mirror of
https://github.com/NVIDIA/cuda-samples.git
synced 2026-10-23 21:28:17 +08:00
add and update samples for CUDA 11.5
This commit is contained in:
Rutwik Choughule
18
Samples/bicubicTexture/.vscode/c_cpp_properties.json
vendored
Normal file
18
Samples/bicubicTexture/.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/bicubicTexture/.vscode/extensions.json
vendored
Normal file
7
Samples/bicubicTexture/.vscode/extensions.json
vendored
Normal file
@@ -0,0 +1,7 @@
|
||||
{
|
||||
"recommendations": [
|
||||
"nvidia.nsight-vscode-edition",
|
||||
"ms-vscode.cpptools",
|
||||
"ms-vscode.makefile-tools"
|
||||
]
|
||||
}
|
||||
10
Samples/bicubicTexture/.vscode/launch.json
vendored
Normal file
10
Samples/bicubicTexture/.vscode/launch.json
vendored
Normal file
@@ -0,0 +1,10 @@
|
||||
{
|
||||
"configurations": [
|
||||
{
|
||||
"name": "CUDA C++: Launch",
|
||||
"type": "cuda-gdb",
|
||||
"request": "launch",
|
||||
"program": "${workspaceFolder}/bicubicTexture"
|
||||
}
|
||||
]
|
||||
}
|
||||
15
Samples/bicubicTexture/.vscode/tasks.json
vendored
Normal file
15
Samples/bicubicTexture/.vscode/tasks.json
vendored
Normal file
@@ -0,0 +1,15 @@
|
||||
{
|
||||
"version": "2.0.0",
|
||||
"tasks": [
|
||||
{
|
||||
"label": "sample",
|
||||
"type": "shell",
|
||||
"command": "make dbg=1",
|
||||
"problemMatcher": ["$nvcc"],
|
||||
"group": {
|
||||
"kind": "build",
|
||||
"isDefault": true
|
||||
}
|
||||
}
|
||||
]
|
||||
}
|
||||
385
Samples/bicubicTexture/Makefile
Normal file
385
Samples/bicubicTexture/Makefile
Normal file
@@ -0,0 +1,385 @@
|
||||
################################################################################
|
||||
# Copyright (c) 2019, NVIDIA CORPORATION. All rights reserved.
|
||||
#
|
||||
# Redistribution and use in source and binary forms, with or without
|
||||
# modification, are permitted provided that the following conditions
|
||||
# are met:
|
||||
# * Redistributions of source code must retain the above copyright
|
||||
# notice, this list of conditions and the following disclaimer.
|
||||
# * Redistributions in binary form must reproduce the above copyright
|
||||
# notice, this list of conditions and the following disclaimer in the
|
||||
# documentation and/or other materials provided with the distribution.
|
||||
# * Neither the name of NVIDIA CORPORATION nor the names of its
|
||||
# contributors may be used to endorse or promote products derived
|
||||
# from this software without specific prior written permission.
|
||||
#
|
||||
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
|
||||
# EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
||||
# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
|
||||
# PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
|
||||
# CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
|
||||
# EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
|
||||
# PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
|
||||
# PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
|
||||
# OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
||||
# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
||||
# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
#
|
||||
################################################################################
|
||||
#
|
||||
# Makefile project only supported on Mac OS X and Linux Platforms)
|
||||
#
|
||||
################################################################################
|
||||
|
||||
# Location of the CUDA Toolkit
|
||||
CUDA_PATH ?= /usr/local/cuda
|
||||
|
||||
##############################
|
||||
# start deprecated interface #
|
||||
##############################
|
||||
ifeq ($(x86_64),1)
|
||||
$(info WARNING - x86_64 variable has been deprecated)
|
||||
$(info WARNING - please use TARGET_ARCH=x86_64 instead)
|
||||
TARGET_ARCH ?= x86_64
|
||||
endif
|
||||
ifeq ($(ARMv7),1)
|
||||
$(info WARNING - ARMv7 variable has been deprecated)
|
||||
$(info WARNING - please use TARGET_ARCH=armv7l instead)
|
||||
TARGET_ARCH ?= armv7l
|
||||
endif
|
||||
ifeq ($(aarch64),1)
|
||||
$(info WARNING - aarch64 variable has been deprecated)
|
||||
$(info WARNING - please use TARGET_ARCH=aarch64 instead)
|
||||
TARGET_ARCH ?= aarch64
|
||||
endif
|
||||
ifeq ($(ppc64le),1)
|
||||
$(info WARNING - ppc64le variable has been deprecated)
|
||||
$(info WARNING - please use TARGET_ARCH=ppc64le instead)
|
||||
TARGET_ARCH ?= ppc64le
|
||||
endif
|
||||
ifneq ($(GCC),)
|
||||
$(info WARNING - GCC variable has been deprecated)
|
||||
$(info WARNING - please use HOST_COMPILER=$(GCC) instead)
|
||||
HOST_COMPILER ?= $(GCC)
|
||||
endif
|
||||
ifneq ($(abi),)
|
||||
$(error ERROR - abi variable has been removed)
|
||||
endif
|
||||
############################
|
||||
# end deprecated interface #
|
||||
############################
|
||||
|
||||
# architecture
|
||||
HOST_ARCH := $(shell uname -m)
|
||||
TARGET_ARCH ?= $(HOST_ARCH)
|
||||
ifneq (,$(filter $(TARGET_ARCH),x86_64 aarch64 sbsa ppc64le armv7l))
|
||||
ifneq ($(TARGET_ARCH),$(HOST_ARCH))
|
||||
ifneq (,$(filter $(TARGET_ARCH),x86_64 aarch64 sbsa ppc64le))
|
||||
TARGET_SIZE := 64
|
||||
else ifneq (,$(filter $(TARGET_ARCH),armv7l))
|
||||
TARGET_SIZE := 32
|
||||
endif
|
||||
else
|
||||
TARGET_SIZE := $(shell getconf LONG_BIT)
|
||||
endif
|
||||
else
|
||||
$(error ERROR - unsupported value $(TARGET_ARCH) for TARGET_ARCH!)
|
||||
endif
|
||||
|
||||
# sbsa and aarch64 systems look similar. Need to differentiate them at host level for now.
|
||||
ifeq ($(HOST_ARCH),aarch64)
|
||||
ifeq ($(CUDA_PATH)/targets/sbsa-linux,$(shell ls -1d $(CUDA_PATH)/targets/sbsa-linux 2>/dev/null))
|
||||
HOST_ARCH := sbsa
|
||||
TARGET_ARCH := sbsa
|
||||
endif
|
||||
endif
|
||||
|
||||
ifneq ($(TARGET_ARCH),$(HOST_ARCH))
|
||||
ifeq (,$(filter $(HOST_ARCH)-$(TARGET_ARCH),aarch64-armv7l x86_64-armv7l x86_64-aarch64 x86_64-sbsa x86_64-ppc64le))
|
||||
$(error ERROR - cross compiling from $(HOST_ARCH) to $(TARGET_ARCH) is not supported!)
|
||||
endif
|
||||
endif
|
||||
|
||||
# When on native aarch64 system with userspace of 32-bit, change TARGET_ARCH to armv7l
|
||||
ifeq ($(HOST_ARCH)-$(TARGET_ARCH)-$(TARGET_SIZE),aarch64-aarch64-32)
|
||||
TARGET_ARCH = armv7l
|
||||
endif
|
||||
|
||||
# operating system
|
||||
HOST_OS := $(shell uname -s 2>/dev/null | tr "[:upper:]" "[:lower:]")
|
||||
TARGET_OS ?= $(HOST_OS)
|
||||
ifeq (,$(filter $(TARGET_OS),linux darwin qnx android))
|
||||
$(error ERROR - unsupported value $(TARGET_OS) for TARGET_OS!)
|
||||
endif
|
||||
|
||||
# host compiler
|
||||
ifeq ($(TARGET_OS),darwin)
|
||||
ifeq ($(shell expr `xcodebuild -version | grep -i xcode | awk '{print $$2}' | cut -d'.' -f1` \>= 5),1)
|
||||
HOST_COMPILER ?= clang++
|
||||
endif
|
||||
else ifneq ($(TARGET_ARCH),$(HOST_ARCH))
|
||||
ifeq ($(HOST_ARCH)-$(TARGET_ARCH),x86_64-armv7l)
|
||||
ifeq ($(TARGET_OS),linux)
|
||||
HOST_COMPILER ?= arm-linux-gnueabihf-g++
|
||||
else ifeq ($(TARGET_OS),qnx)
|
||||
ifeq ($(QNX_HOST),)
|
||||
$(error ERROR - QNX_HOST must be passed to the QNX host toolchain)
|
||||
endif
|
||||
ifeq ($(QNX_TARGET),)
|
||||
$(error ERROR - QNX_TARGET must be passed to the QNX target toolchain)
|
||||
endif
|
||||
export QNX_HOST
|
||||
export QNX_TARGET
|
||||
HOST_COMPILER ?= $(QNX_HOST)/usr/bin/arm-unknown-nto-qnx6.6.0eabi-g++
|
||||
else ifeq ($(TARGET_OS),android)
|
||||
HOST_COMPILER ?= arm-linux-androideabi-g++
|
||||
endif
|
||||
else ifeq ($(TARGET_ARCH),aarch64)
|
||||
ifeq ($(TARGET_OS), linux)
|
||||
HOST_COMPILER ?= aarch64-linux-gnu-g++
|
||||
else ifeq ($(TARGET_OS),qnx)
|
||||
ifeq ($(QNX_HOST),)
|
||||
$(error ERROR - QNX_HOST must be passed to the QNX host toolchain)
|
||||
endif
|
||||
ifeq ($(QNX_TARGET),)
|
||||
$(error ERROR - QNX_TARGET must be passed to the QNX target toolchain)
|
||||
endif
|
||||
export QNX_HOST
|
||||
export QNX_TARGET
|
||||
HOST_COMPILER ?= $(QNX_HOST)/usr/bin/q++
|
||||
else ifeq ($(TARGET_OS), android)
|
||||
HOST_COMPILER ?= aarch64-linux-android-clang++
|
||||
endif
|
||||
else ifeq ($(TARGET_ARCH),sbsa)
|
||||
HOST_COMPILER ?= aarch64-linux-gnu-g++
|
||||
else ifeq ($(TARGET_ARCH),ppc64le)
|
||||
HOST_COMPILER ?= powerpc64le-linux-gnu-g++
|
||||
endif
|
||||
endif
|
||||
HOST_COMPILER ?= g++
|
||||
NVCC := $(CUDA_PATH)/bin/nvcc -ccbin $(HOST_COMPILER)
|
||||
|
||||
# internal flags
|
||||
NVCCFLAGS := -m${TARGET_SIZE}
|
||||
CCFLAGS :=
|
||||
LDFLAGS :=
|
||||
|
||||
# build flags
|
||||
ifeq ($(TARGET_OS),darwin)
|
||||
LDFLAGS += -rpath $(CUDA_PATH)/lib
|
||||
CCFLAGS += -arch $(HOST_ARCH)
|
||||
else ifeq ($(HOST_ARCH)-$(TARGET_ARCH)-$(TARGET_OS),x86_64-armv7l-linux)
|
||||
LDFLAGS += --dynamic-linker=/lib/ld-linux-armhf.so.3
|
||||
CCFLAGS += -mfloat-abi=hard
|
||||
else ifeq ($(TARGET_OS),android)
|
||||
LDFLAGS += -pie
|
||||
CCFLAGS += -fpie -fpic -fexceptions
|
||||
endif
|
||||
|
||||
ifneq ($(TARGET_ARCH),$(HOST_ARCH))
|
||||
ifeq ($(TARGET_ARCH)-$(TARGET_OS),armv7l-linux)
|
||||
ifneq ($(TARGET_FS),)
|
||||
GCCVERSIONLTEQ46 := $(shell expr `$(HOST_COMPILER) -dumpversion` \<= 4.6)
|
||||
ifeq ($(GCCVERSIONLTEQ46),1)
|
||||
CCFLAGS += --sysroot=$(TARGET_FS)
|
||||
endif
|
||||
LDFLAGS += --sysroot=$(TARGET_FS)
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/lib
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/usr/lib
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/usr/lib/arm-linux-gnueabihf
|
||||
endif
|
||||
endif
|
||||
ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-linux)
|
||||
ifneq ($(TARGET_FS),)
|
||||
GCCVERSIONLTEQ46 := $(shell expr `$(HOST_COMPILER) -dumpversion` \<= 4.6)
