CIS565-Fall-2014 · CyborgYL · Sep 22, 2014 · Sep 22, 2014
diff --git a/Answers.txt b/Answers.txt
@@ -0,0 +1,8 @@
+How does changing the tile and block sizes change performance? Why?
+increasing block size will gain performance while registers in a single block is not fully used. However if the block size reaches the level that all registers are used, the performance will decrease. Increasing tile size will gain performance because we can utilize shared memory.
+
+How does changing the number of planets change performance? Why?
+If the number of planets is smaller than the number of threads, it will not affect the performance. Otherwise, increasing the number of planets will slow down the program. Because it is embarrasing parallel problem.all threads are calculated parallelly. If the number of planets are greater than the threads, a thread will have to wait to calculate other planets before completing the current one.
+
+Without running experiments, how would you expect the serial and GPU verions of matrix_math to compare? Why?
+We can compare the runtime of computing matrices of different sizes. We can expect that when size is small, CPU version is faster because CPU runs faster than GPU at single thread. However when matrix size exceed a certain level, GPU version runs faster because it is massively parallel.
diff --git a/Part1/PROJ_WIN/CIS565_PROJ_1.suo b/Part1/PROJ_WIN/CIS565_PROJ_1.suo
diff --git a/Part1/PROJ_WIN/CIS565_PROJ_1/CIS565_PROJ_1.vcxproj b/Part1/PROJ_WIN/CIS565_PROJ_1/CIS565_PROJ_1.vcxproj
@@ -30,7 +30,7 @@
   </PropertyGroup>
   <Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
   <ImportGroup Label="ExtensionSettings">
-    <Import Project="$(VCTargetsPath)\BuildCustomizations\CUDA 5.5.props" />
+    <Import Project="$(VCTargetsPath)\BuildCustomizations\CUDA 6.5.props" />
   </ImportGroup>
   <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
     <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
@@ -114,6 +114,6 @@
   </ItemGroup>
   <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
   <ImportGroup Label="ExtensionTargets">
-    <Import Project="$(VCTargetsPath)\BuildCustomizations\CUDA 5.5.targets" />
+    <Import Project="$(VCTargetsPath)\BuildCustomizations\CUDA 6.5.targets" />
   </ImportGroup>
 </Project>
diff --git a/Part1/PROJ_WIN/CIS565_PROJ_1/vc100.pdb b/Part1/PROJ_WIN/CIS565_PROJ_1/vc100.pdb
diff --git a/Part1/PROJ_WIN/src/kernel.cu.deps b/Part1/PROJ_WIN/src/kernel.cu.deps
diff --git a/Part1/src/kernel.cu b/Part1/src/kernel.cu
@@ -9,11 +9,12 @@
 dim3 threadsPerBlock(blockSize);
 
 int numObjects;
-const float planetMass = 3e8;
+const __device__ float planetMass = 3e8;
 const __device__ float starMass = 5e10;
 
 const float scene_scale = 2e2; //size of the height map in simulation space
 
+
 glm::vec4 * dev_pos;
 glm::vec3 * dev_vel;
 glm::vec3 * dev_acc;
@@ -87,21 +88,63 @@ __global__ void generateCircularVelArray(int time, int N, glm::vec3 * arr, glm::
 //		 HINT : You may want to write a helper function that will help you 
 //              calculate the acceleration contribution of a single body.
 //		 REMEMBER : F = (G * m_a * m_b) / (r_ab ^ 2)
+
+__device__ glm::vec3 acc_pair(int N, glm::vec4 my_pos, glm::vec4 other_pos, int star)
+{
+	float coef=0;
+	glm::vec3 dir(0);
+	dir =  glm::vec3(other_pos.x - my_pos.x,other_pos.y - my_pos.y, 0);
+	float len = glm::length(dir);
+	if(len < ZERO_ABSORPTION_EPSILON)
+		return glm::vec3(0);
+	else
+	{
+		if(star >0 )
+			coef = G * starMass / (len * len *len);
+		else
+			coef = G* planetMass / (len * len *len);
+	}
+	return coef * dir;
+}
+
 __device__  glm::vec3 accelerate(int N, glm::vec4 my_pos, glm::vec4 * their_pos)
 {
-    return glm::vec3(0.0f);
+	int index = (blockIdx.x * blockDim.x) + threadIdx.x;
+	glm::vec3 acc(0);
+	if(index < N)
+	{
+		//star acc
+		acc = acc_pair(N, my_pos,their_pos[0],1);
+		for(int i=0; i < N; ++i){
+			acc +=  acc_pair(N,my_pos,their_pos[i],0);
+		}
+	}
+    return acc;
 }
 
