Commit to make the branch compile

CMakeLists.txt

@@ -21,30 +21,30 @@ if(CUDA_FOUND)
   # to make sure more people can easily run class code without knowing 
   # about this compiler argument
   set(CUDA_NVCC_FLAGS "
-  -ccbin /usr/bin/clang; 
+  -ccbin /usr/bin/gcc; 
   -gencode;arch=compute_30,code=sm_30;  
   -gencode;arch=compute_35,code=sm_35;
-  -gencode;arch=compute_35,code=compute_35;
-  -gencode;arch=compute_20,code=sm_20; 
-  -gencode;arch=compute_11,code=sm_11; 
-  -gencode;arch=compute_12,code=sm_12;
-  -gencode;arch=compute_13,code=sm_13;")
+  -gencode;arch=compute_35,code=compute_35;")
+  #-gencode;arch=compute_20,code=sm_20; 
+  #-gencode;arch=compute_11,code=sm_11; 
+  #-gencode;arch=compute_12,code=sm_12;
+  #-gencode;arch=compute_13,code=sm_13;")
 
   # add -Wextra compiler flag for gcc compilations
   if (UNIX)
     set(CUDA_NVCC_FLAGS ${CUDA_NVCC_FLAGS} "-Xcompiler -Wextra")
-    set(CMAKE_CXX_FLAGS "-stdlib=libstdc++")
+	#set(CMAKE_CXX_FLAGS "-stdlib=libstdc++")
   endif (UNIX)
 
   # add debugging to CUDA NVCC flags.  For NVidia's NSight tools.
   set(CUDA_NVCC_FLAGS_DEBUG ${CUDA_NVCC_FLAGS_DEBUG} "-G")
 
-  add_subdirectory (HW1)
-  add_subdirectory (HW2)
-  add_subdirectory (HW3)
-  add_subdirectory (HW4)
-  add_subdirectory (HW5)
-  add_subdirectory (HW6)
+  add_subdirectory ("Problem Sets/Problem Set 1")
+  add_subdirectory ("Problem Sets/Problem Set 2")
+  add_subdirectory ("Problem Sets/Problem Set 3")
+  add_subdirectory ("Problem Sets/Problem Set 4")
+  add_subdirectory ("Problem Sets/Problem Set 5")
+  add_subdirectory ("Problem Sets/Problem Set 6")
 else(CUDA_FOUND)
   message("CUDA is not installed on this system.")
 endif()

Problem Sets/Problem Set 1/Makefile

@@ -1,54 +1,240 @@
-NVCC=nvcc
+# CMAKE generated file: DO NOT EDIT!
+# Generated by "Unix Makefiles" Generator, CMake Version 3.5
 
-###################################
-# These are the default install   #
-# locations on most linux distros #
-###################################
+# Default target executed when no arguments are given to make.
+default_target: all
 
-OPENCV_LIBPATH=/usr/lib
-OPENCV_INCLUDEPATH=/usr/include
+.PHONY : default_target
 
-###################################################
-# On Macs the default install locations are below #
-###################################################
+# Allow only one "make -f Makefile2" at a time, but pass parallelism.
+.NOTPARALLEL:
 
-#OPENCV_LIBPATH=/usr/local/lib
-#OPENCV_INCLUDEPATH=/usr/local/include
 
-# or if using MacPorts
+#=============================================================================
+# Special targets provided by cmake.
 
-#OPENCV_LIBPATH=/opt/local/lib
-#OPENCV_INCLUDEPATH=/opt/local/include
+# Disable implicit rules so canonical targets will work.
+.SUFFIXES:
 
-OPENCV_LIBS=-lopencv_core -lopencv_imgproc -lopencv_highgui
 
-CUDA_INCLUDEPATH=/usr/local/cuda-5.0/include
+# Remove some rules from gmake that .SUFFIXES does not remove.
+SUFFIXES =
 
-######################################################
-# On Macs the default install locations are below    #
-# ####################################################
+.SUFFIXES: .hpux_make_needs_suffix_list
 