|
||||
ifeq ($(GCCVERSIONLTEQ46),1)
|
||||
CCFLAGS += --sysroot=$(TARGET_FS)
|
||||
endif
|
||||
LDFLAGS += --sysroot=$(TARGET_FS)
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/lib -L$(TARGET_FS)/lib
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/lib/aarch64-linux-gnu -L$(TARGET_FS)/lib/aarch64-linux-gnu
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/usr/lib -L$(TARGET_FS)/usr/lib
|
||||
LDFLAGS += -rpath-link=$(TARGET_FS)/usr/lib/aarch64-linux-gnu -L$(TARGET_FS)/usr/lib/aarch64-linux-gnu
|
||||
LDFLAGS += --unresolved-symbols=ignore-in-shared-libs
|
||||
CCFLAGS += -isystem=$(TARGET_FS)/usr/include -I$(TARGET_FS)/usr/include -I$(TARGET_FS)/usr/include/libdrm
|
||||
CCFLAGS += -isystem=$(TARGET_FS)/usr/include/aarch64-linux-gnu -I$(TARGET_FS)/usr/include/aarch64-linux-gnu
|
||||
endif
|
||||
endif
|
||||
ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-qnx)
|
||||
NVCCFLAGS += --qpp-config 5.4.0,gcc_ntoaarch64le
|
||||
CCFLAGS += -DWIN_INTERFACE_CUSTOM -I/usr/include/aarch64-qnx-gnu
|
||||
LDFLAGS += -lsocket
|
||||
LDFLAGS += -L/usr/lib/aarch64-qnx-gnu
|
||||
CCFLAGS += "-Wl\,-rpath-link\,/usr/lib/aarch64-qnx-gnu"
|
||||
ifdef TARGET_OVERRIDE
|
||||
LDFLAGS += -lslog2
|
||||
endif
|
||||
|
||||
ifneq ($(TARGET_FS),)
|
||||
LDFLAGS += -L$(TARGET_FS)/usr/lib
|
||||
CCFLAGS += "-Wl\,-rpath-link\,$(TARGET_FS)/usr/lib"
|
||||
LDFLAGS += -L$(TARGET_FS)/usr/libnvidia
|
||||
CCFLAGS += "-Wl\,-rpath-link\,$(TARGET_FS)/usr/libnvidia"
|
||||
CCFLAGS += -I$(TARGET_FS)/../include
|
||||
endif
|
||||
endif
|
||||
endif
|
||||
|
||||
ifdef TARGET_OVERRIDE # cuda toolkit targets override
|
||||
NVCCFLAGS += -target-dir $(TARGET_OVERRIDE)
|
||||
endif
|
||||
|
||||
# Install directory of different arch
|
||||
CUDA_INSTALL_TARGET_DIR :=
|
||||
ifeq ($(TARGET_ARCH)-$(TARGET_OS),armv7l-linux)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/armv7-linux-gnueabihf/
|
||||
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-linux)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/aarch64-linux/
|
||||
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),sbsa-linux)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/sbsa-linux/
|
||||
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),armv7l-android)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/armv7-linux-androideabi/
|
||||
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-android)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/aarch64-linux-androideabi/
|
||||
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),armv7l-qnx)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/ARMv7-linux-QNX/
|
||||
else ifeq ($(TARGET_ARCH)-$(TARGET_OS),aarch64-qnx)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/aarch64-qnx/
|
||||
else ifeq ($(TARGET_ARCH),ppc64le)
|
||||
CUDA_INSTALL_TARGET_DIR = targets/ppc64le-linux/
|
||||
endif
|
||||
|
||||
# Debug build flags
|
||||
ifeq ($(dbg),1)
|
||||
NVCCFLAGS += -g -G
|
||||
BUILD_TYPE := debug
|
||||
else
|
||||
BUILD_TYPE := release
|
||||
endif
|
||||
|
||||
ALL_CCFLAGS :=
|
||||
ALL_CCFLAGS += $(NVCCFLAGS)
|
||||
ALL_CCFLAGS += $(EXTRA_NVCCFLAGS)
|
||||
ALL_CCFLAGS += $(addprefix -Xcompiler ,$(CCFLAGS))
|
||||
ALL_CCFLAGS += $(addprefix -Xcompiler ,$(EXTRA_CCFLAGS))
|
||||
|
||||
SAMPLE_ENABLED := 1
|
||||
|
||||
# This sample is not supported on QNX
|
||||
ifeq ($(TARGET_OS),qnx)
|
||||
$(info >>> WARNING - bicubicTexture 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
|
||||
|
||||
#Detect if installed version of GCC supports required C++11
|
||||
ifeq ($(TARGET_OS),linux)
|
||||
empty :=
|
||||
space := $(empty) $(empty)
|
||||
GCCVERSIONSTRING := $(shell expr `$(HOST_COMPILER) -dumpversion`)
|
||||
#Create version number without "."
|
||||
GCCVERSION := $(shell expr `echo $(GCCVERSIONSTRING)` | cut -f1 -d.)
|
||||
GCCVERSION += $(shell expr `echo $(GCCVERSIONSTRING)` | cut -f2 -d.)
|
||||
GCCVERSION += $(shell expr `echo $(GCCVERSIONSTRING)` | cut -f3 -d.)
|
||||
# Make sure the version number has at least 3 decimals
|
||||
GCCVERSION += 00
|
||||
# Remove spaces from the version number
|
||||
GCCVERSION := $(subst $(space),$(empty),$(GCCVERSION))
|
||||
#$(warning $(GCCVERSION))
|
||||
|
||||
IS_MIN_VERSION := $(shell expr `echo $(GCCVERSION)` \>= 51000)
|
||||
|
||||
ifeq ($(IS_MIN_VERSION), 1)
|
||||
$(info >>> GCC Version is greater or equal to 5.1.0 <<<)
|
||||
else
|
||||
$(info >>> Waiving build. Minimum GCC version required is 5.1.0<<<)
|
||||
SAMPLE_ENABLED := 0
|
||||
endif
|
||||
endif
|
||||
|
||||
# Gencode arguments
|
||||
ifeq ($(TARGET_ARCH),$(filter $(TARGET_ARCH),armv7l aarch64))
|
||||
SMS ?= 35 37 50 52 60 61 70 72 75 80 86
|
||||
else
|
||||
SMS ?= 35 37 50 52 60 61 70 75 80 86
|
||||
endif
|
||||
|
||||
ifeq ($(SMS),)
|
||||
$(info >>> WARNING - no SM architectures have been specified - waiving sample <<<)
|
||||
SAMPLE_ENABLED := 0
|
||||
endif
|
||||
|
||||
ifeq ($(GENCODE_FLAGS),)
|
||||
# Generate SASS code for each SM architecture listed in $(SMS)
|
||||
$(foreach sm,$(SMS),$(eval GENCODE_FLAGS += -gencode arch=compute_$(sm),code=sm_$(sm)))
|
||||
|
||||
# Generate PTX code from the highest SM architecture in $(SMS) to guarantee forward-compatibility
|
||||
HIGHEST_SM := $(lastword $(sort $(SMS)))
|
||||
ifneq ($(HIGHEST_SM),)
|
||||
GENCODE_FLAGS += -gencode arch=compute_$(HIGHEST_SM),code=compute_$(HIGHEST_SM)
|
||||
endif
|
||||
endif
|
||||
|
||||
ALL_CCFLAGS += --threads 0 --std=c++11
|
||||
|
||||
ifeq ($(SAMPLE_ENABLED),0)
|
||||
EXEC ?= @echo "[@]"
|
||||
endif
|
||||
|
||||
################################################################################
|
||||
|
||||
# Target rules
|
||||
all: build
|
||||
|
||||
build: bicubicTexture
|
||||
|
||||
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
|
||||
|
||||
bicubicTexture.o:bicubicTexture.cpp
|
||||
$(EXEC) $(NVCC) $(INCLUDES) $(ALL_CCFLAGS) $(GENCODE_FLAGS) -o $@ -c $<
|
||||
|
||||
bicubicTexture_cuda.o:bicubicTexture_cuda.cu
|
||||
$(EXEC) $(NVCC) $(INCLUDES) $(ALL_CCFLAGS) $(GENCODE_FLAGS) -o $@ -c $<
|
||||
|
||||
bicubicTexture: bicubicTexture.o bicubicTexture_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) ./bicubicTexture
|
||||
|
||||
clean:
|
||||
rm -f bicubicTexture bicubicTexture.o bicubicTexture_cuda.o
|
||||
rm -rf ../../bin/$(TARGET_ARCH)/$(TARGET_OS)/$(BUILD_TYPE)/bicubicTexture
|
||||
|
||||
clobber: clean
|
||||
99
Samples/bicubicTexture/NsightEclipse.xml
Normal file
99
Samples/bicubicTexture/NsightEclipse.xml
Normal file
@@ -0,0 +1,99 @@
|
||||
<?xml version="1.0" encoding="UTF-8"?>
|
||||
<!DOCTYPE entry SYSTEM "SamplesInfo.dtd">
|
||||
<entry>
|
||||
<name>bicubicTexture</name>
|
||||
<cuda_api_list>
|
||||
<toolkit>cudaGraphicsMapResources</toolkit>
|
||||
<toolkit>cudaGraphicsUnmapResources</toolkit>
|
||||
<toolkit>cudaGraphicsResourceGetMappedPointer</toolkit>
|
||||
<toolkit>cudaGraphicsRegisterResource</toolkit>
|
||||
<toolkit>cudaGraphicsGLRegisterBuffer</toolkit>
|
||||
<toolkit>cudaGraphicsUnregisterResource</toolkit>
|
||||
</cuda_api_list>
|
||||
<description><![CDATA[This sample demonstrates how to efficiently implement a Bicubic B-spline interpolation filter with CUDA texture.]]></description>
|
||||
<devicecompilation>whole</devicecompilation>
|
||||
<files>
|
||||
<file>data\lena_bw.raw</file>
|
||||
<file>data\0_nearest.ppm</file>
|
||||
<file>data\1_bilinear.ppm</file>
|
||||
<file>data\2_bicubic.ppm</file>
|
||||
<file>data\3_fastbicubic.ppm</file>
|
||||
<file>data\4_catmull-rom.ppm</file>
|
||||
</files>
|
||||
<includepaths>
|
||||
<path>./</path>
|
||||
<path>../</path>
|
||||
<path>../../Common</path>
|
||||
</includepaths>
|
||||
<keyconcepts>
|
||||
<concept level="basic">Graphics Interop</concept>
|
||||
<concept level="basic">Image Processing</concept>
|
||||
</keyconcepts>
|
||||
<keywords>
|
||||
<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>bicubicTexture.cpp</primary_file>
|
||||
<qatests>
|
||||
<qatest>-mode=0 -file=0_nearest.ppm</qatest>
|
||||
<qatest>-mode=1 -file=1_bilinear.ppm</qatest>
|
||||
<qatest>-mode=2 -file=2_bicubic.ppm</qatest>
|
||||
<qatest>-mode=3 -file=3_fastbicubic.ppm</qatest>
|
||||
<qatest>-mode=4 -file=4_catmull-rom.ppm</qatest>
|
||||
</qatests>
|
||||
<required_dependencies>
|
||||
<dependency>X11</dependency>
|
||||
<dependency>GL</dependency>
|
||||
</required_dependencies>
|
||||
<scopes>
|
||||
<scope>1:CUDA Basic Topics</scope>
|
||||
<scope>2:Texture</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>sm60</sm-arch>
|
||||
<sm-arch>sm61</sm-arch>
|
||||
<sm-arch>sm70</sm-arch>
|
||||
<sm-arch>sm72</sm-arch>
|
||||
<sm-arch>sm75</sm-arch>
|
||||
<sm-arch>sm80</sm-arch>
|
||||
<sm-arch>sm86</sm-arch>
|
||||
<supported_envs>
|
||||
<env>
|
||||
<arch>x86_64</arch>
|
||||
<platform>linux</platform>
|
||||
</env>
|
||||
<env>
|
||||
<platform>windows7</platform>
|
||||
</env>
|
||||
<env>
|
||||
<arch>x86_64</arch>
|
||||
<platform>macosx</platform>
|
||||
</env>
|
||||
<env>
|
||||
<arch>arm</arch>
|
||||
</env>
|
||||
</supported_envs>
|
||||
<supported_sm_architectures>
|
||||
<include>all</include>
|
||||
</supported_sm_architectures>
|
||||
<title>Bicubic B-spline Interoplation</title>
|
||||
<type>exe</type>
|
||||
</entry>
|
||||
74
Samples/bicubicTexture/README.md
Normal file
74
Samples/bicubicTexture/README.md
Normal file
@@ -0,0 +1,74 @@
|
||||
# bicubicTexture - Bicubic B-spline Interoplation
|
||||
|
||||
## Description
|
||||
|
||||
This sample demonstrates how to efficiently implement a Bicubic B-spline interpolation filter with CUDA texture.
