init

init

CUDA_BasicMatrixMultiplication/BasicMatrixMultiplication/template.cu

@@ -0,0 +1,125 @@
+
+#include <wb.h>
+
+#define wbCheck(stmt)                                                     \
+  do {                                                                    \
+    cudaError_t err = stmt;                                               \
+    if (err != cudaSuccess) {                                             \
+      wbLog(ERROR, "Failed to run stmt ", #stmt);                         \
+      wbLog(ERROR, "Got CUDA error ...  ", cudaGetErrorString(err));      \
+      return -1;                                                          \
+    }                                                                     \
+  } while (0)
+
+// Compute C = A * B
+__global__ void matrixMultiply(float *A, float *B, float *C, int numARows, int numAColumns, int numBRows, int numBColumns, int numCRows, int numCColumns) {
+  int Row = blockIdx.y * blockDim.y + threadIdx.y;
+  int Col = blockIdx.x * blockDim.x + threadIdx.x;
+
+  if ((Row < numCRows) && (Col < numCColumns)) {
+	float Pvalue = 0;
+	for (int k = 0; k < numAColumns; ++k) {
+		Pvalue += A[Row * numAColumns + k] * B[k * numBColumns + Col];
+	}
+	C[Row * numCColumns + Col] = Pvalue;
+  }
+}
+
+int main(int argc, char **argv) {
+  wbArg_t args;
+  float *hostA; // The A matrix
+  float *hostB; // The B matrix
+  float *hostC; // The output C matrix
+  float *deviceA;
+  float *deviceB;
+  float *deviceC;
+  int numARows;    // number of rows in the matrix A
+  int numAColumns; // number of columns in the matrix A
+  int numBRows;    // number of rows in the matrix B
+  int numBColumns; // number of columns in the matrix B
+  int numCRows;    // number of rows in the matrix C
+  int numCColumns; // number of columns in the matrix C
+
+  hostC = NULL;
+  int sizeA, sizeB, sizeC;
+
+  args = wbArg_read(argc, argv);
+
+  wbTime_start(Generic, "Importing data and creating memory on host");
+  hostA = (float *)wbImport(wbArg_getInputFile(args, 0), &numARows,
+                            &numAColumns);
+  hostB = (float *)wbImport(wbArg_getInputFile(args, 1), &numBRows,
+                            &numBColumns);
+
+  //@@ Set numCRows and numCColumns
+  numCRows    = numARows;
+  numCColumns = numBColumns;
+
+  sizeA = numARows * numAColumns * sizeof(float);
+  sizeB = numBRows * numBColumns * sizeof(float);
+  sizeC = numCRows * numCColumns * sizeof(float);
+
+  //@@ Allocate the hostC matrix
+
+  hostC = (float *) malloc(sizeC);
+
+  wbTime_stop(Generic, "Importing data and creating memory on host");
+
+  wbLog(TRACE, "The dimensions of A are ", numARows, " x ", numAColumns);
+  wbLog(TRACE, "The dimensions of B are ", numBRows, " x ", numBColumns);
+
+  wbTime_start(GPU, "Allocating GPU memory.");
+  //@@ Allocate GPU memory here
+
+  wbCheck(cudaMalloc((void**) &deviceA, sizeA));
+  wbCheck(cudaMalloc((void**) &deviceB, sizeB));
+  wbCheck(cudaMalloc((void**) &deviceC, sizeC));
+
+  wbTime_stop(GPU, "Allocating GPU memory.");
+
+  wbTime_start(GPU, "Copying input memory to the GPU.");
+  //@@ Copy memory to the GPU here
+
+  wbCheck(cudaMemcpy(deviceA, hostA, sizeA, cudaMemcpyHostToDevice));
+  wbCheck(cudaMemcpy(deviceB, hostB, sizeB, cudaMemcpyHostToDevice));
+
+  wbTime_stop(GPU, "Copying input memory to the GPU.");
+
+  //@@ Initialize the grid and block dimensions here
+
+  // TILE_WIDTH = 16
+  dim3 dimBlock(16, 16, 1);
+  dim3 dimGrid(numBColumns / 16, numARows / 16, 1);
+
+  wbTime_start(Compute, "Performing CUDA computation");
+  //@@ Launch the GPU Kernel
+
+  matrixMultiply<<<dimGrid, dimBlock>>>(deviceA, deviceB, deviceC, numARows, numAColumns, numBRows, numBColumns, numCRows, numCColumns);
+  wbCheck(cudaDeviceSynchronize());
+
+  wbTime_stop(Compute, "Performing CUDA computation");
+
+  wbTime_start(Copy, "Copying output memory to the CPU");
+  //@@ Copy the GPU memory back to the CPU
+
+  wbCheck(cudaMemcpy(hostC, deviceC, sizeC, cudaMemcpyDeviceToHost));
+
+  wbTime_stop(Copy, "Copying output memory to the CPU");
+
+  wbTime_start(GPU, "Freeing GPU Memory");
+  //@@ Free the GPU memory
+
+  cudaFree(deviceA);
+  cudaFree(deviceB);
+  cudaFree(deviceC);
+
+  wbTime_stop(GPU, "Freeing GPU Memory");
+
+  wbSolution(args, hostC, numCRows, numCColumns);
+
+  free(hostA);
+  free(hostB);
+  free(hostC);
+
+  return 0;
+}

