clean:
|
||||
rm -f recursiveGaussian recursiveGaussian.o recursiveGaussian_cuda.o
|
||||
rm -rf ../../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)/recursiveGaussian
|
||||
|
||||
clobber: clean
|
||||
111
Samples/5_Domain_Specific/recursiveGaussian/NsightEclipse.xml
Normal file
111
Samples/5_Domain_Specific/recursiveGaussian/NsightEclipse.xml
Normal file
@@ -0,0 +1,111 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<!DOCTYPE entry SYSTEM "SamplesInfo.dtd">
|
||||
<entry>
|
||||
<name>recursiveGaussian</name>
|
||||
<cuda_api_list>
|
||||
<toolkit>cudaFree</toolkit>
|
||||
<toolkit>cudaGraphicsMapResources</toolkit>
|
||||
<toolkit>cudaGraphicsGLRegisterBuffer</toolkit>
|
||||
<toolkit>cudaGraphicsResourceGetMappedPointer</toolkit>
|
||||
<toolkit>cudaDeviceSynchronize</toolkit>
|
||||
<toolkit>cudaMalloc</toolkit>
|
||||
<toolkit>cudaGraphicsUnregisterResource</toolkit>
|
||||
<toolkit>cudaGraphicsUnmapResources</toolkit>
|
||||
<toolkit>cudaMemcpy</toolkit>
|
||||
<toolkit>cudaGetDeviceProperties</toolkit>
|
||||
<toolkit>cudaGetDevice</toolkit>
|
||||
</cuda_api_list>
|
||||
<description><![CDATA[This sample implements a Gaussian blur using Deriche's recursive method. The advantage of this method is that the execution time is independent of the filter width.]]></description>
|
||||
<devicecompilation>whole</devicecompilation>
|
||||
<files>
|
||||
<file>data\teapot512.ppm</file>
|
||||
<file>data\ref_10.ppm</file>
|
||||
<file>data\ref_14.ppm</file>
|
||||
<file>data\ref_18.ppm</file>
|
||||
<file>data\ref_22.ppm</file>
|
||||
</files>
|
||||
<includepaths>
|
||||
<path>./</path>
|
||||
<path>../</path>
|
||||
<path>../../../Common</path>
|
||||
</includepaths>
|
||||
<keyconcepts>
|
||||
<concept level="advanced">Graphics Interop</concept>
|
||||
<concept level="advanced">Image Processing</concept>
|
||||
</keyconcepts>
|
||||
<keywords>
|
||||
<keyword>GPGPU</keyword>
|
||||
<keyword>CUDA</keyword>
|
||||
<keyword>Image Processing</keyword>
|
||||
<keyword>box filter</keyword>
|
||||
<keyword>opengl</keyword>
|
||||
<keyword>openGL</keyword>
|
||||
</keywords>
|
||||
<libraries>
|
||||
<library>GLU</library>
|
||||
<library>GL</library>
|
||||
<library framework="true" os="macosx">GLUT</library>
|
||||
<library os="linux">GLEW</library>
|
||||
<library os="linux">glut</library>
|
||||
<library os="linux">X11</library>
|
||||
</libraries>
|
||||
<librarypaths>
|
||||
<path arch="x86_64" os="linux">../../../common/lib/linux/x86_64</path>
|
||||
<path arch="armv7l" os="linux">../../../common/lib/linux/armv7l</path>
|
||||
<path os="macosx">../../../common/lib/darwin</path>
|
||||
</librarypaths>
|
||||
<nsight_eclipse>true</nsight_eclipse>
|
||||
<primary_file>recursiveGaussian.cpp</primary_file>
|
||||
<qatests>
|
||||
<qatest>-sigma=10 -file=ref_10.ppm</qatest>
|
||||
<qatest>-sigma=14 -file=ref_14.ppm</qatest>
|
||||
<qatest>-sigma=18 -file=ref_18.ppm</qatest>
|
||||
<qatest>-sigma=22 -file=ref_22.ppm</qatest>
|
||||
</qatests>
|
||||
<required_dependencies>
|
||||
<dependency>X11</dependency>
|
||||
<dependency>GL</dependency>
|
||||
</required_dependencies>
|
||||
<scopes>
|
||||
<scope>2:Graphics Interop</scope>
|
||||
<scope>2:Image Processing</scope>
|
||||
<scope>2:Computer Vision</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>
|
||||
</supported_envs>
|
||||
<supported_sm_architectures>
|
||||
<include>all</include>
|
||||
</supported_sm_architectures>
|
||||
<title>Recursive Gaussian Filter</title>
|
||||
<type>exe</type>
|
||||
</entry>
|
||||
74
Samples/5_Domain_Specific/recursiveGaussian/README.md
Normal file
74
Samples/5_Domain_Specific/recursiveGaussian/README.md
Normal file
@@ -0,0 +1,74 @@
|
||||
# recursiveGaussian - Recursive Gaussian Filter
|
||||
|
||||
## Description
|
||||
|
||||
This sample implements a Gaussian blur using Deriche's recursive method. The advantage of this method is that the execution time is independent of the filter width.
|
||||
|
||||
## Key Concepts
|
||||
|
||||
Graphics Interop, Image Processing
|
||||
|
||||
## 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, armv7l
|
||||
|
||||
## CUDA APIs involved
|
||||
|
||||
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
|
||||
cudaFree, cudaGraphicsMapResources, cudaGraphicsGLRegisterBuffer, cudaGraphicsResourceGetMappedPointer, cudaDeviceSynchronize, cudaMalloc, cudaGraphicsUnregisterResource, cudaGraphicsUnmapResources, cudaMemcpy, cudaGetDeviceProperties, cudaGetDevice
|
||||
|
||||
## Dependencies needed to build/run
|
||||
[X11](../../README.md#x11), [GL](../../README.md#gl)
|
||||
|
||||
## 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, 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=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)
|
||||
|
||||
BIN
Samples/5_Domain_Specific/recursiveGaussian/data/ref_10.ppm
Normal file
BIN
Samples/5_Domain_Specific/recursiveGaussian/data/ref_10.ppm
Normal file
Binary file not shown.
BIN
Samples/5_Domain_Specific/recursiveGaussian/data/ref_14.ppm
Normal file
BIN
Samples/5_Domain_Specific/recursiveGaussian/data/ref_14.ppm
Normal file
Binary file not shown.
BIN
Samples/5_Domain_Specific/recursiveGaussian/data/ref_18.ppm
Normal file
BIN
Samples/5_Domain_Specific/recursiveGaussian/data/ref_18.ppm
Normal file
Binary file not shown.
BIN
Samples/5_Domain_Specific/recursiveGaussian/data/ref_22.ppm
Normal file
BIN
Samples/5_Domain_Specific/recursiveGaussian/data/ref_22.ppm
Normal file
Binary file not shown.