|
||||
|
||||
## 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 6.0 ](https://developer.nvidia.com/cuda-gpus) [SM 6.1 ](https://developer.nvidia.com/cuda-gpus) [SM 7.0 ](https://developer.nvidia.com/cuda-gpus) [SM 7.2 ](https://developer.nvidia.com/cuda-gpus) [SM 7.5 ](https://developer.nvidia.com/cuda-gpus) [SM 8.0 ](https://developer.nvidia.com/cuda-gpus) [SM 8.6 ](https://developer.nvidia.com/cuda-gpus)
|
||||
|
||||
## Supported OSes
|
||||
|
||||
Linux, Windows
|
||||
|
||||
## Supported CPU Architecture
|
||||
|
||||
x86_64, armv7l
|
||||
|
||||
## CUDA APIs involved
|
||||
|
||||
### [CUDA Runtime API](http://docs.nvidia.com/cuda/cuda-runtime-api/index.html)
|
||||
cudaGraphicsMapResources, cudaGraphicsUnmapResources, cudaGraphicsResourceGetMappedPointer, cudaGraphicsRegisterResource, cudaGraphicsGLRegisterBuffer, cudaGraphicsUnregisterResource
|
||||
|
||||
## Dependencies needed to build/run
|
||||
[X11](../../README.md#x11), [GL](../../README.md#gl)
|
||||
|
||||
## Prerequisites
|
||||
|
||||
Download and install the [CUDA Toolkit 11.5](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)
|
||||
|
||||
781
Samples/bicubicTexture/bicubicTexture.cpp
Normal file
781
Samples/bicubicTexture/bicubicTexture.cpp
Normal file
@@ -0,0 +1,781 @@
|
||||
/* Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved.
|
||||
*
|
||||
* Redistribution and use in source and binary forms, with or without
|
||||
* modification, are permitted provided that the following conditions
|
||||
* are met:
|
||||
* * Redistributions of source code must retain the above copyright
|
||||
* notice, this list of conditions and the following disclaimer.
|
||||
* * Redistributions in binary form must reproduce the above copyright
|
||||
* notice, this list of conditions and the following disclaimer in the
|
||||
* documentation and/or other materials provided with the distribution.
|
||||
* * Neither the name of NVIDIA CORPORATION nor the names of its
|
||||
* contributors may be used to endorse or promote products derived
|
||||
* from this software without specific prior written permission.
|
||||
*
|
||||
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
|
||||
* EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
||||
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
|
||||
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
|
||||
* CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
|
||||
* EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
|
||||
* PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
|
||||
* PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
|
||||
* OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
||||
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
||||
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
*/
|
||||
|
||||
/*
|
||||
Bicubic texture filtering sample
|
||||
sgreen 6/2008
|
||||
|
||||
This sample demonstrates how to efficiently implement bicubic texture
|
||||
filtering in CUDA.
|
||||
|
||||
Bicubic filtering is a higher order interpolation method that produces
|
||||
smoother results than bilinear interpolation:
|
||||
http://en.wikipedia.org/wiki/Bicubic
|
||||
|
||||
It requires reading a 4 x 4 pixel neighbourhood rather than the
|
||||
2 x 2 area required by bilinear filtering.
|
||||
|
||||
Current graphics hardware doesn't support bicubic filtering natively,
|
||||
but it is possible to compose a bicubic filter using just 4 bilinear
|
||||
lookups by offsetting the sample position within each texel and weighting
|
||||
the samples correctly. The only disadvantage to this method is that the
|
||||
hardware only maintains 9-bits of filtering precision within each texel.
|
||||
|
||||
See "Fast Third-Order Texture Filtering", Sigg & Hadwiger, GPU Gems 2:
|
||||
https://developer.nvidia.com/gpugems/gpugems2/part-iii-high-quality-rendering/chapter-20-fast-third-order-texture-filtering
|
||||
|
||||
v1.1 - updated to include the brute force method using 16 texture lookups.
|
||||
v1.2 - added Catmull-Rom interpolation
|
||||
|
||||
Example performance results from GeForce 8800 GTS:
|
||||
Bilinear - 5500 MPixels/sec
|
||||
Bicubic - 1400 MPixels/sec
|
||||
Fast Bicubic - 2100 MPixels/sec
|
||||
*/
|
||||
|
||||
// OpenGL Graphics includes
|
||||
#include <helper_gl.h>
|
||||
#if defined(__APPLE__) || defined(MACOSX)
|
||||
#pragma clang diagnostic ignored "-Wdeprecated-declarations"
|
||||
#include <GLUT/glut.h>
|
||||
#else
|
||||
#include <GL/freeglut.h>
|
||||
#endif
|
||||
|
||||
// Includes
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
#include <math.h>
|
||||
|
||||
// CUDA system and GL includes
|
||||
#include <cuda_runtime.h>
|
||||
#include <cuda_gl_interop.h>
|
||||
|
||||
// Helper functions
|
||||
#include <helper_functions.h> // CUDA SDK Helper functions
|
||||
#include <helper_cuda.h> // CUDA device initialization helper functions
|
||||
|
||||
typedef unsigned int uint;
|
||||
typedef unsigned char uchar;
|
||||
|
||||
#define USE_BUFFER_TEX 0
|
||||
#ifndef MAX
|
||||
#define MAX(a, b) ((a < b) ? b : a)
|
||||
#endif
|
||||
|
||||
// Auto-Verification Code
|
||||
const int frameCheckNumber = 4;
|
||||
int fpsCount = 0; // FPS count for averaging
|
||||
int fpsLimit = 4; // FPS limit for sampling
|
||||
int g_Index = 0;
|
||||
unsigned int frameCount = 0;
|
||||
unsigned int g_TotalErrors = 0;
|
||||
StopWatchInterface *timer = 0;
|
||||
bool g_Verify = false;
|
||||
|
||||
int *pArgc = NULL;
|
||||
char **pArgv = NULL;
|
||||
|
||||
#define MAX_EPSILON_ERROR 5.0f
|
||||
#define REFRESH_DELAY 10 // ms
|
||||
|
||||
static const char *sSDKsample = "CUDA BicubicTexture";
|
||||
|
||||
// Define the files that are to be save and the reference images for validation
|
||||
const char *sFilterMode[] = {"Nearest", "Bilinear", "Bicubic",
|
||||
"Fast Bicubic", "Catmull-Rom", NULL};
|
||||
|
||||
const char *sOriginal[] = {"0_nearest.ppm", "1_bilinear.ppm",
|
||||
"2_bicubic.ppm", "3_fastbicubic.ppm",
|
||||
"4_catmull-rom.ppm", NULL};
|
||||
|
||||
const char *sReference[] = {"0_nearest.ppm", "1_bilinear.ppm",
|
||||
"2_bicubic.ppm", "3_fastbicubic.ppm",
|
||||
"4_catmull-rom.ppm", NULL};
|
||||
|
||||
const char *srcImageFilename = "lena_bw.pgm";
|
||||
char *dumpFilename = NULL;
|
||||
|
||||
uint width = 512, height = 512;
|
||||
uint imageWidth, imageHeight;
|
||||
dim3 blockSize(16, 16);
|
||||
dim3 gridSize(width / blockSize.x, height / blockSize.y);
|
||||
|
||||
enum eFilterMode {
|
||||
MODE_NEAREST,
|
||||
MODE_BILINEAR,
|
||||
MODE_BICUBIC,
|
||||
MODE_FAST_BICUBIC,
|
||||
MODE_CATMULL_ROM,
|
||||
NUM_MODES
|
||||
};
|
||||
|
||||
eFilterMode g_FilterMode = MODE_FAST_BICUBIC;
|
||||
|
||||
bool drawCurves = false;
|
||||
|
||||
GLuint pbo = 0; // OpenGL pixel buffer object
|
||||
struct cudaGraphicsResource *cuda_pbo_resource; // handles OpenGL-CUDA exchange
|
||||
GLuint displayTex = 0;
|
||||
GLuint bufferTex = 0;
|
||||
GLuint fprog; // fragment program (shader)
|
||||
|
||||
float tx = 9.0f, ty = 10.0f; // image translation
|
||||
float scale = 1.0f / 16.0f; // image scale
|
||||
float cx, cy; // image centre
|
||||
|
||||
void display();
|
||||
void initGLBuffers();
|
||||
void runBenchmark(int iterations);
|
||||
void cleanup();
|
||||
|
||||
#define GL_TEXTURE_TYPE GL_TEXTURE_RECTANGLE_ARB
|
||||
//#define GL_TEXTURE_TYPE GL_TEXTURE_2D
|
||||
|
||||
extern "C" void initGL(int *argc, char **argv);
|
||||
extern "C" void loadImageData(int argc, char **argv);
|
||||
|
||||
extern "C" void initTexture(int imageWidth, int imageHeight, uchar *h_data);
|
||||
extern "C" void freeTexture();
|
||||
extern "C" void render(int width, int height, float tx, float ty, float scale,
|
||||
float cx, float cy, dim3 blockSize, dim3 gridSize,
|
||||
eFilterMode filter_mode, uchar4 *output);
|
||||
|
||||
// w0, w1, w2, and w3 are the four cubic B-spline basis functions