-#CUDA_INCLUDEPATH=/usr/local/cuda/include
-#CUDA_LIBPATH=/usr/local/cuda/lib
 
-NVCC_OPTS=-O3 -arch=sm_20 -Xcompiler -Wall -Xcompiler -Wextra -m64
+# Suppress display of executed commands.
+$(VERBOSE).SILENT:
 
-GCC_OPTS=-O3 -Wall -Wextra -m64
 
-student: main.o student_func.o compare.o reference_calc.o Makefile
-	$(NVCC) -o HW1 main.o student_func.o compare.o reference_calc.o -L $(OPENCV_LIBPATH) $(OPENCV_LIBS) $(NVCC_OPTS)
+# A target that is always out of date.
+cmake_force:
 
-main.o: main.cpp timer.h utils.h reference_calc.cpp compare.cpp HW1.cpp
-	g++ -c main.cpp $(GCC_OPTS) -I $(CUDA_INCLUDEPATH) -I $(OPENCV_INCLUDEPATH)
+.PHONY : cmake_force
 
-student_func.o: student_func.cu utils.h
-	nvcc -c student_func.cu $(NVCC_OPTS)
+#=============================================================================
+# Set environment variables for the build.
 
-compare.o: compare.cpp compare.h
-	g++ -c compare.cpp -I $(OPENCV_INCLUDEPATH) $(GCC_OPTS) -I $(CUDA_INCLUDEPATH)
+# The shell in which to execute make rules.
+SHELL = /bin/sh
 
-reference_calc.o: reference_calc.cpp reference_calc.h
-	g++ -c reference_calc.cpp -I $(OPENCV_INCLUDEPATH) $(GCC_OPTS) -I $(CUDA_INCLUDEPATH)
+# The CMake executable.
+CMAKE_COMMAND = /usr/bin/cmake
 