+
 // TODO : update the acceleration of each body
 __global__ void updateF(int N, float dt, glm::vec4 * pos, glm::vec3 * vel, glm::vec3 * acc)
 {
 	// FILL IN HERE
+	int index = (blockIdx.x * blockDim.x) + threadIdx.x;
+	if(index < N)
+	{
+		acc[index] = accelerate(N,pos[index],pos);
+	}
 }
 
 // TODO : update velocity and position using a simple Euler integration scheme
 __global__ void updateS(int N, float dt, glm::vec4 * pos, glm::vec3 * vel, glm::vec3 * acc)
 {
 	// FILL IN HERE
+	int index = (blockIdx.x * blockDim.x) + threadIdx.x;
+	if(index < N)
+	{
+		vel[index] += acc[index] * dt;
+		pos[index] += glm::vec4(vel[index] *dt,0);
+
+	}
 }
 
 // Update the vertex buffer object
@@ -180,6 +223,10 @@ void initCuda(int N)
 void cudaNBodyUpdateWrapper(float dt)
 {
 	// FILL IN HERE
+	dim3 fullBlocksPerGrid((int)ceil(float(numObjects)/float(blockSize)));
+	updateF<<<fullBlocksPerGrid, blockSize>>>(numObjects,dt,dev_pos,dev_vel,dev_acc);
+	updateS<<<fullBlocksPerGrid, blockSize>>>(numObjects,dt,dev_pos,dev_vel,dev_acc);
+
 }
 
 void cudaUpdateVBO(float * vbodptr, int width, int height)

diff --git a/Part1/src/kernel.h b/Part1/src/kernel.h
@@ -6,7 +6,7 @@
 #include <cuda.h>
 #include <cmath>
 
-#define blockSize 128
+#define blockSize 256
 #define checkCUDAErrorWithLine(msg) checkCUDAError(msg, __LINE__)
 #define SHARED 0
 

diff --git a/Part1/src/main.cpp b/Part1/src/main.cpp
@@ -4,7 +4,7 @@
 
 #include "main.h"
 
-#define N_FOR_VIS 5000
+#define N_FOR_VIS 8000
 #define DT 0.2
 #define VISUALIZE 1
 //-------------------------------

diff --git a/Part1/src/utilities.h b/Part1/src/utilities.h
@@ -23,7 +23,7 @@
 #define E                           2.7182818284590452353602874713526624977572
 #define G                           6.67384e-11
 #define EPSILON                     .000000001
-#define ZERO_ABSORPTION_EPSILON     0.00001
+#define ZERO_ABSORPTION_EPSILON     1
 
 namespace utilityCore {
     extern float clamp(float f, float min, float max);