|
||||
float bspline_w0(float a) {
|
||||
return (1.0f / 6.0f) * (-a * a * a + 3.0f * a * a - 3.0f * a + 1.0f);
|
||||
}
|
||||
|
||||
float bspline_w1(float a) {
|
||||
return (1.0f / 6.0f) * (3.0f * a * a * a - 6.0f * a * a + 4.0f);
|
||||
}
|
||||
|
||||
float bspline_w2(float a) {
|
||||
return (1.0f / 6.0f) * (-3.0f * a * a * a + 3.0f * a * a + 3.0f * a + 1.0f);
|
||||
}
|
||||
|
||||
__host__ __device__ float bspline_w3(float a) {
|
||||
return (1.0f / 6.0f) * (a * a * a);
|
||||
}
|
||||
|
||||
void computeFPS() {
|
||||
frameCount++;
|
||||
fpsCount++;
|
||||
|
||||
if (fpsCount == fpsLimit - 1) {
|
||||
g_Verify = true;
|
||||
}
|
||||
|
||||
if (fpsCount == fpsLimit) {
|
||||
char fps[256];
|
||||
float ifps = 1.f / (sdkGetAverageTimerValue(&timer) / 1000.f);
|
||||
sprintf(fps, "%s %s <%s>: %3.1f fps", "", sSDKsample,
|
||||
sFilterMode[g_FilterMode], ifps);
|
||||
|
||||
glutSetWindowTitle(fps);
|
||||
fpsCount = 0;
|
||||
|
||||
sdkResetTimer(&timer);
|
||||
}
|
||||
}
|
||||
|
||||
void plotCurve(float (*func)(float)) {
|
||||
const int steps = 100;
|
||||
glBegin(GL_LINE_STRIP);
|
||||
|
||||
for (int i = 0; i < steps; i++) {
|
||||
float x = i / (float)(steps - 1);
|
||||
glVertex2f(x, func(x));
|
||||
}
|
||||
|
||||
glEnd();
|
||||
}
|
||||
|
||||
// display results using OpenGL (called by GLUT)
|
||||
void display() {
|
||||
sdkStartTimer(&timer);
|
||||
|
||||
// map PBO to get CUDA device pointer
|
||||
uchar4 *d_output;
|
||||
checkCudaErrors(cudaGraphicsMapResources(1, &cuda_pbo_resource, 0));
|
||||
size_t num_bytes;
|
||||
checkCudaErrors(cudaGraphicsResourceGetMappedPointer(
|
||||
(void **)&d_output, &num_bytes, cuda_pbo_resource));
|
||||
render(imageWidth, imageHeight, tx, ty, scale, cx, cy, blockSize, gridSize,
|
||||
g_FilterMode, d_output);
|
||||
|
||||
checkCudaErrors(cudaGraphicsUnmapResources(1, &cuda_pbo_resource, 0));
|
||||
|
||||
// Common display path
|
||||
{
|
||||
// display results
|
||||
glClear(GL_COLOR_BUFFER_BIT);
|
||||
|
||||
#if USE_BUFFER_TEX
|
||||
// display using buffer texture
|
||||
glBindTexture(GL_TEXTURE_BUFFER_EXT, bufferTex);
|
||||
glBindProgramARB(GL_FRAGMENT_PROGRAM_ARB, fprog);
|
||||
glEnable(GL_FRAGMENT_PROGRAM_ARB);
|
||||
glProgramLocalParameterI4iNV(GL_FRAGMENT_PROGRAM_ARB, 0, width, 0, 0, 0);
|
||||
#else
|
||||
// download image from PBO to OpenGL texture
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, pbo);
|
||||
glBindTexture(GL_TEXTURE_TYPE, displayTex);
|
||||
glPixelStorei(GL_UNPACK_ALIGNMENT, 1);
|
||||
glTexSubImage2D(GL_TEXTURE_TYPE, 0, 0, 0, width, height, GL_BGRA,
|
||||
GL_UNSIGNED_BYTE, 0);
|
||||
glEnable(GL_TEXTURE_TYPE);
|
||||
#endif
|
||||
|
||||
// draw textured quad
|
||||
glDisable(GL_DEPTH_TEST);
|
||||
glBegin(GL_QUADS);
|
||||
glTexCoord2f(0.0f, (GLfloat)height);
|
||||
glVertex2f(0.0f, 0.0f);
|
||||
glTexCoord2f((GLfloat)width, (GLfloat)height);
|
||||
glVertex2f(1.0f, 0.0f);
|
||||
glTexCoord2f((GLfloat)width, 0.0f);
|
||||
glVertex2f(1.0f, 1.0f);
|
||||
glTexCoord2f(0.0f, 0.0f);
|
||||
glVertex2f(0.0f, 1.0f);
|
||||
glEnd();
|
||||
glDisable(GL_TEXTURE_TYPE);
|
||||
glDisable(GL_FRAGMENT_PROGRAM_ARB);
|
||||
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, 0);
|
||||
|
||||
if (drawCurves) {
|
||||
// draw spline curves
|
||||
glPushMatrix();
|
||||
glScalef(0.25, 0.25, 1.0);
|
||||
|
||||
glTranslatef(0.0, 2.0, 0.0);
|
||||
glColor3f(1.0, 0.0, 0.0);
|
||||
plotCurve(bspline_w3);
|
||||
|
||||
glTranslatef(1.0, 0.0, 0.0);
|
||||
glColor3f(0.0, 1.0, 0.0);
|
||||
plotCurve(bspline_w2);
|
||||
|
||||
glTranslatef(1.0, 0.0, 0.0);
|
||||
glColor3f(0.0, 0.0, 1.0);
|
||||
plotCurve(bspline_w1);
|
||||
|
||||
glTranslatef(1.0, 0.0, 0.0);
|
||||
glColor3f(1.0, 0.0, 1.0);
|
||||
plotCurve(bspline_w0);
|
||||
|
||||
glPopMatrix();
|
||||
glColor3f(1.0, 1.0, 1.0);
|
||||
}
|
||||
}
|
||||
|
||||
glutSwapBuffers();
|
||||
glutReportErrors();
|
||||
|
||||
sdkStopTimer(&timer);
|
||||
|
||||
computeFPS();
|
||||
}
|
||||
|
||||
// GLUT callback functions
|
||||
void timerEvent(int value) {
|
||||
if (glutGetWindow()) {
|
||||
glutPostRedisplay();
|
||||
glutTimerFunc(REFRESH_DELAY, timerEvent, 0);
|
||||
}
|
||||
}
|
||||
|
||||
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
|
||||
|
||||
case '1':
|
||||
g_FilterMode = MODE_NEAREST;
|
||||
break;
|
||||
|
||||
case '2':
|
||||
g_FilterMode = MODE_BILINEAR;
|
||||
break;
|
||||
|
||||
case '3':
|
||||
g_FilterMode = MODE_BICUBIC;
|
||||
break;
|
||||
|
||||
case '4':
|
||||
g_FilterMode = MODE_FAST_BICUBIC;
|
||||
break;
|
||||
|
||||
case '5':
|
||||
g_FilterMode = MODE_CATMULL_ROM;
|
||||
break;
|
||||
|
||||
case '=':
|
||||
case '+':
|
||||
scale *= 0.5f;
|
||||
break;
|
||||
|
||||
case '-':
|
||||
scale *= 2.0f;
|
||||
break;
|
||||
|
||||
case 'r':
|
||||
scale = 1.0f;
|
||||
tx = ty = 0.0f;
|
||||
break;
|
||||
|
||||
case 'd':
|
||||
printf("%f, %f, %f\n", tx, ty, scale);
|
||||
break;
|
||||
|
||||
case 'b':
|
||||
runBenchmark(500);
|
||||
break;
|
||||
|
||||
case 'c':
|
||||
drawCurves ^= 1;
|
||||
break;
|
||||
|
||||
default:
|
||||
break;
|
||||
}
|
||||
|
||||
if (key >= '1' && key <= '5') {
|
||||
printf("> FilterMode[%d] = %s\n", g_FilterMode + 1,
|
||||
sFilterMode[g_FilterMode]);
|
||||
}
|
||||
}
|
||||
|
||||
int ox, oy;
|
||||
int buttonState = 0;
|
||||
|
||||
void mouse(int button, int state, int x, int y) {
|
||||
if (state == GLUT_DOWN) {
|
||||
buttonState |= 1 << button;
|
||||
} else if (state == GLUT_UP) {
|
||||
buttonState = 0;
|
||||
}
|
||||
|
||||
ox = x;
|
||||
oy = y;
|
||||
}
|
||||
|
||||
void motion(int x, int y) {
|
||||
float dx, dy;
|
||||
dx = (float)(x - ox);
|
||||
dy = (float)(y - oy);
|
||||
|
||||
if (buttonState & 1) {
|
||||
// left = translate
|
||||
tx -= dx * scale;
|
||||
ty -= dy * scale;
|
||||
} else if (buttonState & 2) {
|
||||
// middle = zoom
|
||||
scale -= dy / 1000.0f;
|
||||
}
|
||||
|
||||
ox = x;
|
||||
oy = y;
|
||||
}
|
||||
|
||||
void reshape(int x, int y) {
|
||||
width = x;
|
||||
height = y;
|
||||
imageWidth = width;
|
||||
imageHeight = height;
|
||||
|
||||
initGLBuffers();
|
||||
|
||||
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 cleanup() {
|
||||
freeTexture();
|
||||
checkCudaErrors(cudaGraphicsUnregisterResource(cuda_pbo_resource));
|
||||
|
||||
glDeleteBuffers(1, &pbo);
|
||||
|
||||
#if USE_BUFFER_TEX
|
||||
glDeleteTextures(1, &bufferTex);
|
||||
glDeleteProgramsARB(1, &fprog);
|
||||
#else
|
||||
glDeleteTextures(1, &displayTex);
|
||||
#endif
|
||||
|
||||
sdkDeleteTimer(&timer);
|
||||
}
|
||||
|
||||
int iDivUp(int a, int b) { return (a % b != 0) ? (a / b + 1) : (a / b); }
|
||||
|
||||
void initGLBuffers() {
|
||||
if (pbo) {
|
||||
// delete old buffer
|
||||
checkCudaErrors(cudaGraphicsUnregisterResource(cuda_pbo_resource));
|
||||
glDeleteBuffers(1, &pbo);
|
||||
}
|
||||
|
||||
// create pixel buffer object for display
|
||||
glGenBuffers(1, &pbo);
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, pbo);
|
||||
glBufferData(GL_PIXEL_UNPACK_BUFFER_ARB, width * height * sizeof(uchar4), 0,
|
||||
GL_STREAM_DRAW_ARB);
|
||||
glBindBuffer(GL_PIXEL_UNPACK_BUFFER_ARB, 0);
|
||||
|
||||
checkCudaErrors(cudaGraphicsGLRegisterBuffer(
|
||||
&cuda_pbo_resource, pbo, cudaGraphicsMapFlagsWriteDiscard));
|
||||
|
||||
#if USE_BUFFER_TEX
|
||||
|
||||
// create buffer texture, attach to pbo
|
||||
if (bufferTex) {
|
||||
glDeleteTextures(1, &bufferTex);
|
||||
}
|
||||
|
||||
glGenTextures(1, &bufferTex);
|
||||
glBindTexture(GL_TEXTURE_BUFFER_EXT, bufferTex);
|
||||
glTexBufferEXT(GL_TEXTURE_BUFFER_EXT, GL_RGBA8, pbo);
|
||||
glBindTexture(GL_TEXTURE_BUFFER_EXT, 0);
|
||||
#else
|
||||
|
||||
// create texture for display
|
||||
if (displayTex) {
|
||||
glDeleteTextures(1, &displayTex);
|
||||
}
|
||||
|
||||
glGenTextures(1, &displayTex);
|
||||