BIN
Samples/5_Domain_Specific/recursiveGaussian/data/teapot512.ppm
Normal file
BIN
Samples/5_Domain_Specific/recursiveGaussian/data/teapot512.ppm
Normal file
Binary file not shown.
145
Samples/5_Domain_Specific/recursiveGaussian/findgllib.mk
Normal file
145
Samples/5_Domain_Specific/recursiveGaussian/findgllib.mk
Normal file
@@ -0,0 +1,145 @@
|
||||
################################################################################
|
||||
# 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.
|
||||
#
|
||||
################################################################################
|
||||
#
|
||||
# findgllib.mk is used to find the necessary GL Libraries for specific distributions
|
||||
# this is supported on Mac OSX and Linux Platforms
|
||||
#
|
||||
################################################################################
|
||||
|
||||
# Determine OS platform and unix distribution
|
||||
ifeq ("$(TARGET_OS)","linux")
|
||||
# first search lsb_release
|
||||
DISTRO := $(shell lsb_release -i -s 2>/dev/null | tr "[:upper:]" "[:lower:]")
|
||||
ifeq ("$(DISTRO)","")
|
||||
# second search and parse /etc/issue
|
||||
DISTRO := $(shell awk '{print $$1}' /etc/issue | tr -d "[:space:]" | sed -e "/^$$/d" | tr "[:upper:]" "[:lower:]")
|
||||
# ensure data from /etc/issue is valid
|
||||
ifeq (,$(filter $(DISTRO),ubuntu fedora red rhel centos suse))
|
||||
DISTRO :=
|
||||
endif
|
||||
ifeq ("$(DISTRO)","")
|
||||
# third, we can search in /etc/os-release or /etc/{distro}-release
|
||||
DISTRO := $(shell awk '/ID/' /etc/*-release | sed 's/ID=//' | grep -v "VERSION" | grep -v "ID" | grep -v "DISTRIB")
|
||||
endif
|
||||
endif
|
||||
endif
|
||||
|
||||
ifeq ("$(TARGET_OS)","linux")
|
||||
# $(info) >> findgllib.mk -> LINUX path <<<)
|
||||
# Each set of Linux Distros have different paths for where to find their OpenGL libraries reside
|
||||
UBUNTU = $(shell echo $(DISTRO) | grep -i ubuntu >/dev/null 2>&1; echo $$?)
|
||||
FEDORA = $(shell echo $(DISTRO) | grep -i fedora >/dev/null 2>&1; echo $$?)
|
||||
RHEL = $(shell echo $(DISTRO) | grep -i 'red\|rhel' >/dev/null 2>&1; echo $$?)
|
||||
CENTOS = $(shell echo $(DISTRO) | grep -i centos >/dev/null 2>&1; echo $$?)
|
||||
SUSE = $(shell echo $(DISTRO) | grep -i 'suse\|sles' >/dev/null 2>&1; echo $$?)
|
||||
ifeq ("$(UBUNTU)","0")
|
||||
ifeq ($(HOST_ARCH)-$(TARGET_ARCH),x86_64-armv7l)
|
||||
GLPATH := /usr/arm-linux-gnueabihf/lib
|
||||
GLLINK := -L/usr/arm-linux-gnueabihf/lib
|
||||
ifneq ($(TARGET_FS),)
|
||||
GLPATH += $(TARGET_FS)/usr/lib/arm-linux-gnueabihf
|
||||
GLLINK += -L$(TARGET_FS)/usr/lib/arm-linux-gnueabihf
|
||||
endif
|
||||
else ifeq ($(HOST_ARCH)-$(TARGET_ARCH),x86_64-aarch64)
|
||||
GLPATH := /usr/aarch64-linux-gnu/lib
|
||||
GLLINK := -L/usr/aarch64-linux-gnu/lib
|
||||
ifneq ($(TARGET_FS),)
|
||||
GLPATH += $(TARGET_FS)/usr/lib
|
||||
GLPATH += $(TARGET_FS)/usr/lib/aarch64-linux-gnu
|
||||
GLLINK += -L$(TARGET_FS)/usr/lib/aarch64-linux-gnu
|
||||
endif
|
||||
else ifeq ($(HOST_ARCH)-$(TARGET_ARCH),x86_64-ppc64le)
|
||||
GLPATH := /usr/powerpc64le-linux-gnu/lib
|
||||
GLLINK := -L/usr/powerpc64le-linux-gnu/lib
|
||||
else
|
||||
UBUNTU_PKG_NAME = $(shell which dpkg >/dev/null 2>&1 && dpkg -l 'nvidia-*' | grep '^ii' | awk '{print $$2}' | head -1)
|
||||
ifneq ("$(UBUNTU_PKG_NAME)","")
|
||||
GLPATH ?= /usr/lib/$(UBUNTU_PKG_NAME)
|
||||
GLLINK ?= -L/usr/lib/$(UBUNTU_PKG_NAME)
|
||||
endif
|
||||
|
||||
DFLT_PATH ?= /usr/lib
|
||||
endif
|
||||
endif
|
||||
ifeq ("$(SUSE)","0")
|
||||
GLPATH ?= /usr/X11R6/lib64
|
||||
GLLINK ?= -L/usr/X11R6/lib64
|
||||
DFLT_PATH ?= /usr/lib64
|
||||
endif
|
||||
ifeq ("$(FEDORA)","0")
|
||||
GLPATH ?= /usr/lib64/nvidia
|
||||
GLLINK ?= -L/usr/lib64/nvidia
|
||||
DFLT_PATH ?= /usr/lib64
|
||||
endif
|
||||
ifeq ("$(RHEL)","0")
|
||||
GLPATH ?= /usr/lib64/nvidia
|
||||
GLLINK ?= -L/usr/lib64/nvidia
|
||||
DFLT_PATH ?= /usr/lib64
|
||||
endif
|
||||
ifeq ("$(CENTOS)","0")
|
||||
GLPATH ?= /usr/lib64/nvidia
|
||||
GLLINK ?= -L/usr/lib64/nvidia
|
||||
DFLT_PATH ?= /usr/lib64
|
||||
endif
|
||||
|
||||
# find libGL, libGLU
|
||||
GLLIB := $(shell find -L $(GLPATH) $(DFLT_PATH) -name libGL.so -print 2>/dev/null)
|
||||
GLULIB := $(shell find -L $(GLPATH) $(DFLT_PATH) -name libGLU.so -print 2>/dev/null)
|
||||
|
||||
ifeq ("$(GLLIB)","")
|
||||
$(info >>> WARNING - libGL.so not found, refer to CUDA Getting Started Guide for how to find and install them. <<<)
|
||||
SAMPLE_ENABLED := 0
|
||||
endif
|
||||
ifeq ("$(GLULIB)","")
|
||||
$(info >>> WARNING - libGLU.so not found, refer to CUDA Getting Started Guide for how to find and install them. <<<)
|
||||
SAMPLE_ENABLED := 0
|
||||
endif
|
||||
|
||||
HEADER_SEARCH_PATH ?= $(TARGET_FS)/usr/include
|
||||
ifeq ($(HOST_ARCH)-$(TARGET_ARCH)-$(TARGET_OS),x86_64-armv7l-linux)
|
||||
HEADER_SEARCH_PATH += /usr/arm-linux-gnueabihf/include
|
||||
else ifeq ($(HOST_ARCH)-$(TARGET_ARCH)-$(TARGET_OS),x86_64-aarch64-linux)
|
||||
HEADER_SEARCH_PATH += /usr/aarch64-linux-gnu/include
|
||||
endif
|
||||
|
||||
GLHEADER := $(shell find -L $(HEADER_SEARCH_PATH) -name gl.h -print 2>/dev/null)
|
||||
GLUHEADER := $(shell find -L $(HEADER_SEARCH_PATH) -name glu.h -print 2>/dev/null)
|
||||
|
||||
ifeq ("$(GLHEADER)","")
|
||||
$(info >>> WARNING - gl.h not found, refer to CUDA Getting Started Guide for how to find and install them. <<<)
|
||||
SAMPLE_ENABLED := 0
|
||||
endif
|
||||
ifeq ("$(GLUHEADER)","")
|
||||
$(info >>> WARNING - glu.h not found, refer to CUDA Getting Started Guide for how to find and install them. <<<)
|
||||
SAMPLE_ENABLED := 0
|
||||
endif
|
||||
else
|
||||
# This would be the Mac OS X path if we had to do anything special
|
||||
endif
|
||||
|
||||
@@ -0,0 +1,535 @@
|
||||
/* 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.