diff --git a/Part2/Matrix_math/Matrix_math.sln b/Part2/Matrix_math/Matrix_math.sln
@@ -0,0 +1,34 @@
+
+Microsoft Visual Studio Solution File, Format Version 11.00
+# Visual Studio 2010
+Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "Matrix_math", "Matrix_math\Matrix_math.vcxproj", "{88D7C9B9-8065-4912-8561-A15539ABE405}"
+EndProject
+Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "matrix_cpu", "matrix_cpu\matrix_cpu.vcxproj", "{7AE7EE73-9D50-4A8B-AAE6-0ED6A81C5D43}"
+EndProject
+Global
+	GlobalSection(SolutionConfigurationPlatforms) = preSolution
+		Debug|Win32 = Debug|Win32
+		Debug|x64 = Debug|x64
+		Release|Win32 = Release|Win32
+		Release|x64 = Release|x64
+	EndGlobalSection
+	GlobalSection(ProjectConfigurationPlatforms) = postSolution
+		{88D7C9B9-8065-4912-8561-A15539ABE405}.Debug|Win32.ActiveCfg = Debug|Win32
+		{88D7C9B9-8065-4912-8561-A15539ABE405}.Debug|Win32.Build.0 = Debug|Win32
+		{88D7C9B9-8065-4912-8561-A15539ABE405}.Debug|x64.ActiveCfg = Debug|x64
+		{88D7C9B9-8065-4912-8561-A15539ABE405}.Debug|x64.Build.0 = Debug|x64
+		{88D7C9B9-8065-4912-8561-A15539ABE405}.Release|Win32.ActiveCfg = Release|Win32
+		{88D7C9B9-8065-4912-8561-A15539ABE405}.Release|Win32.Build.0 = Release|Win32
+		{88D7C9B9-8065-4912-8561-A15539ABE405}.Release|x64.ActiveCfg = Release|x64
+		{88D7C9B9-8065-4912-8561-A15539ABE405}.Release|x64.Build.0 = Release|x64
+		{7AE7EE73-9D50-4A8B-AAE6-0ED6A81C5D43}.Debug|Win32.ActiveCfg = Debug|Win32
+		{7AE7EE73-9D50-4A8B-AAE6-0ED6A81C5D43}.Debug|Win32.Build.0 = Debug|Win32
+		{7AE7EE73-9D50-4A8B-AAE6-0ED6A81C5D43}.Debug|x64.ActiveCfg = Debug|Win32
+		{7AE7EE73-9D50-4A8B-AAE6-0ED6A81C5D43}.Release|Win32.ActiveCfg = Release|Win32
+		{7AE7EE73-9D50-4A8B-AAE6-0ED6A81C5D43}.Release|Win32.Build.0 = Release|Win32
+		{7AE7EE73-9D50-4A8B-AAE6-0ED6A81C5D43}.Release|x64.ActiveCfg = Release|Win32
+	EndGlobalSection
+	GlobalSection(SolutionProperties) = preSolution
+		HideSolutionNode = FALSE
+	EndGlobalSection
+EndGlobal
diff --git a/Part2/Matrix_math/Matrix_math/Matrix_math.vcxproj b/Part2/Matrix_math/Matrix_math/Matrix_math.vcxproj
@@ -0,0 +1,158 @@
+<?xml version="1.0" encoding="utf-8"?>
+<Project DefaultTargets="Build" ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
+  <ItemGroup Label="ProjectConfigurations">
+    <ProjectConfiguration Include="Debug|Win32">
+      <Configuration>Debug</Configuration>
+      <Platform>Win32</Platform>
+    </ProjectConfiguration>
+    <ProjectConfiguration Include="Debug|x64">
+      <Configuration>Debug</Configuration>
+      <Platform>x64</Platform>
+    </ProjectConfiguration>
+    <ProjectConfiguration Include="Release|Win32">
+      <Configuration>Release</Configuration>
+      <Platform>Win32</Platform>
+    </ProjectConfiguration>
+    <ProjectConfiguration Include="Release|x64">
+      <Configuration>Release</Configuration>
+      <Platform>x64</Platform>
+    </ProjectConfiguration>
+  </ItemGroup>
+  <PropertyGroup Label="Globals">
+    <ProjectGuid>{88D7C9B9-8065-4912-8561-A15539ABE405}</ProjectGuid>
+    <RootNamespace>Matrix_math</RootNamespace>
+  </PropertyGroup>
+  <Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'" Label="Configuration">
+    <ConfigurationType>Application</ConfigurationType>
+    <UseDebugLibraries>true</UseDebugLibraries>
+    <CharacterSet>MultiByte</CharacterSet>
+  </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'" Label="Configuration">
+    <ConfigurationType>Application</ConfigurationType>
+    <UseDebugLibraries>true</UseDebugLibraries>