+# The command to remove a file.
+RM = /usr/bin/cmake -E remove -f
+
+# Escaping for special characters.
+EQUALS = =
+
+# The top-level source directory on which CMake was run.
+CMAKE_SOURCE_DIR = /home/wuqiongs/cs344
+
+# The top-level build directory on which CMake was run.
+CMAKE_BINARY_DIR = /home/wuqiongs/cs344
+
+#=============================================================================
+# Targets provided globally by CMake.
+
+# Special rule for the target edit_cache
+edit_cache:
+	@$(CMAKE_COMMAND) -E cmake_echo_color --switch=$(COLOR) --cyan "No interactive CMake dialog available..."
+	/usr/bin/cmake -E echo No\ interactive\ CMake\ dialog\ available.
+.PHONY : edit_cache
+
+# Special rule for the target edit_cache
+edit_cache/fast: edit_cache
+
+.PHONY : edit_cache/fast
+
+# Special rule for the target rebuild_cache
+rebuild_cache:
+	@$(CMAKE_COMMAND) -E cmake_echo_color --switch=$(COLOR) --cyan "Running CMake to regenerate build system..."
+	/usr/bin/cmake -H$(CMAKE_SOURCE_DIR) -B$(CMAKE_BINARY_DIR)
+.PHONY : rebuild_cache
+
+# Special rule for the target rebuild_cache
+rebuild_cache/fast: rebuild_cache
+
+.PHONY : rebuild_cache/fast
+
+# The main all target
+all: cmake_check_build_system
+	cd /home/wuqiongs/cs344 && $(CMAKE_COMMAND) -E cmake_progress_start /home/wuqiongs/cs344/CMakeFiles "/home/wuqiongs/cs344/Problem Sets/Problem Set 1/CMakeFiles/progress.marks"
+	cd /home/wuqiongs/cs344 && $(MAKE) -f CMakeFiles/Makefile2 "Problem Sets/Problem Set 1/all"
+	$(CMAKE_COMMAND) -E cmake_progress_start /home/wuqiongs/cs344/CMakeFiles 0
+.PHONY : all
+
+# The main clean target
 clean:
-	rm -f *.o *.png hw
+	cd /home/wuqiongs/cs344 && $(MAKE) -f CMakeFiles/Makefile2 "Problem Sets/Problem Set 1/clean"
+.PHONY : clean
+
+# The main clean target
+clean/fast: clean
+
+.PHONY : clean/fast
+
+# Prepare targets for installation.
+preinstall: all
+	cd /home/wuqiongs/cs344 && $(MAKE) -f CMakeFiles/Makefile2 "Problem Sets/Problem Set 1/preinstall"
+.PHONY : preinstall
+
+# Prepare targets for installation.
+preinstall/fast:
+	cd /home/wuqiongs/cs344 && $(MAKE) -f CMakeFiles/Makefile2 "Problem Sets/Problem Set 1/preinstall"
+.PHONY : preinstall/fast
+
+# clear depends
+depend:
+	cd /home/wuqiongs/cs344 && $(CMAKE_COMMAND) -H$(CMAKE_SOURCE_DIR) -B$(CMAKE_BINARY_DIR) --check-build-system CMakeFiles/Makefile.cmake 1
+.PHONY : depend
+
+# Convenience name for target.
+Problem\ Sets/Problem\ Set\ 1/CMakeFiles/HW1.dir/rule:
+	cd /home/wuqiongs/cs344 && $(MAKE) -f CMakeFiles/Makefile2 "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/rule"
+.PHONY : Problem\ Sets/Problem\ Set\ 1/CMakeFiles/HW1.dir/rule
+
+# Convenience name for target.
+HW1: Problem\ Sets/Problem\ Set\ 1/CMakeFiles/HW1.dir/rule
+
+.PHONY : HW1
+
+# fast build rule for target.
+HW1/fast:
+	cd /home/wuqiongs/cs344 && $(MAKE) -f "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/build.make" "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/build"
+.PHONY : HW1/fast
+
+compare.o: compare.cpp.o
+
+.PHONY : compare.o
+
+# target to build an object file
+compare.cpp.o:
+	cd /home/wuqiongs/cs344 && $(MAKE) -f "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/build.make" "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/compare.cpp.o"
+.PHONY : compare.cpp.o
+
+compare.i: compare.cpp.i
+
+.PHONY : compare.i
+
+# target to preprocess a source file
+compare.cpp.i:
+	cd /home/wuqiongs/cs344 && $(MAKE) -f "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/build.make" "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/compare.cpp.i"
+.PHONY : compare.cpp.i
+
+compare.s: compare.cpp.s
+
+.PHONY : compare.s
+
+# target to generate assembly for a file
+compare.cpp.s:
+	cd /home/wuqiongs/cs344 && $(MAKE) -f "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/build.make" "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/compare.cpp.s"
+.PHONY : compare.cpp.s
+
+main.o: main.cpp.o
+
+.PHONY : main.o
+
+# target to build an object file
+main.cpp.o:
+	cd /home/wuqiongs/cs344 && $(MAKE) -f "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/build.make" "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/main.cpp.o"
+.PHONY : main.cpp.o
+
+main.i: main.cpp.i
+
+.PHONY : main.i
+
+# target to preprocess a source file
+main.cpp.i:
+	cd /home/wuqiongs/cs344 && $(MAKE) -f "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/build.make" "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/main.cpp.i"
+.PHONY : main.cpp.i
+
+main.s: main.cpp.s
+
+.PHONY : main.s
+
+# target to generate assembly for a file
+main.cpp.s:
+	cd /home/wuqiongs/cs344 && $(MAKE) -f "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/build.make" "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/main.cpp.s"
+.PHONY : main.cpp.s
+
+reference_calc.o: reference_calc.cpp.o
+
+.PHONY : reference_calc.o
+
+# target to build an object file
+reference_calc.cpp.o:
+	cd /home/wuqiongs/cs344 && $(MAKE) -f "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/build.make" "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/reference_calc.cpp.o"
+.PHONY : reference_calc.cpp.o
+
+reference_calc.i: reference_calc.cpp.i
+
+.PHONY : reference_calc.i
+
+# target to preprocess a source file
+reference_calc.cpp.i:
+	cd /home/wuqiongs/cs344 && $(MAKE) -f "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/build.make" "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/reference_calc.cpp.i"
+.PHONY : reference_calc.cpp.i
+
+reference_calc.s: reference_calc.cpp.s
+
+.PHONY : reference_calc.s
+
+# target to generate assembly for a file
+reference_calc.cpp.s:
+	cd /home/wuqiongs/cs344 && $(MAKE) -f "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/build.make" "Problem Sets/Problem Set 1/CMakeFiles/HW1.dir/reference_calc.cpp.s"
+.PHONY : reference_calc.cpp.s
+
+# Help Target
+help:
+	@echo "The following are some of the valid targets for this Makefile:"
+	@echo "... all (the default if no target is provided)"
+	@echo "... clean"
+	@echo "... depend"
+	@echo "... edit_cache"
+	@echo "... HW1"
+	@echo "... rebuild_cache"
+	@echo "... compare.o"
+	@echo "... compare.i"
+	@echo "... compare.s"
+	@echo "... main.o"
+	@echo "... main.i"
+	@echo "... main.s"
+	@echo "... reference_calc.o"
+	@echo "... reference_calc.i"
+	@echo "... reference_calc.s"
+.PHONY : help
+
+
+
+#=============================================================================
+# Special targets to cleanup operation of make.
+
+# Special rule to run CMake to check the build system integrity.
+# No rule that depends on this can have commands that come from listfiles
+# because they might be regenerated.
+cmake_check_build_system:
+	cd /home/wuqiongs/cs344 && $(CMAKE_COMMAND) -H$(CMAKE_SOURCE_DIR) -B$(CMAKE_BINARY_DIR) --check-build-system CMakeFiles/Makefile.cmake 0
+.PHONY : cmake_check_build_system
+