glBindTexture(GL_TEXTURE_TYPE, displayTex);
|
||||
glTexImage2D(GL_TEXTURE_TYPE, 0, GL_RGBA8, width, height, 0, GL_RGBA,
|
||||
GL_UNSIGNED_BYTE, NULL);
|
||||
glTexParameteri(GL_TEXTURE_TYPE, GL_TEXTURE_MIN_FILTER, GL_NEAREST);
|
||||
glTexParameteri(GL_TEXTURE_TYPE, GL_TEXTURE_MAG_FILTER, GL_NEAREST);
|
||||
glBindTexture(GL_TEXTURE_TYPE, 0);
|
||||
#endif
|
||||
|
||||
// calculate new grid size
|
||||
gridSize = dim3(iDivUp(width, blockSize.x), iDivUp(height, blockSize.y));
|
||||
}
|
||||
|
||||
void mainMenu(int i) { keyboard(i, 0, 0); }
|
||||
|
||||
void initMenus() {
|
||||
glutCreateMenu(mainMenu);
|
||||
glutAddMenuEntry("Nearest [1]", '1');
|
||||
glutAddMenuEntry("Bilinear [2]", '2');
|
||||
glutAddMenuEntry("Bicubic [3]", '3');
|
||||
glutAddMenuEntry("Fast Bicubic [4]", '4');
|
||||
glutAddMenuEntry("Catmull-Rom [5]", '5');
|
||||
glutAddMenuEntry("Zoom in [=]", '=');
|
||||
glutAddMenuEntry("Zoom out [-]", '-');
|
||||
glutAddMenuEntry("Benchmark [b]", 'b');
|
||||
glutAddMenuEntry("DrawCurves [c]", 'c');
|
||||
glutAddMenuEntry("Quit [esc]", 27);
|
||||
glutAttachMenu(GLUT_RIGHT_BUTTON);
|
||||
}
|
||||
|
||||
void runBenchmark(int iterations) {
|
||||
printf("[%s] (Benchmark Mode)\n", sSDKsample);
|
||||
|
||||
sdkCreateTimer(&timer);
|
||||
|
||||
uchar4 *d_output;
|
||||
checkCudaErrors(cudaGraphicsMapResources(1, &cuda_pbo_resource, 0));
|
||||
size_t num_bytes;
|
||||
checkCudaErrors(cudaGraphicsResourceGetMappedPointer(
|
||||
(void **)&d_output, &num_bytes, cuda_pbo_resource));
|
||||
|
||||
sdkStartTimer(&timer);
|
||||
|
||||
for (int i = 0; i < iterations; ++i) {
|
||||
render(imageWidth, imageHeight, tx, ty, scale, cx, cy, blockSize, gridSize,
|
||||
g_FilterMode, d_output);
|
||||
}
|
||||
|
||||
cudaDeviceSynchronize();
|
||||
sdkStopTimer(&timer);
|
||||
float time = sdkGetTimerValue(&timer) / (float)iterations;
|
||||
|
||||
checkCudaErrors(cudaGraphicsUnmapResources(1, &cuda_pbo_resource, 0));
|
||||
|
||||
printf("time: %0.3f ms, %f Mpixels/sec\n", time,
|
||||
(width * height / (time * 0.001f)) / 1e6);
|
||||
}
|
||||
|
||||
void runAutoTest(int argc, char **argv, const char *dump_filename,
|
||||
eFilterMode filter_mode) {
|
||||
cudaDeviceProp deviceProps;
|
||||
|
||||
int devID = findCudaDevice(argc, (const char **)argv);
|
||||
|
||||
checkCudaErrors(cudaGetDeviceProperties(&deviceProps, devID));
|
||||
|
||||
printf("[%s] (automated testing w/ readback)\n", sSDKsample);
|
||||
printf("CUDA device [%s] has %d Multi-Processors\n", deviceProps.name,
|
||||
deviceProps.multiProcessorCount);
|
||||
|
||||
loadImageData(argc, argv);
|
||||
|
||||
uchar4 *d_output;
|
||||
checkCudaErrors(cudaMalloc((void **)&d_output, imageWidth * imageHeight * 4));
|
||||
unsigned int *h_result =
|
||||
(unsigned int *)malloc(width * height * sizeof(unsigned int));
|
||||
|
||||
printf("AutoTest: %s Filter Mode: <%s>\n", sSDKsample,
|
||||
sFilterMode[g_FilterMode]);
|
||||
|
||||
render(imageWidth, imageHeight, tx, ty, scale, cx, cy, blockSize, gridSize,
|
||||
filter_mode, d_output);
|
||||
|
||||
// check if kernel execution generated an error
|
||||
getLastCudaError("Error: render (bicubicTexture) Kernel execution FAILED");
|
||||
checkCudaErrors(cudaDeviceSynchronize());
|
||||
|
||||
cudaMemcpy(h_result, d_output, imageWidth * imageHeight * 4,
|
||||
cudaMemcpyDeviceToHost);
|
||||
|
||||
sdkSavePPM4ub(dump_filename, (unsigned char *)h_result, imageWidth,
|
||||
imageHeight);
|
||||
|
||||
checkCudaErrors(cudaFree(d_output));
|
||||
free(h_result);
|
||||
}
|
||||
|
||||
#if USE_BUFFER_TEX
|
||||
// fragment program for reading from buffer texture
|
||||
static const char *shaderCode =
|
||||
"!!NVfp4.0\n"
|
||||
"INT PARAM width = program.local[0];\n"
|
||||
"INT TEMP index;\n"
|
||||
"FLR.S index, fragment.texcoord;\n"
|
||||
"MAD.S index.x, index.y, width, index.x;\n" // compute 1D index from 2D
|
||||
// coords
|
||||
"TXF result.color, index.x, texture[0], BUFFER;\n"
|
||||
"END";
|
||||
#endif
|
||||
|
||||
GLuint compileASMShader(GLenum program_type, const char *code) {
|
||||
GLuint program_id;
|
||||
glGenProgramsARB(1, &program_id);
|
||||
glBindProgramARB(program_type, program_id);
|
||||
glProgramStringARB(program_type, GL_PROGRAM_FORMAT_ASCII_ARB,
|
||||
(GLsizei)strlen(code), (GLubyte *)code);
|
||||
|
||||
GLint error_pos;
|
||||
glGetIntegerv(GL_PROGRAM_ERROR_POSITION_ARB, &error_pos);
|
||||
|
||||
if (error_pos != -1) {
|
||||
const GLubyte *error_string;
|
||||
error_string = glGetString(GL_PROGRAM_ERROR_STRING_ARB);
|
||||
fprintf(stderr, "Program error at position: %d\n%s\n", (int)error_pos,
|
||||
error_string);
|
||||
return 0;
|
||||
}
|
||||
|
||||
return program_id;
|
||||
}
|
||||
|
||||
void initialize(int argc, char **argv) {
|
||||
printf("[%s] (OpenGL Mode)\n", sSDKsample);
|
||||
|
||||
initGL(&argc, argv);
|
||||
|
||||
// use command-line specified CUDA device, otherwise use device with highest
|
||||
// Gflops/s
|
||||
int devID = findCudaDevice(argc, (const char **)argv);
|
||||
|
||||
// get number of SMs on this GPU
|
||||
cudaDeviceProp deviceProps;
|
||||
checkCudaErrors(cudaGetDeviceProperties(&deviceProps, devID));
|
||||
printf("CUDA device [%s] has %d Multi-Processors\n", deviceProps.name,
|
||||
deviceProps.multiProcessorCount);
|
||||
|
||||
// Create the timer (for fps measurement)
|
||||
sdkCreateTimer(&timer);
|
||||
|
||||
// load image from disk
|
||||
loadImageData(argc, argv);
|
||||
|
||||
printf(
|
||||
"\n"
|
||||
"\tControls\n"
|
||||
"\t=/- : Zoom in/out\n"
|
||||
"\tb : Run Benchmark g_FilterMode\n"
|
||||
"\tc : Draw Bicubic Spline Curve\n"
|
||||
"\t[esc] - Quit\n\n"
|
||||
|
||||
"\tPress number keys to change filtering g_FilterMode:\n\n"
|
||||
"\t1 : nearest filtering\n"
|
||||
"\t2 : bilinear filtering\n"
|
||||
"\t3 : bicubic filtering\n"
|
||||
"\t4 : fast bicubic filtering\n"
|
||||
"\t5 : Catmull-Rom filtering\n\n");
|
||||
|
||||
initGLBuffers();
|
||||
|
||||
#if USE_BUFFER_TEX
|
||||
fprog = compileASMShader(GL_FRAGMENT_PROGRAM_ARB, shaderCode);
|
||||
|
||||
if (!fprog) {
|
||||
exit(EXIT_SUCCESS);
|
||||
}
|
||||
|
||||
#endif
|
||||
}
|
||||
|
||||
void initGL(int *argc, char **argv) {
|
||||
// initialize GLUT callback functions
|
||||
glutInit(argc, argv);
|
||||
glutInitDisplayMode(GLUT_RGBA | GLUT_ALPHA | GLUT_DOUBLE | GLUT_DEPTH);
|
||||
glutInitWindowSize(width, height);
|
||||
glutCreateWindow("CUDA bicubic texture filtering");
|
||||
glutDisplayFunc(display);
|
||||
glutKeyboardFunc(keyboard);
|
||||
glutMouseFunc(mouse);
|
||||
glutMotionFunc(motion);
|
||||
glutReshapeFunc(reshape);
|
||||
glutTimerFunc(REFRESH_DELAY, timerEvent, 0);
|
||||
|
||||
#if defined(__APPLE__) || defined(MACOSX)
|
||||
atexit(cleanup);
|
||||
#else
|
||||
glutCloseFunc(cleanup);
|
||||
#endif
|
||||
|
||||
initMenus();
|
||||
|
||||
if (!isGLVersionSupported(2, 0) ||
|
||||
!areGLExtensionsSupported("GL_ARB_pixel_buffer_object")) {
|
||||
fprintf(stderr, "Required OpenGL extensions are missing.");
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
|
||||
#if USE_BUFFER_TEX
|
||||
|
||||
if (!areGLExtensionsSupported("GL_EXT_texture_buffer_object")) {
|
||||
fprintf(stderr,
|
||||
"OpenGL extension: GL_EXT_texture_buffer_object missing.\n");
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
|
||||
if (!areGLExtensionsSupported("GL_NV_gpu_program4")) {
|
||||
fprintf(stderr, "OpenGL extension: GL_NV_gpu_program4 missing.\n");
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
|
||||
#endif
|
||||
}
|
||||
|
||||
void loadImageData(int argc, char **argv) {
|
||||
// load image from disk
|
||||
uchar *h_data = NULL;
|
||||
char *srcImagePath = NULL;
|
||||
|
||||
if ((srcImagePath = sdkFindFilePath(srcImageFilename, argv[0])) == NULL) {
|
||||
printf("bicubicTexture loadImageData() could not find <%s>\nExiting...\n",
|
||||
srcImageFilename);