+    <CharacterSet>MultiByte</CharacterSet>
+  </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'" Label="Configuration">
+    <ConfigurationType>Application</ConfigurationType>
+    <UseDebugLibraries>false</UseDebugLibraries>
+    <WholeProgramOptimization>true</WholeProgramOptimization>
+    <CharacterSet>MultiByte</CharacterSet>
+  </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'" Label="Configuration">
+    <ConfigurationType>Application</ConfigurationType>
+    <UseDebugLibraries>false</UseDebugLibraries>
+    <WholeProgramOptimization>true</WholeProgramOptimization>
+    <CharacterSet>MultiByte</CharacterSet>
+  </PropertyGroup>
+  <Import Project="$(VCTargetsPath)\Microsoft.Cpp.props" />
+  <ImportGroup Label="ExtensionSettings">
+    <Import Project="$(VCTargetsPath)\BuildCustomizations\CUDA 6.5.props" />
+  </ImportGroup>
+  <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
+    <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
+  </ImportGroup>
+  <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
+    <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
+  </ImportGroup>
+  <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
+    <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
+  </ImportGroup>
+  <ImportGroup Label="PropertySheets" Condition="'$(Configuration)|$(Platform)'=='Release|x64'">
+    <Import Project="$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props" Condition="exists('$(UserRootDir)\Microsoft.Cpp.$(Platform).user.props')" Label="LocalAppDataPlatform" />
+  </ImportGroup>
+  <PropertyGroup Label="UserMacros" />
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
+    <LinkIncremental>true</LinkIncremental>
+  </PropertyGroup>
+  <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
+    <LinkIncremental>true</LinkIncremental>
+  </PropertyGroup>
+  <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'">
+    <ClCompile>
+      <WarningLevel>Level3</WarningLevel>
+      <Optimization>Disabled</Optimization>
+      <PreprocessorDefinitions>WIN32;_DEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+      <AdditionalIncludeDirectories>$(SolutionDir)/shared/freeglut/include/;$(SolutionDir)/shared/glew/include/;%(AdditionalIncludeDirectories)</AdditionalIncludeDirectories>
+    </ClCompile>
+    <Link>
+      <GenerateDebugInformation>true</GenerateDebugInformation>
+      <SubSystem>Console</SubSystem>
+      <AdditionalDependencies>opengl32.lib;glut32.lib;glew32.lib;freeglut.lib;cudart.lib;%(AdditionalDependencies)</AdditionalDependencies>
+      <AdditionalLibraryDirectories>$(SolutionDir)/shared/freeglut/lib;$(SolutionDir)/shared/glew/lib;%(AdditionalLibraryDirectories)</AdditionalLibraryDirectories>
+    </Link>
+    <PostBuildEvent>
+      <Command>echo copy "$(CudaToolkitBinDir)\cudart*.dll" "$(OutDir)"
+copy "$(CudaToolkitBinDir)\cudart*.dll" "$(OutDir)"</Command>
+    </PostBuildEvent>
+  </ItemDefinitionGroup>
+  <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Debug|x64'">
+    <ClCompile>
+      <WarningLevel>Level3</WarningLevel>
+      <Optimization>Disabled</Optimization>
+      <PreprocessorDefinitions>WIN32;WIN64;_DEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+    </ClCompile>
+    <Link>
+      <GenerateDebugInformation>true</GenerateDebugInformation>
+      <SubSystem>Console</SubSystem>
+      <AdditionalDependencies>cudart.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
+    </Link>
+    <PostBuildEvent>
+      <Command>echo copy "$(CudaToolkitBinDir)\cudart*.dll" "$(OutDir)"
