Submission: Ziwei Zong

Project1-Part1/CMakeLists.txt

@@ -40,8 +40,6 @@ set(CORELIBS
 # Enable C++11 for host code
 set(CMAKE_CXX_STANDARD 11)
 
-list(APPEND CUDA_NVCC_FLAGS -G -g)
-
 # OSX-specific hacks/fixes
 if(${CMAKE_SYSTEM_NAME} MATCHES "Darwin")
     list(APPEND CORELIBS "-framework IOKit")

Project1-Part1/src/kernel.cu

@@ -3,6 +3,7 @@
 #include <cuda.h>
 #include <cmath>
 #include <glm/glm.hpp>
+#include <cuda_runtime.h>
 #include "utilityCore.hpp"
 #include "kernel.h"
 
@@ -28,7 +29,7 @@ void checkCUDAError(const char *msg, int line = -1) {
  *****************/
 
 /*! Block size used for CUDA kernel launch. */
-#define blockSize 128
+#define blockSize 1024
 
 /*! Mass of one "planet." */
 #define planetMass 3e8f
@@ -39,6 +40,7 @@ void checkCUDAError(const char *msg, int line = -1) {
 /*! Size of the starting area in simulation space. */
 const float scene_scale = 1e2;
 
+const int ApprStep = 1;
 
 /***********************************************
  * Kernel state (pointers are device pointers) *
@@ -172,6 +174,15 @@ void Nbody::copyPlanetsToVBO(float *vbodptr) {
 /**
  * Compute the acceleration on a body at `my_pos` due to the `N` bodies in the array `other_planets`.
  */
+
+__device__ glm::vec3 accBySingleOtherPlanet(glm::vec3 this_planet,glm::vec3 other_planet,float mass)
+{
+	float dist = glm::length(other_planet - this_planet);
+	dist *= dist;
+	float gMag = G*mass / (dist + ZERO_ABSORPTION_EPSILON);//!!! divide by zero
+	glm::vec3 g = glm::normalize(other_planet - this_planet)*gMag;
+	return g;
+}
 __device__  glm::vec3 accelerate(int N, int iSelf, glm::vec3 this_planet, const glm::vec3 *other_planets) {
     // TODO: Compute the acceleration on `my_pos` due to:
     //   * The star at the origin (with mass `starMass`)
@@ -189,8 +200,21 @@ __device__  glm::vec3 accelerate(int N, int iSelf, glm::vec3 this_planet, const
     //    * G is the universal gravitational constant (already defined for you)
     //    * M is the mass of the other object
     //    * r is the distance between this object and the other object
-
-    return glm::vec3(0.0f);
+	glm::vec3 g(0.0f);
+
+	//g by Origin Star
+	g += accBySingleOtherPlanet(this_planet,glm::vec3(0.0f),starMass);
+	//g by others
+
+	for (int i = 0; i < N; i += ApprStep)
+	{
+		if (i != iSelf) 
+		{
+			g += accBySingleOtherPlanet(this_planet, other_planets[i],planetMass);
+		}
+	}
+    //return glm::vec3(0.0f);
+	return g;
 }
 
 /**
@@ -199,6 +223,11 @@ __device__  glm::vec3 accelerate(int N, int iSelf, glm::vec3 this_planet, const
  */
 __global__ void kernUpdateAcc(int N, float dt, const glm::vec3 *pos, glm::vec3 *acc) {
     // TODO: implement updateAccArray.
+	int idx = (blockIdx.x * blockDim.x) + threadIdx.x;
+	if (idx < N)
+	{
+		acc[idx] = accelerate(N, idx, pos[idx], pos);
+	}
     // This function body runs once on each CUDA thread.
     // To avoid race conditions, each instance should only write ONE value to `acc`!
 }
@@ -209,6 +238,11 @@ __global__ void kernUpdateAcc(int N, float dt, const glm::vec3 *pos, glm::vec3 *
  */
 __global__ void kernUpdateVelPos(int N, float dt, glm::vec3 *pos, glm::vec3 *vel, const glm::vec3 *acc) {
     // TODO: implement updateVelocityPosition
+	int idx = (blockIdx.x * blockDim.x) + threadIdx.x;
+	if (idx < N){
+		vel[idx] += acc[idx] * dt;
+		pos[idx] += vel[idx] * dt;
+	}
 }
 