|
||||
*/
|
||||
|
||||
/*
|
||||
Recursive Gaussian filter
|
||||
sgreen 8/1/08
|
||||
|
||||
This code sample implements a Gaussian blur using Deriche's recursive method:
|
||||
http://citeseer.ist.psu.edu/deriche93recursively.html
|
||||
|
||||
This is similar to the box filter sample in the SDK, but it uses the previous
|
||||
outputs of the filter as well as the previous inputs. This is also known as an
|
||||
IIR (infinite impulse response) filter, since its response to an input impulse
|
||||
can last forever.
|
||||
|
||||
The main advantage of this method is that the execution time is independent of
|
||||
the filter width.
|
||||
|
||||
The GPU processes columns of the image in parallel. To avoid uncoalesced reads
|
||||
for the row pass we transpose the image and then transpose it back again
|
||||
afterwards.
|
||||
|
||||
The implementation is based on code from the CImg library:
|
||||
http://cimg.sourceforge.net/
|
||||
Thanks to David Tschumperl<72> and all the CImg contributors!
|
||||
*/
|
||||
|
||||
#pragma warning(disable : 4819)
|
||||
|
||||
// OpenGL Graphics includes
|
||||
#include <helper_gl.h>
|
||||
#if defined(__APPLE__) || defined(MACOSX)
|
||||
#pragma clang diagnostic ignored "-Wdeprecated-declarations"
|
||||
#include <GLUT/glut.h>
|
||||
#ifndef glutCloseFunc
|
||||
#define glutCloseFunc glutWMCloseFunc
|
||||
#endif
|
||||
#else
|
||||
#include <GL/freeglut.h>
|
||||
#endif
|
||||
|
||||
// CUDA includes and interop headers
|
||||
#include <cuda_runtime.h>
|
||||
#include <cuda_gl_interop.h>
|
||||
|
||||
// CUDA utilities and system includes
|
||||
#include <helper_functions.h>
|
||||
#include <helper_cuda.h> // includes cuda.h and cuda_runtime_api.h
|
||||
|
||||
// Includes
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
#include <math.h>
|
||||
|
||||
#define MAX(a, b) ((a > b) ? a : b)
|
||||
|
||||
#define USE_SIMPLE_FILTER 0
|
||||
|
||||
#define MAX_EPSILON_ERROR 5.0f
|
||||
#define THRESHOLD 0.15f
|
||||
|
||||
// Define the files that are to be save and the reference images for validation
|
||||
const char *sOriginal[] = {"teapot512_10.ppm", "teapot512_14.ppm", "teapot512_18.ppm",
|
||||
"teapot512_22.ppm", NULL};
|
||||
|
||||
const char *sReference[] = {"ref_10.ppm", "ref_14.ppm", "ref_18.ppm",
|
||||
"ref_22.ppm", NULL};
|
||||
|
||||
const char *image_filename = "teapot512.ppm";
|
||||
float sigma = 10.0f;
|
||||
int order = 0;
|
||||
int nthreads = 64; // number of threads per block
|
||||
|
||||
unsigned int width, height;
|
||||
unsigned int *h_img = NULL;
|
||||
unsigned int *d_img = NULL;
|
||||
unsigned int *d_temp = NULL;
|
||||
|
||||
GLuint pbo = 0; // OpenGL pixel buffer object
|
||||
GLuint texid = 0; // texture
|
||||
|
||||
cudaGraphicsResource_t cuda_vbo_resource;
|
||||
|
||||
StopWatchInterface *timer = 0;
|
||||
|
||||
// Auto-Verification Code
|
||||
const int frameCheckNumber = 4;
|
||||
int fpsCount = 0; // FPS count for averaging
|
||||
int fpsLimit = 1; // FPS limit for sampling
|
||||
unsigned int frameCount = 0;
|
||||
|
||||
int *pArgc = NULL;
|
||||
char **pArgv = NULL;
|
||||
|
||||
bool runBenchmark = false;
|
||||
|
||||
const char *sSDKsample = "CUDA Recursive Gaussian";
|
||||
|
||||
extern "C" void transpose(unsigned int *d_src, unsigned int *d_dest,
|
||||
unsigned int width, int height);
|
||||
|
||||
extern "C" void gaussianFilterRGBA(unsigned int *d_src, unsigned int *d_dest,
|
||||
unsigned int *d_temp, int width, int height,
|
||||
float sigma, int order, int nthreads);
|
||||
|
||||
void cleanup();
|
||||
|
||||
void computeFPS() {
|
||||
frameCount++;
|
||||
fpsCount++;
|
||||
|
||||
if (fpsCount == fpsLimit) {
|
||||
char fps[256];
|
||||
float ifps = 1.f / (sdkGetAverageTimerValue(&timer) / 1000.f);
|
||||
sprintf(fps, "%s (sigma=%4.2f): %3.1f fps", sSDKsample, sigma, ifps);
|
||||
|
||||
glutSetWindowTitle(fps);
|
||||
fpsCount = 0;
|
||||
|
||||
fpsLimit = ftoi(MAX(ifps, 1.f));
|
||||
sdkResetTimer(&timer);
|
||||
}
|
||||
}
|
||||
|
||||
// display results using OpenGL
|
||||
void display() {
|
||||
sdkStartTimer(&timer);
|
||||
|
||||
// execute filter, writing results to pbo
|
||||
unsigned int *d_result;
|
||||
checkCudaErrors(cudaGraphicsMapResources(1, &cuda_vbo_resource, 0));
|
||||
size_t num_bytes;
|
||||
checkCudaErrors(cudaGraphicsResourceGetMappedPointer(
|
||||
(void **)&d_result, &num_bytes, cuda_vbo_resource));
|
||||
gaussianFilterRGBA(d_img, d_result, d_temp, width, height, sigma, order,
|
||||
nthreads);
|
||||
|
||||
// unmap buffer object
|
||||
checkCudaErrors(cudaGraphicsUnmapResources(1, &cuda_vbo_resource, 0));
|
||||
|
||||
// load texture from pbo
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, pbo);
|
||||
glBindTexture(GL_TEXTURE_2D, texid);
|
||||
glPixelStorei(GL_UNPACK_ALIGNMENT, 1);
|
||||
glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, width, height, GL_RGBA,
|
||||
GL_UNSIGNED_BYTE, 0);
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, 0);
|
||||