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
|
||||
sdkLoadPGM<unsigned char>(srcImagePath, &h_data, &imageWidth, &imageHeight);
|
||||
|
||||
printf("Loaded '%s', %d x %d pixels\n", srcImageFilename, imageWidth,
|
||||
imageHeight);
|
||||
|
||||
cx = imageWidth * 0.5f;
|
||||
cy = imageHeight * 0.5f;
|
||||
|
||||
// initialize texture
|
||||
initTexture(imageWidth, imageHeight, h_data);
|
||||
}
|
||||
|
||||
void printHelp() {
|
||||
printf("bicubicTexture Usage:\n");
|
||||
printf("\t-file=output.ppm (output file to save to disk)\n");
|
||||
printf(
|
||||
"\t-mode=n (0=Nearest, 1=Bilinear, 2=Bicubic, 3=Fast-Bicubic, "
|
||||
"4=Catmull-Rom\n");
|
||||
}
|
||||
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
// Program main
|
||||
////////////////////////////////////////////////////////////////////////////////
|
||||
int main(int argc, char **argv) {
|
||||
pArgc = &argc;
|
||||
pArgv = argv;
|
||||
|
||||
// parse arguments
|
||||
char *filename;
|
||||
|
||||
#if defined(__linux__)
|
||||
setenv("DISPLAY", ":0", 0);
|
||||
#endif
|
||||
|
||||
printf("Starting bicubicTexture\n");
|
||||
|
||||
if (checkCmdLineFlag(argc, (const char **)argv, "help")) {
|
||||
printHelp();
|
||||
exit(EXIT_SUCCESS);
|
||||
}
|
||||
|
||||
if (checkCmdLineFlag(argc, (const char **)argv, "mode")) {
|
||||
g_FilterMode =
|
||||
(eFilterMode)getCmdLineArgumentInt(argc, (const char **)argv, "mode");
|
||||
|
||||
if (g_FilterMode < 0 || g_FilterMode >= NUM_MODES) {
|
||||
printf("Invalid Mode setting %d\n", g_FilterMode);
|
||||
exit(EXIT_FAILURE);
|
||||
}
|
||||
}
|
||||
|
||||
if (getCmdLineArgumentString(argc, (const char **)argv, "file", &filename)) {
|
||||
dumpFilename = filename;
|
||||
fpsLimit = frameCheckNumber;
|
||||
|
||||
// Running CUDA kernel (bicubicFiltering) without visualization (QA
|
||||
// Testing/Verification)
|
||||
runAutoTest(argc, argv, (const char *)dumpFilename, g_FilterMode);
|
||||
} else {
|
||||
// This runs the CUDA kernel (bicubicFiltering) + OpenGL visualization
|
||||
initialize(argc, argv);
|
||||
glutMainLoop();
|
||||
}
|
||||
|
||||
exit(EXIT_SUCCESS);
|
||||
}
|
||||
128
Samples/bicubicTexture/bicubicTexture_cuda.cu
Normal file
128
Samples/bicubicTexture/bicubicTexture_cuda.cu
Normal file
@@ -0,0 +1,128 @@
|
||||
/* Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved.
|
||||
*
|
||||
* Redistribution and use in source and binary forms, with or without
|
||||
* modification, are permitted provided that the following conditions
|
||||
* are met:
|
||||
* * Redistributions of source code must retain the above copyright
|
||||
* notice, this list of conditions and the following disclaimer.
|
||||
* * Redistributions in binary form must reproduce the above copyright
|
||||
* notice, this list of conditions and the following disclaimer in the
|
||||
* documentation and/or other materials provided with the distribution.
|
||||
* * Neither the name of NVIDIA CORPORATION nor the names of its
|
||||
* contributors may be used to endorse or promote products derived
|
||||
* from this software without specific prior written permission.
|
||||
*
|
||||
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
|
||||
* EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
||||
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
|
||||
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
|
||||
* CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
|
||||
* EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
|
||||
* PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
|
||||
* PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
|
||||
* OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
||||
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
||||
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
*/
|
||||
|
||||
#ifndef _BICUBICTEXTURE_CU_
|
||||
#define _BICUBICTEXTURE_CU_
|
||||
|
||||
#include <stdlib.h>
|
||||
#include <stdio.h>
|
||||
#include <string.h>
|
||||
|
||||
#include <helper_math.h>
|
||||
|
||||
// includes, cuda
|
||||
#include <helper_cuda.h>
|
||||
|
||||
typedef unsigned int uint;
|
||||
typedef unsigned char uchar;
|
||||
|
||||
#include "bicubicTexture_kernel.cuh"
|
||||
|
||||
cudaArray *d_imageArray = 0;
|
||||
|
||||
extern "C" void initTexture(int imageWidth, int imageHeight, uchar *h_data) {
|
||||
// allocate array and copy image data
|
||||
cudaChannelFormatDesc channelDesc =
|
||||
cudaCreateChannelDesc(8, 0, 0, 0, cudaChannelFormatKindUnsigned);
|
||||
checkCudaErrors(
|
||||
cudaMallocArray(&d_imageArray, &channelDesc, imageWidth, imageHeight));
|
||||
checkCudaErrors(cudaMemcpy2DToArray(
|
||||
d_imageArray, 0, 0, h_data, imageWidth * sizeof(uchar),
|
||||
imageWidth * sizeof(uchar), imageHeight, cudaMemcpyHostToDevice));
|
||||
free(h_data);
|
||||
|
||||
cudaResourceDesc texRes;
|
||||
memset(&texRes, 0, sizeof(cudaResourceDesc));
|
||||
|
||||
texRes.resType = cudaResourceTypeArray;
|
||||
texRes.res.array.array = d_imageArray;
|
||||
|
||||
cudaTextureDesc texDescr;
|
||||
memset(&texDescr, 0, sizeof(cudaTextureDesc));
|
||||
|
||||
texDescr.normalizedCoords = false;
|
||||
texDescr.filterMode = cudaFilterModeLinear;
|
||||
texDescr.addressMode[0] = cudaAddressModeClamp;
|
||||
texDescr.addressMode[1] = cudaAddressModeClamp;
|
||||
texDescr.readMode = cudaReadModeNormalizedFloat;
|
||||
|
||||
checkCudaErrors(
|
||||
cudaCreateTextureObject(&texObjLinear, &texRes, &texDescr, NULL));
|
||||
|
||||
memset(&texDescr, 0, sizeof(cudaTextureDesc));
|
||||
texDescr.normalizedCoords = false;
|
||||
texDescr.filterMode = cudaFilterModePoint;
|
||||
texDescr.addressMode[0] = cudaAddressModeClamp;
|
||||
texDescr.addressMode[1] = cudaAddressModeClamp;
|
||||
texDescr.readMode = cudaReadModeNormalizedFloat;
|
||||
|
||||
checkCudaErrors(
|
||||
cudaCreateTextureObject(&texObjPoint, &texRes, &texDescr, NULL));
|
||||
}
|
||||
|
||||
extern "C" void freeTexture() {
|
||||
checkCudaErrors(cudaDestroyTextureObject(texObjPoint));
|
||||
checkCudaErrors(cudaDestroyTextureObject(texObjLinear));
|
||||
checkCudaErrors(cudaFreeArray(d_imageArray));
|
||||
}
|
||||
|
||||
// render image using CUDA
|
||||
extern "C" void render(int width, int height, float tx, float ty, float scale,
|
||||
float cx, float cy, dim3 blockSize, dim3 gridSize,
|
||||
int filter_mode, uchar4 *output) {
|
||||
// call CUDA kernel, writing results to PBO memory
|
||||
switch (filter_mode) {
|
||||
case MODE_NEAREST:
|
||||
d_render<<<gridSize, blockSize>>>(output, width, height, tx, ty, scale,
|
||||
cx, cy, texObjPoint);
|
||||
break;
|
||||
|
||||
case MODE_BILINEAR:
|
||||
d_render<<<gridSize, blockSize>>>(output, width, height, tx, ty, scale,
|
||||
cx, cy, texObjLinear);
|
||||
break;
|
||||
|
||||
case MODE_BICUBIC:
|
||||
d_renderBicubic<<<gridSize, blockSize>>>(output, width, height, tx, ty,
|
||||
scale, cx, cy, texObjPoint);
|
||||
break;
|
||||
|
||||
case MODE_FAST_BICUBIC:
|
||||
d_renderFastBicubic<<<gridSize, blockSize>>>(
|
||||
output, width, height, tx, ty, scale, cx, cy, texObjLinear);
|
||||
break;
|
||||
|
||||
case MODE_CATROM:
|
||||
d_renderCatRom<<<gridSize, blockSize>>>(output, width, height, tx, ty,
|
||||
scale, cx, cy, texObjPoint);
|
||||
break;
|
||||
}
|
||||
|
||||
getLastCudaError("kernel failed");
|
||||
}
|
||||
|
||||
#endif
|
||||
327
Samples/bicubicTexture/bicubicTexture_kernel.cuh
Normal file
327
Samples/bicubicTexture/bicubicTexture_kernel.cuh
Normal file
@@ -0,0 +1,327 @@
|
||||
/* Copyright (c) 2021, NVIDIA CORPORATION. All rights reserved.
|
||||
*
|
||||
* Redistribution and use in source and binary forms, with or without
|
||||
* modification, are permitted provided that the following conditions
|
||||
* are met:
|
||||
* * Redistributions of source code must retain the above copyright
|
||||
* notice, this list of conditions and the following disclaimer.