+copy "$(CudaToolkitBinDir)\cudart*.dll" "$(OutDir)"</Command>
+    </PostBuildEvent>
+    <CudaCompile>
+      <TargetMachinePlatform>64</TargetMachinePlatform>
+    </CudaCompile>
+  </ItemDefinitionGroup>
+  <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|Win32'">
+    <ClCompile>
+      <WarningLevel>Level3</WarningLevel>
+      <Optimization>MaxSpeed</Optimization>
+      <FunctionLevelLinking>true</FunctionLevelLinking>
+      <IntrinsicFunctions>true</IntrinsicFunctions>
+      <PreprocessorDefinitions>WIN32;NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+    </ClCompile>
+    <Link>
+      <GenerateDebugInformation>true</GenerateDebugInformation>
+      <EnableCOMDATFolding>true</EnableCOMDATFolding>
+      <OptimizeReferences>true</OptimizeReferences>
+      <SubSystem>Console</SubSystem>
+      <AdditionalDependencies>cudart.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
+    </Link>
+    <PostBuildEvent>
+      <Command>echo copy "$(CudaToolkitBinDir)\cudart*.dll" "$(OutDir)"
+copy "$(CudaToolkitBinDir)\cudart*.dll" "$(OutDir)"</Command>
+    </PostBuildEvent>
+  </ItemDefinitionGroup>
+  <ItemDefinitionGroup Condition="'$(Configuration)|$(Platform)'=='Release|x64'">
+    <ClCompile>
+      <WarningLevel>Level3</WarningLevel>
+      <Optimization>MaxSpeed</Optimization>
+      <FunctionLevelLinking>true</FunctionLevelLinking>
+      <IntrinsicFunctions>true</IntrinsicFunctions>
+      <PreprocessorDefinitions>WIN32;WIN64;NDEBUG;_CONSOLE;%(PreprocessorDefinitions)</PreprocessorDefinitions>
+    </ClCompile>
+    <Link>
+      <GenerateDebugInformation>true</GenerateDebugInformation>
+      <EnableCOMDATFolding>true</EnableCOMDATFolding>
+      <OptimizeReferences>true</OptimizeReferences>
+      <SubSystem>Console</SubSystem>
+      <AdditionalDependencies>cudart.lib;kernel32.lib;user32.lib;gdi32.lib;winspool.lib;comdlg32.lib;advapi32.lib;shell32.lib;ole32.lib;oleaut32.lib;uuid.lib;odbc32.lib;odbccp32.lib;%(AdditionalDependencies)</AdditionalDependencies>
+    </Link>
+    <PostBuildEvent>
+      <Command>echo copy "$(CudaToolkitBinDir)\cudart*.dll" "$(OutDir)"
+copy "$(CudaToolkitBinDir)\cudart*.dll" "$(OutDir)"</Command>
+    </PostBuildEvent>
+    <CudaCompile>
+      <TargetMachinePlatform>64</TargetMachinePlatform>
+    </CudaCompile>
+  </ItemDefinitionGroup>
+  <ItemGroup>
+    <CudaCompile Include="math_matrix.cu" />
+  </ItemGroup>
+  <Import Project="$(VCTargetsPath)\Microsoft.Cpp.targets" />
+  <ImportGroup Label="ExtensionTargets">
+    <Import Project="$(VCTargetsPath)\BuildCustomizations\CUDA 6.5.targets" />
+  </ImportGroup>
+</Project>
diff --git a/Part2/Matrix_math/Matrix_math/Matrix_math.vcxproj.filters b/Part2/Matrix_math/Matrix_math/Matrix_math.vcxproj.filters
@@ -0,0 +1,16 @@
+<?xml version="1.0" encoding="utf-8"?>
+<Project ToolsVersion="4.0" xmlns="http://schemas.microsoft.com/developer/msbuild/2003">
+  <ItemGroup>
+    <Filter Include="Source">
+      <UniqueIdentifier>{46dd5b81-f4ff-446d-9f33-5d8f44abded7}</UniqueIdentifier>
+    </Filter>
+    <Filter Include="Header">
+      <UniqueIdentifier>{7ea994f3-d441-413d-8588-2e4950ab67bc}</UniqueIdentifier>
+    </Filter>
+  </ItemGroup>
+  <ItemGroup>
+    <CudaCompile Include="math_matrix.cu">
+      <Filter>Source</Filter>
+    </CudaCompile>
+  </ItemGroup>
+</Project>
diff --git a/Part2/Matrix_math/Matrix_math/main.cpp b/Part2/Matrix_math/Matrix_math/main.cpp
@@ -0,0 +1,7 @@
+#include "main.h"
+
+int main(int argc, char** argv)
+{
+
+	return 0;
+}