|
||||
// display results
|
||||
glClear(GL_COLOR_BUFFER_BIT);
|
||||
|
||||
glEnable(GL_TEXTURE_2D);
|
||||
glDisable(GL_DEPTH_TEST);
|
||||
|
||||
glBegin(GL_QUADS);
|
||||
glTexCoord2f(0, 1);
|
||||
glVertex2f(0, 0);
|
||||
glTexCoord2f(1, 1);
|
||||
glVertex2f(1, 0);
|
||||
glTexCoord2f(1, 0);
|
||||
glVertex2f(1, 1);
|
||||
glTexCoord2f(0, 0);
|
||||
glVertex2f(0, 1);
|
||||
glEnd();
|
||||
|
||||
glDisable(GL_TEXTURE_2D);
|
||||
glutSwapBuffers();
|
||||
|
||||
sdkStopTimer(&timer);
|
||||
|
||||
computeFPS();
|
||||
}
|
||||
|
||||
void idle() { glutPostRedisplay(); }
|
||||
|
||||
void cleanup() {
|
||||
sdkDeleteTimer(&timer);
|
||||
|
||||
checkCudaErrors(cudaFree(d_img));
|
||||
checkCudaErrors(cudaFree(d_temp));
|
||||
|
||||
if (!runBenchmark) {
|
||||
if (pbo) {
|
||||
// unregister this buffer object with CUDA
|
||||
checkCudaErrors(cudaGraphicsUnregisterResource(cuda_vbo_resource));
|
||||
glDeleteBuffers(1, &pbo);
|
||||
}
|
||||
|
||||
if (texid) {
|
||||
glDeleteTextures(1, &texid);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
void keyboard(unsigned char key, int x, int y) {
|
||||
switch (key) {
|
||||
case 27:
|
||||
#if defined(__APPLE__) || defined(MACOSX)
|
||||
exit(EXIT_SUCCESS);
|
||||
#else
|
||||
glutDestroyWindow(glutGetWindow());
|
||||
return;
|
||||
#endif
|
||||
break;
|
||||
|
||||
case '=':
|
||||
sigma += 0.1f;
|
||||
break;
|
||||
|
||||
case '-':
|
||||
sigma -= 0.1f;
|
||||
|
||||
if (sigma < 0.0) {
|
||||
sigma = 0.0f;
|
||||
}
|
||||
|
||||
break;
|
||||
|
||||
case '+':
|
||||
sigma += 1.0f;
|
||||
break;
|
||||
|
||||
case '_':
|
||||
sigma -= 1.0f;
|
||||
|
||||
if (sigma < 0.0) {
|
||||
sigma = 0.0f;
|
||||
}
|
||||
|
||||
break;
|
||||
|
||||
case '0':
|
||||
order = 0;
|
||||
break;
|
||||
|
||||
case '1':
|
||||
order = 1;
|
||||
sigma = 0.5f;
|
||||
break;
|
||||
|
||||
case '2':
|
||||
order = 2;
|
||||
sigma = 0.5f;
|
||||
break;
|
||||
|
||||
default:
|
||||
break;
|
||||
}
|
||||
|
||||
printf("sigma = %f\n", sigma);
|
||||
glutPostRedisplay();
|
||||
}
|
||||
|
||||
void reshape(int x, int y) {
|
||||
glViewport(0, 0, x, y);
|
||||
|
||||
glMatrixMode(GL_MODELVIEW);
|
||||
glLoadIdentity();
|
||||
|
||||
glMatrixMode(GL_PROJECTION);
|
||||
glLoadIdentity();
|
||||
glOrtho(0.0, 1.0, 0.0, 1.0, 0.0, 1.0);
|
||||
}
|
||||
|
||||
void initCudaBuffers() {
|
||||
unsigned int size = width * height * sizeof(unsigned int);
|
||||
|
||||
// allocate device memory
|
||||
checkCudaErrors(cudaMalloc((void **)&d_img, size));
|
||||
checkCudaErrors(cudaMalloc((void **)&d_temp, size));
|
||||
|
||||
checkCudaErrors(cudaMemcpy(d_img, h_img, size, cudaMemcpyHostToDevice));
|
||||
|
||||
sdkCreateTimer(&timer);
|
||||
}
|
||||
|
||||
void initGLBuffers() {
|
||||
// create pixel buffer object to store final image
|
||||
glGenBuffers(1, &pbo);
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, pbo);
|
||||
glBufferData(GL_PIXEL_UNPACK_BUFFER_ARB, width * height * sizeof(GLubyte) * 4,
|
||||
h_img, GL_STREAM_DRAW_ARB);
|
||||
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, 0);
|
||||
checkCudaErrors(cudaGraphicsGLRegisterBuffer(
|
||||
&cuda_vbo_resource, pbo, cudaGraphicsRegisterFlagsWriteDiscard));
|
||||
|
||||
// create texture for display
|
||||
glGenTextures(1, &texid);
|
||||
glBindTexture(GL_TEXTURE_2D, texid);
|
||||
glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA8, width, height, 0, GL_RGBA,
|
||||
GL_UNSIGNED_BYTE, NULL);
|
||||
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
|
||||
glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
|
||||
glBindTexture(GL_TEXTURE_2D, 0);
|
||||
}
|
||||
|
||||
void initGL(int *argc, char **argv) {
|
||||
glutInit(argc, argv);
|
||||
glutInitDisplayMode(GLUT_RGB | GLUT_DOUBLE);
|
||||
glutInitWindowSize(width, height);
|
||||
glutCreateWindow(sSDKsample);
|
||||
glutDisplayFunc(display);
|
||||
glutKeyboardFunc(keyboard);
|
||||
glutReshapeFunc(reshape);
|
||||
glutIdleFunc(idle);
|
||||
|
||||
#if defined(__APPLE__) || defined(MACOSX)
|
||||
atexit(cleanup);
|
||||
#else
|
||||
glutCloseFunc(cleanup);
|
||||
#endif
|
||||
|
||||
printf("Press '+' and '-' to change filter width\n");
|
||||
printf("0, 1, 2 - change filter order\n");
|
||||
|
||||
if (!isGLVersionSupported(2, 0) ||
|
||||
!areGLExtensionsSupported(
|
||||
"GL_ARB_vertex_buffer_object GL_ARB_pixel_buffer_object")) {
|
||||
fprintf(stderr, "Required OpenGL extensions missing.");
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
}
|
||||
|
||||
void benchmark(int iterations) {
|
||||
// allocate memory for result
|
||||
unsigned int *d_result;
|
||||
unsigned int size = width * height * sizeof(unsigned int);
|
||||
checkCudaErrors(cudaMalloc((void **)&d_result, size));
|
||||
|
||||
// warm-up
|
||||
gaussianFilterRGBA(d_img, d_result, d_temp, width, height, sigma, order,
|
||||
nthreads);
|
||||
|
||||
checkCudaErrors(cudaDeviceSynchronize());
|
||||
sdkStartTimer(&timer);
|
||||
|
||||
// execute the kernel
|
||||