|
||||
* * Redistributions in binary form must reproduce the above copyright
|
||||
* notice, this list of conditions and the following disclaimer in the
|
||||
* documentation and/or other materials provided with the distribution.
|
||||
* * Neither the name of NVIDIA CORPORATION nor the names of its
|
||||
* contributors may be used to endorse or promote products derived
|
||||
* from this software without specific prior written permission.
|
||||
*
|
||||
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS ``AS IS'' AND ANY
|
||||
* EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
||||
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
|
||||
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR
|
||||
* CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
|
||||
* EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
|
||||
* PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
|
||||
* PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY
|
||||
* OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
||||
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
||||
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
*/
|
||||
|
||||
/*
|
||||
Bicubic filtering
|
||||
See GPU Gems 2: "Fast Third-Order Texture Filtering", Sigg & Hadwiger
|
||||
https://developer.nvidia.com/gpugems/gpugems2/part-iii-high-quality-rendering/chapter-20-fast-third-order-texture-filtering
|
||||
|
||||
Reformulation thanks to Keenan Crane
|
||||
*/
|
||||
|
||||
#ifndef _BICUBICTEXTURE_KERNEL_CUH_
|
||||
#define _BICUBICTEXTURE_KERNEL_CUH_
|
||||
|
||||
enum Mode {
|
||||
MODE_NEAREST,
|
||||
MODE_BILINEAR,
|
||||
MODE_BICUBIC,
|
||||
MODE_FAST_BICUBIC,
|
||||
MODE_CATROM
|
||||
};
|
||||
|
||||
cudaTextureObject_t texObjPoint, texObjLinear;
|
||||
|
||||
// w0, w1, w2, and w3 are the four cubic B-spline basis functions
|
||||
__host__ __device__ float w0(float a) {
|
||||
// return (1.0f/6.0f)*(-a*a*a + 3.0f*a*a - 3.0f*a + 1.0f);
|
||||
return (1.0f / 6.0f) * (a * (a * (-a + 3.0f) - 3.0f) + 1.0f); // optimized
|
||||
}
|
||||
|
||||
__host__ __device__ float w1(float a) {
|
||||
// return (1.0f/6.0f)*(3.0f*a*a*a - 6.0f*a*a + 4.0f);
|
||||
return (1.0f / 6.0f) * (a * a * (3.0f * a - 6.0f) + 4.0f);
|
||||
}
|
||||
|
||||
__host__ __device__ float w2(float a) {
|
||||
// return (1.0f/6.0f)*(-3.0f*a*a*a + 3.0f*a*a + 3.0f*a + 1.0f);
|
||||
return (1.0f / 6.0f) * (a * (a * (-3.0f * a + 3.0f) + 3.0f) + 1.0f);
|
||||
}
|
||||
|
||||
__host__ __device__ float w3(float a) { return (1.0f / 6.0f) * (a * a * a); }
|
||||
|
||||
// g0 and g1 are the two amplitude functions
|
||||
__device__ float g0(float a) { return w0(a) + w1(a); }
|
||||
|
||||
__device__ float g1(float a) { return w2(a) + w3(a); }
|
||||
|
||||
// h0 and h1 are the two offset functions
|
||||
__device__ float h0(float a) {
|
||||
// note +0.5 offset to compensate for CUDA linear filtering convention
|
||||
return -1.0f + w1(a) / (w0(a) + w1(a)) + 0.5f;
|
||||
}
|
||||
|
||||
__device__ float h1(float a) { return 1.0f + w3(a) / (w2(a) + w3(a)) + 0.5f; }
|
||||
|
||||
// filter 4 values using cubic splines
|
||||
template <class T>
|
||||
__device__ T cubicFilter(float x, T c0, T c1, T c2, T c3) {
|
||||
T r;
|
||||
r = c0 * w0(x);
|
||||
r += c1 * w1(x);
|
||||
r += c2 * w2(x);
|
||||
r += c3 * w3(x);
|
||||
return r;
|
||||
}
|
||||
|
||||
// slow but precise bicubic lookup using 16 texture lookups
|
||||
template <class T, class R> // texture data type, return type
|
||||
__device__ R tex2DBicubic(const cudaTextureObject_t tex, float x, float y) {
|
||||
x -= 0.5f;
|
||||
y -= 0.5f;
|
||||
float px = floorf(x);
|
||||
float py = floorf(y);
|
||||
float fx = x - px;
|
||||
float fy = y - py;
|
||||
|
||||
return cubicFilter<R>(
|
||||
fy, cubicFilter<R>(
|
||||
fx, tex2D<R>(tex, px - 1, py - 1), tex2D<R>(tex, px, py - 1),
|
||||
tex2D<R>(tex, px + 1, py - 1), tex2D<R>(tex, px + 2, py - 1)),
|
||||
cubicFilter<R>(fx, tex2D<R>(tex, px - 1, py), tex2D<R>(tex, px, py),
|
||||
tex2D<R>(tex, px + 1, py), tex2D<R>(tex, px + 2, py)),
|
||||
cubicFilter<R>(fx, tex2D<R>(tex, px - 1, py + 1),
|
||||
tex2D<R>(tex, px, py + 1), tex2D<R>(tex, px + 1, py + 1),
|
||||
tex2D<R>(tex, px + 2, py + 1)),
|
||||
cubicFilter<R>(fx, tex2D<R>(tex, px - 1, py + 2),
|
||||
tex2D<R>(tex, px, py + 2), tex2D<R>(tex, px + 1, py + 2),
|
||||
tex2D<R>(tex, px + 2, py + 2)));
|
||||
}
|
||||
|
||||
// fast bicubic texture lookup using 4 bilinear lookups
|
||||
// assumes texture is set to non-normalized coordinates, point sampling
|
||||
template <class T, class R> // texture data type, return type
|
||||
__device__ R tex2DFastBicubic(const cudaTextureObject_t tex, float x, float y) {
|
||||
x -= 0.5f;
|
||||
y -= 0.5f;
|
||||
float px = floorf(x);
|
||||
float py = floorf(y);
|
||||
float fx = x - px;
|
||||
float fy = y - py;
|
||||
|
||||
// note: we could store these functions in a lookup table texture, but maths
|
||||
// is cheap
|
||||
float g0x = g0(fx);
|
||||
float g1x = g1(fx);
|
||||
float h0x = h0(fx);
|
||||
float h1x = h1(fx);
|
||||
float h0y = h0(fy);
|
||||
float h1y = h1(fy);
|
||||
|
||||
R r = g0(fy) * (g0x * tex2D<R>(tex, px + h0x, py + h0y) +
|
||||
g1x * tex2D<R>(tex, px + h1x, py + h0y)) +
|
||||
g1(fy) * (g0x * tex2D<R>(tex, px + h0x, py + h1y) +
|
||||
g1x * tex2D<R>(tex, px + h1x, py + h1y));
|
||||
return r;
|
||||
}
|
||||
|
||||
// higher-precision 2D bilinear lookup
|
||||
template <class T, class R> // texture data type, return type
|
||||
__device__ R tex2DBilinear(const cudaTextureObject_t tex, float x, float y) {
|
||||
x -= 0.5f;
|
||||
y -= 0.5f;
|
||||
float px = floorf(x); // integer position
|
||||
float py = floorf(y);
|
||||
float fx = x - px; // fractional position
|
||||
float fy = y - py;
|
||||
px += 0.5f;
|
||||
py += 0.5f;
|
||||
|
||||
return lerp(lerp(tex2D<R>(tex, px, py), tex2D<R>(tex, px + 1.0f, py), fx),
|
||||
lerp(tex2D<R>(tex, px, py + 1.0f),
|
||||
tex2D<R>(tex, px + 1.0f, py + 1.0f), fx),
|
||||
fy);
|
||||
}
|
||||
|
||||
#if !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 200
|
||||
|
||||
/*
|
||||
bilinear 2D texture lookup using tex2Dgather function
|
||||
- tex2Dgather() returns the four neighbouring samples in a single texture
|
||||
lookup
|
||||
- it is only supported on the Fermi architecture
|
||||
- you can select which component to fetch using the "comp" parameter
|
||||
- it can be used to efficiently implement custom texture filters
|
||||
|
||||
The samples are returned in a 4-vector in the following order:
|
||||
x: (0, 1)
|
||||
y: (1, 1)
|
||||
z: (1, 0)
|
||||
w: (0, 0)
|
||||
*/
|
||||
|
||||
template <class T, class R> // texture data type, return type
|
||||
__device__ float tex2DBilinearGather(const cudaTextureObject_t tex, float x,
|
||||
float y, int comp = 0) {
|
||||
x -= 0.5f;
|
||||
y -= 0.5f;
|
||||
float px = floorf(x); // integer position
|
||||
float py = floorf(y);
|
||||
float fx = x - px; // fractional position
|
||||
float fy = y - py;
|
||||
|
||||
R samples = tex2Dgather<R>(tex, px + 0.5f, py + 0.5f, comp);
|
||||
|
||||
return lerp(lerp((float)samples.w, (float)samples.z, fx),
|
||||
lerp((float)samples.x, (float)samples.y, fx), fy);
|
||||
}
|
||||
|
||||
#endif
|
||||
|
||||
// Catmull-Rom interpolation
|
||||
|
||||
__host__ __device__ float catrom_w0(float a) {
|
||||
// return -0.5f*a + a*a - 0.5f*a*a*a;
|
||||
return a * (-0.5f + a * (1.0f - 0.5f * a));
|
||||
}
|
||||
|
||||
__host__ __device__ float catrom_w1(float a) {
|
||||
// return 1.0f - 2.5f*a*a + 1.5f*a*a*a;
|
||||
return 1.0f + a * a * (-2.5f + 1.5f * a);
|
||||
}
|
||||
|
||||
__host__ __device__ float catrom_w2(float a) {
|
||||
// return 0.5f*a + 2.0f*a*a - 1.5f*a*a*a;
|
||||
return a * (0.5f + a * (2.0f - 1.5f * a));
|
||||
}
|
||||
|
||||
__host__ __device__ float catrom_w3(float a) {
|
||||
// return -0.5f*a*a + 0.5f*a*a*a;
|
||||
return a * a * (-0.5f + 0.5f * a);
|
||||
}
|
||||
|
||||
template <class T>
|
||||
__device__ T catRomFilter(float x, T c0, T c1, T c2, T c3) {
|
||||
T r;
|
||||
r = c0 * catrom_w0(x);
|
||||
r += c1 * catrom_w1(x);
|
||||
r += c2 * catrom_w2(x);
|
||||
r += c3 * catrom_w3(x);
|
||||
return r;
|
||||
}
|
||||
|
||||
// Note - can't use bilinear trick here because of negative lobes
|
||||
template <class T, class R> // texture data type, return type
|
||||
__device__ R tex2DCatRom(const cudaTextureObject_t tex, float x, float y) {
|
||||
x -= 0.5f;
|
||||
y -= 0.5f;
|
||||
float px = floorf(x);
|
||||
float py = floorf(y);
|
||||
float fx = x - px;
|
||||
float fy = y - py;
|
||||
|
||||
return catRomFilter<R>(
|
||||
fy, catRomFilter<R>(
|
||||
fx, tex2D<R>(tex, px - 1, py - 1), tex2D<R>(tex, px, py - 1),
|
||||
tex2D<R>(tex, px + 1, py - 1), tex2D<R>(tex, px + 2, py - 1)),
|
||||
catRomFilter<R>(fx, tex2D<R>(tex, px - 1, py), tex2D<R>(tex, px, py),
|
||||
tex2D<R>(tex, px + 1, py), tex2D<R>(tex, px + 2, py)),
|
||||
catRomFilter<R>(fx, tex2D<R>(tex, px - 1, py + 1),
|
||||
tex2D<R>(tex, px, py + 1), tex2D<R>(tex, px + 1, py + 1),
|
||||
tex2D<R>(tex, px + 2, py + 1)),
|
||||
catRomFilter<R>(fx, tex2D<R>(tex, px - 1, py + 2),
|
||||
tex2D<R>(tex, px, py + 2), tex2D<R>(tex, px + 1, py + 2),
|
||||
tex2D<R>(tex, px + 2, py + 2)));
|
||||
}
|
||||
|
||||
// test functions
|
||||
|
||||