CUDA_BasicMatrixMultiplication/libwb/.clang-format

@@ -0,0 +1,21 @@
+---
+Language:        Cpp
+BasedOnStyle:  LLVM
+ColumnLimit:     75
+AccessModifierOffset: -2
+BreakBeforeBraces: Attach
+AlignTrailingComments: true
+AlignEscapedNewlinesLeft: false
+AlignConsecutiveAssignments: true
+AlignOperands: true
+AllowShortFunctionsOnASingleLine: false
+AllowShortIfStatementsOnASingleLine: false
+AllowShortLoopsOnASingleLine: false
+AlwaysBreakTemplateDeclarations: true
+IndentCaseLabels: true
+SpacesBeforeTrailingComments: 1
+Standard:        Cpp11
+TabWidth:        4
+UseTab:          Never
+SortIncludes: false
+...

CUDA_BasicMatrixMultiplication/libwb/.travis.yml

@@ -0,0 +1,65 @@
+# Use new trusty images, should yield newer compilers and packages
+sudo: required
+dist: precise
+language: cpp
+
+matrix:
+  include:
+    - compiler: gcc
+      addons:
+        apt:
+          sources:
+            - ubuntu-toolchain-r-test
+          packages:
+            - g++-4.9
+      env: COMPILER=g++-4.9
+    - compiler: gcc
+      addons:
+        apt:
+          sources:
+            - ubuntu-toolchain-r-test
+          packages:
+            - g++-5
+      env: COMPILER=g++-5
+    - compiler: clang
+      addons:
+        apt:
+          sources:
+            - ubuntu-toolchain-r-test
+            - llvm-toolchain-precise-3.6
+          packages:
+            - clang-3.6
+      env: COMPILER=clang++-3.6
+    - compiler: clang
+      addons:
+        apt:
+          sources:
+            - ubuntu-toolchain-r-test
+            - llvm-toolchain-precise-3.7
+          packages:
+            - clang-3.7
+      env: COMPILER=clang++-3.7
+#    - compiler: gcc
+#      addons:
+#        apt:
+#          sources:
+#            - ubuntu-toolchain-r-test
+#          packages:
+#            - g++-6
+#      env: COMPILER=g++-6
+    # - compiler: clang
+    #   addons:
+    #     apt:
+    #       sources:
+    #         - ubuntu-toolchain-r-test
+    #         - llvm-toolchain-precise-3.8
+    #       packages:
+    #         - clang-3.8
+    #   env: COMPILER=clang++-3.8
+
+before_install:
+  - sudo apt-get update -qq
+script: 
+  - $COMPILER --version
+  - make CXX=$COMPILER test
+  - ./test

CUDA_BasicMatrixMultiplication/libwb/CMakeLists.txt

@@ -0,0 +1,34 @@
+cmake_minimum_required(VERSION 2.8 FATAL_ERROR)
+
+set(CMAKE_BUILD_TYPE Release)
+set(CMAKE_EXPORT_COMPILE_COMMANDS ON)
+set(CMAKE_COLOR_MAKEFILE ON)
+set(VERBOSE_BUILD ON)
+set(CMAKE_INSTALL_RPATH_USE_LINK_PATH TRUE)
+set(CMAKE_MACOSX_RPATH TRUE)
+set_property(GLOBAL PROPERTY USE_FOLDERS ON)
+
+
+project(wb)
+
+set(current_dir "${CMAKE_CURRENT_SOURCE_DIR}")
+
+if(NOT EXISTS "${CMAKE_CURRENT_BINARY_DIR}/cotire.cmake")
+  file(DOWNLOAD "https://raw.githubusercontent.com/sakra/cotire/master/CMake/cotire.cmake" "${CMAKE_CURRENT_BINARY_DIR}/cotire.cmake")
+endif()
+
+
+
+include(${current_dir}/sources.cmake)
+
+set(WBLIB_STATIC ${WBLIB})
+set(WBLIB_SHARED lib${WBLIB})
+
+add_library(${WBLIB_STATIC} STATIC ${LIBWB_SOURCE_FILES} )
+add_library(${WBLIB_SHARED} SHARED ${LIBWB_SOURCE_FILES} )
+set_property(TARGET ${WBLIB_STATIC} PROPERTY CXX_STANDARD 11)
+set_property(TARGET ${WBLIB_SHARED} PROPERTY CXX_STANDARD 11)
+if (UNIX)
+  set_target_properties(${WBLIB_SHARED} PROPERTIES OUTPUT_NAME ${WBLIB_STATIC})
+endif (UNIX)
+set_property(TARGET ${WBLIB} PROPERTY CXX_STANDARD 11)