 /**
@@ -217,6 +251,28 @@ __global__ void kernUpdateVelPos(int N, float dt, glm::vec3 *pos, glm::vec3 *vel
 void Nbody::stepSimulation(float dt) {
     // TODO: Using the CUDA kernels you wrote above, write a function that
     // calls the kernels to perform a full simulation step.
+	dim3 fullBlocksPerGrid((numObjects + blockSize - 1) / blockSize);
+
+	cudaEvent_t start, stop;
+	float elapsedTime;
+	cudaEventCreate(&start);
+	cudaEventCreate(&stop);
+	cudaEventRecord(start,0);
+
+	kernUpdateAcc <<< fullBlocksPerGrid, threadsPerBlock >>>(numObjects, dt, dev_pos, dev_acc);
+
+	cudaEventRecord(stop,0);
+	cudaEventSynchronize(stop);
+	cudaEventElapsedTime(&elapsedTime, start, stop);
+	printf( "\n1 -- kernUpdateAcc Time = %.2fms",elapsedTime );
+
+
+	cudaEventRecord(start, 0);
+	kernUpdateVelPos <<< fullBlocksPerGrid, threadsPerBlock >> > (numObjects, dt, dev_pos, dev_vel, dev_acc);
+	cudaEventRecord(stop, 0);
+	cudaEventSynchronize(stop);
+	cudaEventElapsedTime(&elapsedTime, start, stop);
+	printf("\n2 -- kernUpdateVelPos Time = %.4fms", elapsedTime);
 }
 
 void Nbody::endSimulation()

Project1-Part1/src/main.cpp

@@ -14,7 +14,7 @@
 
 #define VISUALIZE 1
 
-const int N_FOR_VIS = 5000;
+const int N_FOR_VIS = 10000;
 const float DT = 0.2f;
 
 /**
@@ -238,9 +238,9 @@ void mainLoop() {
 
         glUseProgram(0);
         glBindVertexArray(0);
-#endif
 
         glfwSwapBuffers(window);
+#endif
     }
     glfwDestroyWindow(window);
     glfwTerminate();

Project1-Part1/src/utilityCore.hpp

@@ -24,7 +24,7 @@
 #define E                           2.7182818284590452353602874713526624977572
 #define G                           6.67384e-11
 #define EPSILON                     .000000001
-#define ZERO_ABSORPTION_EPSILON     0.00001
+#define ZERO_ABSORPTION_EPSILON     0.01
 
 namespace utilityCore {
 extern float clamp(float f, float min, float max);