// render image using normal bilinear texture lookup
|
||||
__global__ void d_render(uchar4 *d_output, uint width, uint height, float tx,
|
||||
float ty, float scale, float cx, float cy,
|
||||
cudaTextureObject_t texObj) {
|
||||
uint x = __umul24(blockIdx.x, blockDim.x) + threadIdx.x;
|
||||
uint y = __umul24(blockIdx.y, blockDim.y) + threadIdx.y;
|
||||
uint i = __umul24(y, width) + x;
|
||||
|
||||
float u = (x - cx) * scale + cx + tx;
|
||||
float v = (y - cy) * scale + cy + ty;
|
||||
|
||||
if ((x < width) && (y < height)) {
|
||||
// write output color
|
||||
float c = tex2D<float>(texObj, u, v);
|
||||
// float c = tex2DBilinear<uchar, float>(tex, u, v);
|
||||
// float c = tex2DBilinearGather<uchar, uchar4>(tex2, u, v, 0) / 255.0f;
|
||||
d_output[i] = make_uchar4(c * 0xff, c * 0xff, c * 0xff, 0);
|
||||
}
|
||||
}
|
||||
|
||||
// render image using bicubic texture lookup
|
||||
__global__ void d_renderBicubic(uchar4 *d_output, uint width, uint height,
|
||||
float tx, float ty, float scale, float cx,
|
||||
float cy, cudaTextureObject_t texObj) {
|
||||
uint x = __umul24(blockIdx.x, blockDim.x) + threadIdx.x;
|
||||
uint y = __umul24(blockIdx.y, blockDim.y) + threadIdx.y;
|
||||
uint i = __umul24(y, width) + x;
|
||||
|
||||
float u = (x - cx) * scale + cx + tx;
|
||||
float v = (y - cy) * scale + cy + ty;
|
||||
|
||||
if ((x < width) && (y < height)) {
|
||||
// write output color
|
||||
float c = tex2DBicubic<uchar, float>(texObj, u, v);
|
||||
d_output[i] = make_uchar4(c * 0xff, c * 0xff, c * 0xff, 0);
|
||||
}
|
||||
}
|
||||
|
||||
// render image using fast bicubic texture lookup
|
||||
__global__ void d_renderFastBicubic(uchar4 *d_output, uint width, uint height,
|
||||
float tx, float ty, float scale, float cx,
|
||||
float cy, cudaTextureObject_t texObj) {
|
||||
uint x = __umul24(blockIdx.x, blockDim.x) + threadIdx.x;
|
||||
uint y = __umul24(blockIdx.y, blockDim.y) + threadIdx.y;
|
||||
uint i = __umul24(y, width) + x;
|
||||
|
||||
float u = (x - cx) * scale + cx + tx;
|
||||
float v = (y - cy) * scale + cy + ty;
|
||||
|
||||
if ((x < width) && (y < height)) {
|
||||
// write output color
|
||||
float c = tex2DFastBicubic<uchar, float>(texObj, u, v);
|
||||
d_output[i] = make_uchar4(c * 0xff, c * 0xff, c * 0xff, 0);
|
||||
}
|
||||
}
|
||||
|
||||
// render image using Catmull-Rom texture lookup
|
||||
__global__ void d_renderCatRom(uchar4 *d_output, uint width, uint height,
|
||||
float tx, float ty, float scale, float cx,
|
||||
float cy, cudaTextureObject_t texObj) {
|
||||
uint x = __umul24(blockIdx.x, blockDim.x) + threadIdx.x;
|
||||
uint y = __umul24(blockIdx.y, blockDim.y) + threadIdx.y;
|
||||
uint i = __umul24(y, width) + x;
|
||||
|
||||
float u = (x - cx) * scale + cx + tx;
|
||||
float v = (y - cy) * scale + cy + ty;
|
||||
|
||||
if ((x < width) && (y < height)) {
|
||||
// write output color
|
||||
float c = tex2DCatRom<uchar, float>(texObj, u, v);
|
||||
d_output[i] = make_uchar4(c * 0xff, c * 0xff, c * 0xff, 0);
|
||||
}
|
||||
}
|
||||
|
||||
#endif // _BICUBICTEXTURE_KERNEL_CUH_
|
||||
20
Samples/bicubicTexture/bicubicTexture_vs2017.sln
Normal file
20
Samples/bicubicTexture/bicubicTexture_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}") = "bicubicTexture", "bicubicTexture_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
|
||||
124
Samples/bicubicTexture/bicubicTexture_vs2017.vcxproj
Normal file
124
Samples/bicubicTexture/bicubicTexture_vs2017.vcxproj
Normal file
@@ -0,0 +1,124 @@
|
||||
<?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>bicubicTexture_vs2017</RootNamespace>
|
||||
<ProjectName>bicubicTexture</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 11.5.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)/bicubicTexture.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
|
||||
<Include>./;../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
<AdditionalOptions>--threads 0</AdditionalOptions>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Configuration)'=='Debug'">
|
||||
<ClCompile>
|
||||
<Optimization>Disabled</Optimization>
|
||||
<RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>true</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>Default</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MTd</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Configuration)'=='Release'">
|
||||
<ClCompile>
|
||||
<Optimization>MaxSpeed</Optimization>
|
||||
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MT</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Platform)'=='Win32'">
|
||||
<Link>
|
||||
|
||||
</Link>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Platform)'=='x64'">
|
||||
<Link>
|
||||
<AdditionalDependencies>glew64.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
</Link>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<ClCompile Include="bicubicTexture.cpp" />
|
||||
<CudaCompile Include="bicubicTexture_cuda.cu" />
|
||||
<None Include="bicubicTexture_kernel.cuh" />
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.5.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
20
Samples/bicubicTexture/bicubicTexture_vs2019.sln
Normal file
20
Samples/bicubicTexture/bicubicTexture_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}") = "bicubicTexture", "bicubicTexture_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
|
||||
120
Samples/bicubicTexture/bicubicTexture_vs2019.vcxproj
Normal file
120
Samples/bicubicTexture/bicubicTexture_vs2019.vcxproj
Normal file
@@ -0,0 +1,120 @@
|
||||
<?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>bicubicTexture_vs2019</RootNamespace>
|
||||
<ProjectName>bicubicTexture</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 11.5.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)/bicubicTexture.exe</OutputFile>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<CodeGeneration>compute_35,sm_35;compute_37,sm_37;compute_50,sm_50;compute_52,sm_52;compute_60,sm_60;compute_61,sm_61;compute_70,sm_70;compute_75,sm_75;compute_80,sm_80;compute_86,sm_86;</CodeGeneration>
|
||||
<AdditionalOptions>-Xcompiler "/wd 4819" %(AdditionalOptions)</AdditionalOptions>
|
||||
<Include>./;../../Common</Include>
|
||||
<Defines>WIN32</Defines>
|
||||
<AdditionalOptions>--threads 0</AdditionalOptions>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Configuration)'=='Debug'">
|
||||
<ClCompile>
|
||||
<Optimization>Disabled</Optimization>
|
||||
<RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>true</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>Default</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MTd</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Configuration)'=='Release'">
|
||||
<ClCompile>
|
||||
<Optimization>MaxSpeed</Optimization>
|
||||
<RuntimeLibrary>MultiThreaded</RuntimeLibrary>
|
||||
</ClCompile>
|
||||
<Link>
|
||||
<GenerateDebugInformation>false</GenerateDebugInformation>
|
||||
<LinkTimeCodeGeneration>UseLinkTimeCodeGeneration</LinkTimeCodeGeneration>
|
||||
</Link>
|
||||
<CudaCompile>
|
||||
<Runtime>MT</Runtime>
|
||||
<TargetMachinePlatform>64</TargetMachinePlatform>
|
||||
</CudaCompile>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Platform)'=='Win32'">
|
||||
<Link>
|
||||
|
||||
</Link>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemDefinitionGroup Condition="'$(Platform)'=='x64'">
|
||||
<Link>
|
||||
<AdditionalDependencies>glew64.lib;%(AdditionalDependencies)</AdditionalDependencies>
|
||||
</Link>
|
||||
</ItemDefinitionGroup>
|
||||
<ItemGroup>
|
||||
<ClCompile Include="bicubicTexture.cpp" />
|
||||
<CudaCompile Include="bicubicTexture_cuda.cu" />
|
||||
<None Include="bicubicTexture_kernel.cuh" />
|
||||
</ItemGroup>
|
||||
<Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
|
||||
<ImportGroup Label="ExtensionTargets">
|
||||
<Import Project="$(CUDAPropsPath)\CUDA 11.5.targets" />
|
||||
</ImportGroup>
|
||||
</Project>
|
||||
5
Samples/bicubicTexture/data/0_nearest.ppm
Normal file
5
Samples/bicubicTexture/data/0_nearest.ppm
Normal file
File diff suppressed because one or more lines are too long
5
Samples/bicubicTexture/data/1_bilinear.ppm
Normal file
5
Samples/bicubicTexture/data/1_bilinear.ppm
Normal file
File diff suppressed because one or more lines are too long
5
Samples/bicubicTexture/data/2_bicubic.ppm
Normal file
5
Samples/bicubicTexture/data/2_bicubic.ppm
Normal file
File diff suppressed because one or more lines are too long
5
Samples/bicubicTexture/data/3_fastbicubic.ppm
Normal file
5
Samples/bicubicTexture/data/3_fastbicubic.ppm
Normal file
File diff suppressed because one or more lines are too long
5
Samples/bicubicTexture/data/4_catmull-rom.ppm
Normal file
5
Samples/bicubicTexture/data/4_catmull-rom.ppm
Normal file
File diff suppressed because one or more lines are too long
5
Samples/bicubicTexture/data/lena_bw.pgm
Normal file
5
Samples/bicubicTexture/data/lena_bw.pgm
Normal file
File diff suppressed because one or more lines are too long
145
Samples/bicubicTexture/findgllib.mk
Normal file
145
Samples/bicubicTexture/findgllib.mk
Normal file
@@ -0,0 +1,145 @@
|
||||
################################################################################
|
||||
# Copyright (c) 2020, 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
|
||||
|
||||
Reference in New Issue
Block a user