Problem Sets/Problem Set 1/student_func.cu

@@ -33,12 +33,18 @@
 
 #include "utils.h"
 
+// This is used for Quadro P4000. 
+// In total, a 2D block has 32 * 32 = 1024 threads, 
+// the maximum on a P4000.
+// Somehow 16 x 16 block size performs better than
+// 32 x 32 block size. 
+int BLOCK_DIM = 16; 
+
 __global__
 void rgba_to_greyscale(const uchar4* const rgbaImage,
                        unsigned char* const greyImage,
                        int numRows, int numCols)
 {
-  //TODO
   //Fill in the kernel to convert from color to greyscale
   //the mapping from components of a uchar4 to RGBA is:
   // .x -> R ; .y -> G ; .z -> B ; .w -> A
@@ -50,15 +56,31 @@ void rgba_to_greyscale(const uchar4* const rgbaImage,
   //First create a mapping from the 2D block and grid locations
   //to an absolute 2D location in the image, then use that to
   //calculate a 1D offset
+  int x = blockDim.x * blockIdx.x + threadIdx.x; 
+  int y = blockDim.y * blockIdx.y + threadIdx.y; 
+
+  if (x < numCols && y < numRows) {
+    int offset = y * numCols + x; 
+    uchar4 rgba = rgbaImage[offset];
+
+    float channelSum = 0.299f * rgba.x + 0.587f * rgba.y + 0.114f * rgba.z; 
+    greyImage[offset] = channelSum; 
+  }
 }
 
 void your_rgba_to_greyscale(const uchar4 * const h_rgbaImage, uchar4 * const d_rgbaImage,
                             unsigned char* const d_greyImage, size_t numRows, size_t numCols)
 {
   //You must fill in the correct sizes for the blockSize and gridSize
   //currently only one block with one thread is being launched
-  const dim3 blockSize(1, 1, 1);  //TODO
-  const dim3 gridSize( 1, 1, 1);  //TODO
+  const dim3 blockSize(BLOCK_DIM, BLOCK_DIM, 1);  
+
+  //Compute the sizes of the grid
+  int gridRow = (numRows + BLOCK_DIM - 1) / BLOCK_DIM; 
+  int gridCol = (numCols + BLOCK_DIM - 1) / BLOCK_DIM; 
+
+  // Crucial! Col is for x index, Row is for y index! 
+  const dim3 gridSize(gridCol, gridRow, 1);  
   rgba_to_greyscale<<<gridSize, blockSize>>>(d_rgbaImage, d_greyImage, numRows, numCols);
 
   cudaDeviceSynchronize(); checkCudaErrors(cudaGetLastError());