for (int i = 0; i < iterations; i++) {
|
||||
gaussianFilterRGBA(d_img, d_result, d_temp, width, height, sigma, order,
|
||||
nthreads);
|
||||
}
|
||||
|
||||
checkCudaErrors(cudaDeviceSynchronize());
|
||||
sdkStopTimer(&timer);
|
||||
|
||||
// check if kernel execution generated an error
|
||||
getLastCudaError("Kernel execution failed");
|
||||
|
||||
printf("Processing time: %f (ms)\n", sdkGetTimerValue(&timer));
|
||||
printf("%.2f Mpixels/sec\n",
|
||||
(width * height * iterations / (sdkGetTimerValue(&timer) / 1000.0f)) /
|
||||
1e6);
|
||||
|
||||
checkCudaErrors(cudaFree(d_result));
|
||||
}
|
||||
|
||||
bool runSingleTest(const char *ref_file, const char *exec_path) {
|
||||
// allocate memory for result
|
||||
int nTotalErrors = 0;
|
||||
unsigned int *d_result;
|
||||
unsigned int size = width * height * sizeof(unsigned int);
|
||||
checkCudaErrors(cudaMalloc((void **)&d_result, size));
|
||||
|
||||
// warm-up
|
||||
gaussianFilterRGBA(d_img, d_result, d_temp, width, height, sigma, order,
|
||||
nthreads);
|
||||
|
||||
checkCudaErrors(cudaDeviceSynchronize());
|
||||
sdkStartTimer(&timer);
|
||||
|
||||
gaussianFilterRGBA(d_img, d_result, d_temp, width, height, sigma, order,
|
||||
nthreads);
|
||||
checkCudaErrors(cudaDeviceSynchronize());
|
||||
getLastCudaError("Kernel execution failed");
|
||||
sdkStopTimer(&timer);
|
||||
|
||||
unsigned char *h_result = (unsigned char *)malloc(width * height * 4);
|
||||
checkCudaErrors(cudaMemcpy(h_result, d_result, width * height * 4,
|
||||
cudaMemcpyDeviceToHost));
|
||||
|
||||
char dump_file[1024];
|
||||
sprintf(dump_file, "teapot512_%02d.ppm", (int)sigma);
|
||||
sdkSavePPM4ub(dump_file, h_result, width, height);
|
||||
|
||||
if (!sdkComparePPM(dump_file, sdkFindFilePath(ref_file, exec_path),
|
||||
MAX_EPSILON_ERROR, THRESHOLD, false)) {
|
||||
nTotalErrors++;
|
||||
}
|
||||
|
||||
printf("Processing time: %f (ms)\n", sdkGetTimerValue(&timer));
|
||||
printf("%.2f Mpixels/sec\n",
|
||||
(width * height / (sdkGetTimerValue(&timer) / 1000.0f)) / 1e6);
|
||||
|
||||
checkCudaErrors(cudaFree(d_result));
|
||||
free(h_result);
|
||||
|
||||
printf("Summary: %d errors!\n", nTotalErrors);
|
||||
|
||||
printf(nTotalErrors == 0 ? "Test passed\n" : "Test failed!\n");
|
||||
return (nTotalErrors == 0);
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Program main
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
int main(int argc, char **argv) {
|
||||
pArgc = &argc;
|
||||
pArgv = argv;
|
||||
char *ref_file = NULL;
|
||||
|
||||
#if defined(__linux__)
|
||||
setenv("DISPLAY", ":0", 0);
|
||||
#endif
|
||||
|
||||
printf("%s Starting...\n\n", sSDKsample);
|
||||
|
||||
printf(
|
||||
"NOTE: The CUDA Samples are not meant for performance measurements. "
|
||||
"Results may vary when GPU Boost is enabled.\n\n");
|
||||
|
||||
// use command-line specified CUDA device, otherwise use device with highest
|
||||
// Gflops/s
|
||||
if (argc > 1) {
|
||||
if (checkCmdLineFlag(argc, (const char **)argv, "file")) {
|
||||
getCmdLineArgumentString(argc, (const char **)argv, "file", &ref_file);
|
||||
fpsLimit = frameCheckNumber;
|
||||
}
|
||||
}
|
||||
|
||||
// Get the path of the filename
|
||||
char *filename;
|
||||
|
||||
if (getCmdLineArgumentString(argc, (const char **)argv, "image", &filename)) {
|
||||
image_filename = filename;
|
||||
}
|
||||
|
||||
// load image
|
||||
char *image_path = sdkFindFilePath(image_filename, argv[0]);
|
||||
|
||||
if (image_path == NULL) {
|
||||
fprintf(stderr, "Error unable to find and load image file: '%s'\n",
|
||||
image_filename);
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
|
||||
sdkLoadPPM4ub(image_path, (unsigned char **)&h_img, &width, &height);
|
||||
|
||||
if (!h_img) {
|
||||
printf("Error unable to load PPM file: '%s'\n", image_path);
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
|
||||
printf("Loaded '%s', %d x %d pixels\n", image_path, width, height);
|
||||
|
||||
if (checkCmdLineFlag(argc, (const char **)argv, "threads")) {
|
||||
nthreads = getCmdLineArgumentInt(argc, (const char **)argv, "threads");
|
||||
}
|
||||
|
||||
if (checkCmdLineFlag(argc, (const char **)argv, "sigma")) {
|
||||
sigma = getCmdLineArgumentFloat(argc, (const char **)argv, "sigma");
|
||||
}
|
||||
|
||||
runBenchmark = checkCmdLineFlag(argc, (const char **)argv, "benchmark");
|
||||
|
||||
int device;
|
||||
struct cudaDeviceProp prop;
|
||||
cudaGetDevice(&device);
|
||||
cudaGetDeviceProperties(&prop, device);
|
||||
|
||||
if (!strncmp("Tesla", prop.name, 5)) {
|
||||
printf(
|
||||
"Tesla card detected, running the test in benchmark mode (no OpenGL "
|
||||
"display)\n");
|
||||
// runBenchmark = true;
|
||||
runBenchmark = true;
|
||||
}
|
||||
|
||||
// Benchmark or AutoTest mode detected, no OpenGL
|
||||
if (runBenchmark == true || ref_file != NULL) {
|
||||
findCudaDevice(argc, (const char **)argv);
|
||||
} else {
|
||||
// First initialize OpenGL context, and then select CUDA device.