Project1-Part2/CMakeLists.txt

@@ -0,0 +1,83 @@
+cmake_minimum_required(VERSION 3.0)
+
+project(cis565_nbody)
+
+set(CMAKE_MODULE_PATH "${CMAKE_SOURCE_DIR}/cmake" ${CMAKE_MODULE_PATH})
+
+# Set up include and lib paths
+set(EXTERNAL "external")
+include_directories("${EXTERNAL}/include")
+include_directories("${EXTERNAL}/src")
+if(${CMAKE_SYSTEM_NAME} MATCHES "Darwin")
+    set(EXTERNAL_LIB_PATH "${EXTERNAL}/lib/osx")
+elseif(${CMAKE_SYSTEM_NAME} MATCHES "Linux")
+    set(EXTERNAL_LIB_PATH "${EXTERNAL}/lib/linux" "/usr/lib64")
+elseif(WIN32)
+    set(EXTERNAL_LIB_PATH "${EXTERNAL}/lib/win")
+endif()
+link_directories(${EXTERNAL_LIB_PATH})
+list(APPEND CMAKE_LIBRARY_PATH "${EXTERNAL_LIB_PATH}")
+
+# Find up and set up core dependency libs
+
+set(GLFW_INCLUDE_DIR "${EXTERNAL}/include")
+set(GLFW_LIBRARY_DIR "${CMAKE_LIBRARY_PATH}")
+find_library(GLFW_LIBRARY "glfw3" HINTS "${GLFW_LIBRARY_DIR}")
+
+set(GLEW_INCLUDE_DIR "${EXTERNAL}/include")
+set(GLEW_LIBRARY_DIR "${CMAKE_LIBRARY_PATH}")
+add_definitions(-DGLEW_STATIC)
+find_package(GLEW)
+
+find_package(OpenGL)
+
+set(CORELIBS
+    "${GLFW_LIBRARY}"
+    "${OPENGL_LIBRARY}"
+    "${GLEW_LIBRARY}"
+    )
+
+# Enable C++11 for host code
+set(CMAKE_CXX_STANDARD 11)
+
+# OSX-specific hacks/fixes
+if(${CMAKE_SYSTEM_NAME} MATCHES "Darwin")
+    list(APPEND CORELIBS "-framework IOKit")
+    list(APPEND CORELIBS "-framework Cocoa")
+    list(APPEND CORELIBS "-framework CoreVideo")
+endif()
+
+# Linux-specific hacks/fixes
+if(${CMAKE_SYSTEM_NAME} MATCHES "Linux")
+    list(APPEND CMAKE_EXE_LINKER_FLAGS "-lX11 -lXxf86vm -lXrandr -lpthread -lXi")
+endif()
+
+# Crucial magic for CUDA linking
+find_package(Threads REQUIRED)
+find_package(CUDA REQUIRED)
+
+set(CUDA_ATTACH_VS_BUILD_RULE_TO_CUDA_FILE ON)
+set(CUDA_SEPARABLE_COMPILATION ON)
+
+if(${CMAKE_SYSTEM_NAME} MATCHES "Darwin")
+    set(CUDA_PROPAGATE_HOST_FLAGS OFF)
+endif()
+
+add_subdirectory(src)
+
+cuda_add_executable(${CMAKE_PROJECT_NAME}
+    "src/main.cpp"
+    )
+
+target_link_libraries(${CMAKE_PROJECT_NAME}
+    src
+    ${CORELIBS}
+    )
+
+add_custom_command(
+    TARGET ${CMAKE_PROJECT_NAME}
+    POST_BUILD
+    COMMAND ${CMAKE_COMMAND} -E copy_directory
+        ${CMAKE_SOURCE_DIR}/shaders
+        ${CMAKE_BINARY_DIR}/shaders
+    )

Project1-Part2/GNUmakefile

@@ -0,0 +1,31 @@
+CMAKE_ALT1 := /usr/local/bin/cmake
+CMAKE_ALT2 := /Applications/CMake.app/Contents/bin/cmake
+CMAKE := $(shell \
+	which cmake 2>/dev/null || \
+	([ -e ${CMAKE_ALT1} ] && echo "${CMAKE_ALT1}") || \
+	([ -e ${CMAKE_ALT2} ] && echo "${CMAKE_ALT2}") \
+	)
+
+all: RelWithDebugInfo
+
+
+Debug: build
+	(cd build && ${CMAKE} -DCMAKE_BUILD_TYPE=$@ .. && make)
+
+MinSizeRel: build
+	(cd build && ${CMAKE} -DCMAKE_BUILD_TYPE=$@ .. && make)
+
+Release: build
+	(cd build && ${CMAKE} -DCMAKE_BUILD_TYPE=$@ .. && make)
+
+RelWithDebugInfo: build
+	(cd build && ${CMAKE} -DCMAKE_BUILD_TYPE=$@ .. && make)
+
+
+build:
+	(mkdir -p build && cd build)
+
+clean:
+	((cd build && make clean) 2>&- || true)
+
+.PHONY: all Debug MinSizeRel Release RelWithDebugInfo clean