|
||||
initGL(&argc, argv);
|
||||
findCudaDevice(argc, (const char **)argv);
|
||||
}
|
||||
|
||||
initCudaBuffers();
|
||||
|
||||
if (ref_file) {
|
||||
printf("(Automated Testing)\n");
|
||||
bool testPassed = runSingleTest(ref_file, argv[0]);
|
||||
|
||||
cleanup();
|
||||
exit(testPassed ? EXIT_SUCCESS : EXIT_FAILURE);
|
||||
}
|
||||
|
||||
if (runBenchmark) {
|
||||
printf("(Run Benchmark)\n");
|
||||
benchmark(100);
|
||||
|
||||
cleanup();
|
||||
exit(EXIT_SUCCESS);
|
||||
}
|
||||
|
||||
initGLBuffers();
|
||||
glutMainLoop();
|
||||
|
||||
exit(EXIT_SUCCESS);
|
||||
}
|
||||
@@ -0,0 +1,160 @@
|
||||
/* 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.
|
||||
*/
|
||||
|
||||
/*
|
||||
Recursive Gaussian filter
|
||||
sgreen 8/1/08
|
||||
|
||||
This code sample implements a Gaussian blur using Deriche's recursive method:
|
||||
http://citeseer.ist.psu.edu/deriche93recursively.html
|
||||
|
||||
This is similar to the box filter sample in the SDK, but it uses the previous
|
||||
outputs of the filter as well as the previous inputs. This is also known as an
|
||||
IIR (infinite impulse response) filter, since its response to an input impulse
|
||||
can last forever.
|
||||
|
||||
The main advantage of this method is that the execution time is independent of
|
||||
the filter width.
|
||||
|
||||
The GPU processes columns of the image in parallel. To avoid uncoalesced reads
|
||||
for the row pass we transpose the image and then transpose it back again
|
||||
afterwards.
|
||||
|
||||
The implementation is based on code from the CImg library:
|
||||
http://cimg.sourceforge.net/
|
||||
Thanks to David Tschumperl<72> and all the CImg contributors!
|
||||
*/
|
||||
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
|
||||
#include <cuda_runtime.h>
|
||||
#include <helper_cuda.h>
|
||||
#include <helper_math.h>
|
||||
|
||||
#include "recursiveGaussian_kernel.cuh"
|
||||
|
||||
#define USE_SIMPLE_FILTER 0
|
||||
|
||||
// Round a / b to nearest higher integer value
|
||||
int iDivUp(int a, int b) { return (a % b != 0) ? (a / b + 1) : (a / b); }
|
||||
|
||||
/*
|
||||
Transpose a 2D array (see SDK transpose example)
|
||||
*/
|
||||
extern "C" void transpose(uint *d_src, uint *d_dest, uint width, int height) {
|
||||
dim3 grid(iDivUp(width, BLOCK_DIM), iDivUp(height, BLOCK_DIM), 1);
|
||||
dim3 threads(BLOCK_DIM, BLOCK_DIM, 1);
|
||||
d_transpose<<<grid, threads>>>(d_dest, d_src, width, height);
|
||||
getLastCudaError("Kernel execution failed");
|
||||
}
|
||||
|
||||
/*
|
||||
Perform Gaussian filter on a 2D image using CUDA
|
||||
|
||||
Parameters:
|
||||
d_src - pointer to input image in device memory
|
||||
d_dest - pointer to destination image in device memory
|
||||
d_temp - pointer to temporary storage in device memory
|
||||
width - image width
|
||||
height - image height
|
||||
sigma - sigma of Gaussian
|
||||
order - filter order (0, 1 or 2)
|
||||
*/
|
||||
|
||||
// 8-bit RGBA version
|
||||
extern "C" void gaussianFilterRGBA(uint *d_src, uint *d_dest, uint *d_temp,
|
||||
int width, int height, float sigma,
|
||||
int order, int nthreads) {
|
||||
// compute filter coefficients
|
||||
const float nsigma = sigma < 0.1f ? 0.1f : sigma, alpha = 1.695f / nsigma,
|
||||
ema = (float)std::exp(-alpha), ema2 = (float)std::exp(-2 * alpha),
|
||||
b1 = -2 * ema, b2 = ema2;
|
||||
|
||||
float a0 = 0, a1 = 0, a2 = 0, a3 = 0, coefp = 0, coefn = 0;
|
||||
|
||||
switch (order) {
|
||||
case 0: {
|
||||
const float k = (1 - ema) * (1 - ema) / (1 + 2 * alpha * ema - ema2);
|
||||
a0 = k;
|
||||
a1 = k * (alpha - 1) * ema;
|
||||
a2 = k * (alpha + 1) * ema;
|
||||
a3 = -k * ema2;
|
||||
} break;
|
||||
|
||||
case 1: {
|
||||
const float k = (1 - ema) * (1 - ema) / ema;
|
||||
a0 = k * ema;
|
||||
a1 = a3 = 0;
|
||||
a2 = -a0;
|
||||
} break;
|
||||
|
||||
case 2: {
|
||||
const float ea = (float)std::exp(-alpha),
|
||||
k = -(ema2 - 1) / (2 * alpha * ema),
|
||||
kn = (-2 * (-1 + 3 * ea - 3 * ea * ea + ea * ea * ea) /
|
||||
(3 * ea + 1 + 3 * ea * ea + ea * ea * ea));
|
||||
a0 = kn;
|
||||
a1 = -kn * (1 + k * alpha) * ema;
|
||||
a2 = kn * (1 - k * alpha) * ema;
|
||||
a3 = -kn * ema2;
|
||||
} break;
|
||||
|
||||
default:
|
||||
fprintf(stderr, "gaussianFilter: invalid order parameter!\n");
|
||||
return;
|
||||
}
|
||||
|
||||
coefp = (a0 + a1) / (1 + b1 + b2);
|
||||
coefn = (a2 + a3) / (1 + b1 + b2);
|
||||
|
||||
// process columns
|
||||
#if USE_SIMPLE_FILTER
|
||||
d_simpleRecursive_rgba<<<iDivUp(width, nthreads), nthreads>>>(
|
||||
d_src, d_temp, width, height, ema);
|
||||
#else
|
||||
d_recursiveGaussian_rgba<<<iDivUp(width, nthreads), nthreads>>>(
|
||||
d_src, d_temp, width, height, a0, a1, a2, a3, b1, b2, coefp, coefn);
|
||||
#endif
|
||||
getLastCudaError("Kernel execution failed");
|
||||
|
||||
transpose(d_temp, d_dest, width, height);
|
||||
getLastCudaError("transpose: Kernel execution failed");
|
||||
|
||||
// process rows
|
||||
#if USE_SIMPLE_FILTER
|
||||
d_simpleRecursive_rgba<<<iDivUp(height, nthreads), nthreads>>>(
|
||||
d_dest, d_temp, height, width, ema);
|
||||
#else
|
||||
d_recursiveGaussian_rgba<<<iDivUp(height, nthreads), nthreads>>>(
|
||||
d_dest, d_temp, height, width, a0, a1, a2, a3, b1, b2, coefp, coefn);
|
||||
#endif
|
||||
getLastCudaError("Kernel execution failed");
|
||||
|
||||
transpose(d_temp, d_dest, height, width);
|
||||
}
|
||||
@@ -0,0 +1,223 @@
|
||||
/* 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.
|
||||
*/
|
||||
|
||||
/*
|
||||
Recursive Gaussian filter
|
||||
*/
|
||||
|
||||
#ifndef _RECURSIVEGAUSSIAN_KERNEL_CU_
|
||||
#define _RECURSIVEGAUSSIAN_KERNEL_CU_
|
||||
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
#include <cooperative_groups.h>
|
||||
|
||||
namespace cg = cooperative_groups;
|
||||
|
||||
#include <helper_cuda.h>
|
||||
#include <helper_math.h>
|
||||
|
||||
#define BLOCK_DIM 16
|
||||
#define CLAMP_TO_EDGE 1
|
||||
|
||||
// Transpose kernel (see transpose CUDA Sample for details)
|
||||
__global__ void d_transpose(uint *odata, uint *idata, int width, int height) {
|
||||
// Handle to thread block group
|
||||
cg::thread_block cta = cg::this_thread_block();
|
||||
|
||||
__shared__ uint block[BLOCK_DIM][BLOCK_DIM + 1];
|
||||
|
||||
// read the matrix tile into shared memory
|
||||
unsigned int xIndex = blockIdx.x * BLOCK_DIM + threadIdx.x;
|
||||
unsigned int yIndex = blockIdx.y * BLOCK_DIM + threadIdx.y;
|
||||
|
||||
if ((xIndex < width) && (yIndex < height)) {
|
||||
unsigned int index_in = yIndex * width + xIndex;
|
||||
block[threadIdx.y][threadIdx.x] = idata[index_in];
|
||||
}
|
||||
|
||||
cg::sync(cta);
|
||||
|
||||
// write the transposed matrix tile to global memory
|
||||
xIndex = blockIdx.y * BLOCK_DIM + threadIdx.x;
|
||||
yIndex = blockIdx.x * BLOCK_DIM + threadIdx.y;
|
||||
|
||||
if ((xIndex < height) && (yIndex < width)) {
|
||||
unsigned int index_out = yIndex * height + xIndex;
|
||||
odata[index_out] = block[threadIdx.x][threadIdx.y];
|
||||
}
|
||||
}
|
||||
|
||||
// RGBA version
|
||||
// reads from 32-bit uint array holding 8-bit RGBA
|
||||
|
||||
// convert floating point rgba color to 32-bit integer
|
||||
__device__ uint rgbaFloatToInt(float4 rgba) {
|
||||
rgba.x = __saturatef(rgba.x); // clamp to [0.0, 1.0]
|
||||
rgba.y = __saturatef(rgba.y);
|
||||
rgba.z = __saturatef(rgba.z);
|
||||
rgba.w = __saturatef(rgba.w);
|
||||
return (uint(rgba.w * 255) << 24) | (uint(rgba.z * 255) << 16) |
|
||||
(uint(rgba.y * 255) << 8) | uint(rgba.x * 255);
|
||||
}
|
||||
|
||||
// convert from 32-bit int to float4
|
||||
__device__ float4 rgbaIntToFloat(uint c) {
|
||||
float4 rgba;
|
||||
rgba.x = (c & 0xff) / 255.0f;
|
||||
rgba.y = ((c >> 8) & 0xff) / 255.0f;
|
||||
rgba.z = ((c >> 16) & 0xff) / 255.0f;
|
||||
rgba.w = ((c >> 24) & 0xff) / 255.0f;
|
||||
return rgba;
|
||||
}
|
||||
|
||||
/*
|
||||
simple 1st order recursive filter
|
||||
- processes one image column per thread
|
||||
|
||||
parameters:
|
||||
id - pointer to input data (RGBA image packed into 32-bit integers)
|
||||
od - pointer to output data
|
||||
w - image width
|
||||
h - image height
|
||||
a - blur parameter
|
||||
*/
|
||||
|
||||
__global__ void d_simpleRecursive_rgba(uint *id, uint *od, int w, int h,
|
||||
float a) {
|
||||
unsigned int x = blockIdx.x * blockDim.x + threadIdx.x;
|
||||
|
||||
if (x >= w) return;
|
||||
|
||||
id += x; // advance pointers to correct column
|
||||
od += x;
|
||||
|
||||
// forward pass
|
||||
float4 yp = rgbaIntToFloat(*id); // previous output
|
||||
|
||||
for (int y = 0; y < h; y++) {
|
||||
float4 xc = rgbaIntToFloat(*id);
|
||||
float4 yc =
|
||||
xc + a * (yp - xc); // simple lerp between current and previous value
|
||||
*od = rgbaFloatToInt(yc);
|
||||
id += w;
|
||||
od += w; // move to next row
|
||||
yp = yc;
|
||||
}
|
||||
|
||||
// reset pointers to point to last element in column
|
||||
id -= w;
|
||||
od -= w;
|
||||
|
||||
// reverse pass
|
||||
// ensures response is symmetrical
|
||||
yp = rgbaIntToFloat(*id);
|
||||
|
||||
for (int y = h - 1; y >= 0; y--) {
|
||||
float4 xc = rgbaIntToFloat(*id);
|
||||
float4 yc = xc + a * (yp - xc);
|
||||
*od = rgbaFloatToInt((rgbaIntToFloat(*od) + yc) * 0.5f);
|
||||
id -= w;
|
||||
od -= w; // move to previous row
|
||||
yp = yc;
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
recursive Gaussian filter
|
||||
|
||||
parameters:
|
||||
id - pointer to input data (RGBA image packed into 32-bit integers)
|
||||
od - pointer to output data
|
||||
w - image width
|
||||
h - image height
|
||||
a0-a3, b1, b2, coefp, coefn - filter parameters
|
||||
*/
|
||||
|
||||
__global__ void d_recursiveGaussian_rgba(uint *id, uint *od, int w, int h,
|
||||
float a0, float a1, float a2, float a3,
|
||||
float b1, float b2, float coefp,
|
||||
float coefn) {
|
||||
unsigned int x = blockIdx.x * blockDim.x + threadIdx.x;
|
||||
|
||||
if (x >= w) return;
|
||||
|
||||
id += x; // advance pointers to correct column
|
||||
od += x;
|
||||
|
||||
// forward pass
|
||||
float4 xp = make_float4(0.0f); // previous input
|
||||
float4 yp = make_float4(0.0f); // previous output
|
||||
float4 yb = make_float4(0.0f); // previous output by 2
|
||||
#if CLAMP_TO_EDGE
|
||||
xp = rgbaIntToFloat(*id);
|
||||
yb = coefp * xp;
|
||||
yp = yb;
|
||||
#endif
|
||||
|
||||
for (int y = 0; y < h; y++) {
|
||||
float4 xc = rgbaIntToFloat(*id);
|
||||
float4 yc = a0 * xc + a1 * xp - b1 * yp - b2 * yb;
|
||||
*od = rgbaFloatToInt(yc);
|
||||
id += w;
|
||||
od += w; // move to next row
|
||||
xp = xc;
|
||||
yb = yp;
|
||||
yp = yc;
|
||||
}
|
||||
|
||||
// reset pointers to point to last element in column
|
||||
id -= w;
|
||||
od -= w;
|
||||
|
||||
// reverse pass
|
||||
// ensures response is symmetrical
|
||||
float4 xn = make_float4(0.0f);
|
||||
float4 xa = make_float4(0.0f);
|
||||
float4 yn = make_float4(0.0f);
|
||||
float4 ya = make_float4(0.0f);
|
||||
#if CLAMP_TO_EDGE
|
||||
xn = xa = rgbaIntToFloat(*id);
|
||||
yn = coefn * xn;
|
||||
ya = yn;
|
||||
#endif
|
||||
|
||||
for (int y = h - 1; y >= 0; y--) {
|
||||
float4 xc = rgbaIntToFloat(*id);
|
||||
float4 yc = a2 * xn + a3 * xa - b1 * yn - b2 * ya;
|
||||
xa = xn;
|
||||
xn = xc;
|
||||
ya = yn;
|
||||
yn = yc;
|
||||
*od = rgbaFloatToInt(rgbaIntToFloat(*od) + yc);
|
||||
id -= w;
|
||||
od -= w; // move to previous row
|
||||
}
|
||||
}
|
||||
|
||||
#endif // #ifndef _GAUSSIAN_KERNEL_H_
|
||||
@@ -0,0 +1,20 @@
|
||||
|
||||
Microsoft Visual Studio Solution File, Format Version 12.00
|
||||
# Visual Studio 2017
|
||||
Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "recursiveGaussian", "recursiveGaussian_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 +1,123 @@
|
||||
<?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)'==''">$(VCTargetsPath)\BuildCustomizations</CUDAPropsPath>
|
||||
</PropertyGroup>
|
||||
<ItemGroup Label="ProjectConfigurations">
|
||||
<ProjectConfiguration Include="Debug|x64">
|
||||
<Configuration>Debug</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
<ProjectConfiguration Include="Release|x64">
|
||||
<Configuration>Release</Configuration>
|
||||
<Platform>x64</Platform>
|
||||
</ProjectConfiguration>
|
||||
</ItemGroup>
|
||||
<PropertyGroup Label="Globals">
|
||||
<ProjectGuid>{997E0757-EA74-4A4E-A0FC-47D8C8831A15}</ProjectGuid>
|
||||
<RootNamespace>recursiveGaussian_vs2017</RootNamespace>
|
||||
<ProjectName>recursiveGaussian</ProjectName>
|
||||
<CudaToolkitCustomDir />
|
||||
</PropertyGroup>
|
||||
<PropertyGroup Condition="'$(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>./;$(CudaToolkitDir)/include;../../../Common;</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cudart_static.lib;freeglut.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);../../../Common/lib/$(PlatformName);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/recursiveGaussian.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>
|
||||
<ItemDefinitionGroup Condition="'$(Platform)'=='Win32'">
|
||||
<Link>
|
||||
|
||||
</Link>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Platform)'=='x64'">
|
||||
<Link>
|
||||
<AdditionalDependencies>glew64.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
</Link>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<ClCompile Include="recursiveGaussian.cpp" />
|
||||
<CudaCompile Include="recursiveGaussian_cuda.cu" />
|
||||
<None Include="recursiveGaussian_kernel.cuh" />
|
||||
</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}") = "recursiveGaussian", "recursiveGaussian_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 +1,119 @@
|
||||
<?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>recursiveGaussian_vs2019</RootNamespace>
|
||||
<ProjectName>recursiveGaussian</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>./;$(CudaToolkitDir)/include;../../../Common;</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cudart_static.lib;freeglut.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);../../../Common/lib/$(PlatformName);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/recursiveGaussian.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>
|
||||
<ItemDefinitionGroup Condition="'$(Platform)'=='Win32'">
|
||||
<Link>
|
||||
|
||||
</Link>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Platform)'=='x64'">
|
||||
<Link>
|
||||
<AdditionalDependencies>glew64.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
</Link>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<ClCompile Include="recursiveGaussian.cpp" />
|
||||
<CudaCompile Include="recursiveGaussian_cuda.cu" />
|
||||
<None Include="recursiveGaussian_kernel.cuh" />
|
||||
</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}") = "recursiveGaussian", "recursiveGaussian_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 +1,119 @@
|
||||
<?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>recursiveGaussian_vs2022</RootNamespace>
|
||||
<ProjectName>recursiveGaussian</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>./;$(CudaToolkitDir)/include;../../../Common;</AdditionalIncludeDirectories>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<SubSystem>Console</SubSystem>
|
||||
<AdditionalDependencies>cudart_static.lib;freeglut.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
<AdditionalLibraryDirectories>$(CudaToolkitLibDir);../../../Common/lib/$(PlatformName);</AdditionalLibraryDirectories>
|
||||
<OutputFile>$(OutDir)/recursiveGaussian.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>
|
||||
<ItemDefinitionGroup Condition="'$(Platform)'=='Win32'">
|
||||
<Link>
|
||||
|
||||
</Link>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Platform)'=='x64'">
|
||||
<Link>
|
||||
<AdditionalDependencies>glew64.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
</Link>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<ClCompile Include="recursiveGaussian.cpp" />
|
||||
<CudaCompile Include="recursiveGaussian_cuda.cu" />
|
||||
<None Include="recursiveGaussian_kernel.cuh" />
|
||||
</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