CUDA_BasicMatrixMultiplication/libwb/LICENSE.TXT

@@ -0,0 +1,36 @@
+Copyright (c) 2016, Abdul Dakkak All rights reserved.
+
+Developed by: 	Abdul Dakkak
+                IMPACT Group
+                University of Illinois, Urbana-Champaign
+                impact.crhc.illinois.edu
+
+Permission  is  hereby  granted,  free  of  charge,  to  any
+person  obtaining a  copy  of this  software and  associated
+documentation  files  (the  "Software"), to  deal  with  the
+Software without  restriction, including  without limitation
+the rights to use, copy, modify, merge, publish, distribute,
+sublicense,  and/or  sell copies  of  the  Software, and  to
+permit persons to  whom the Software is furnished  to do so,
+subject to the following conditions:
+
+Redistributions  of  source  code   must  retain  the  above
+copyright notice, this list  of conditions and the following
+disclaimers.
+Redistributions  in binary  form  must  reproduce the  above
+copyright notice, this list  of conditions and the following
+disclaimers  in  the  documentation and/or  other  materials
+provided with the distribution.
+Neither the  names of  Abdul Dakkak,  Impact Group,  nor the
+names of its contributors may  be used to endorse or promote
+products derived  from this Software without  specific prior
+written permission.
+THE  SOFTWARE  IS  PROVIDED  "AS IS",  WITHOUT  WARRANTY  OF
+ANY  KIND, EXPRESS  OR  IMPLIED, INCLUDING  BUT NOT  LIMITED
+TO  THE   WARRANTIES  OF  MERCHANTABILITY,  FITNESS   FOR  A
+PARTICULAR PURPOSE  AND NONINFRINGEMENT.  IN NO  EVENT SHALL
+THE  CONTRIBUTORS OR  COPYRIGHT  HOLDERS BE  LIABLE FOR  ANY
+CLAIM, DAMAGES OR  OTHER LIABILITY, WHETHER IN  AN ACTION OF
+CONTRACT,  TORT OR  OTHERWISE, ARISING  FROM, OUT  OF OR  IN
+CONNECTION WITH  THE SOFTWARE OR  THE USE OR  OTHER DEALINGS
+WITH THE SOFTWARE.

CUDA_BasicMatrixMultiplication/libwb/Makefile

@@ -0,0 +1,56 @@
+##########################################
+# Options
+##########################################
+WB_LIB_PATH=$(CURDIR)/lib
+WB_SRC_PATH=$(CURDIR)
+
+##########################################
+##########################################
+
+DEFINES=
+CXX_FLAGS=-fPIC -Wno-unused-function -x c++ -O3 -g -std=c++11 -Wall -Wno-unused-function -pedantic -I . -I $(WB_SRC_PATH) $(DEFINES)
+LIBS=-lm -lstdc++ 
+
+##########################################
+##########################################
+
+UNAME_S := $(shell uname -s)
+ifeq ($(UNAME_S),Linux)
+    LIBS += -lrt
+endif
+
+##########################################
+##########################################
+
+SOURCES := $(shell find $(WB_SRC_PATH) ! -name "*_test.cpp" -name "*.cpp")
+TESTS :=  $(shell find $(WB_SRC_PATH) -name "*_test.cpp")
+
+OBJECTS = $(SOURCES:.cpp=.o)
+TESTOBJECTS = $(TESTS:.cpp=.o)
+
+##############################################
+# OUTPUT
+##############################################
+
+.PHONY: all
+.SUFFIXES: .o .cpp
+all: libwb.so
+
+.cpp.o:
+	$(CXX) $(DEFINES) $(CXX_FLAGS) -c -o $@ $<
+
+libwb.so: $(OBJECTS)
+	mkdir -p $(WB_LIB_PATH)
+	$(CXX) -fPIC -shared -o $(WB_LIB_PATH)/$@ $(OBJECTS) $(LIBS)
+
+libwb.a: $(OBJECTS)
+	mkdir -p $(WB_LIB_PATH)
+	ar rcs -o $(WB_LIB_PATH)/$@ $(OBJECTS)
+
+test: $(TESTOBJECTS) $(OBJECTS)
+	$(CXX) -fPIC -o $@ $(TESTOBJECTS) $(OBJECTS) $(LIBS)
+
+
+clean:
+	rm -fr $(ARCH)
+	-rm -f $(EXES) *.o *~

CUDA_BasicMatrixMultiplication/libwb/README.md

@@ -0,0 +1,7 @@
+
+# libWB
+
+[![Travis Build Status](https://travis-ci.org/abduld/libwb.svg?branch=master)](https://travis-ci.org/abduld/libwb)
+[![AppVeyor status](https://ci.appveyor.com/api/projects/status/0nx5ie7gn5c0e6ai/branch/master?svg=true)](https://ci.appveyor.com/project/abduld/libwb/branch/master)
+<!-- [![Coverity Scan Build Status](https://img.shields.io/coverity/scan/8295.svg)](https://scan.coverity.com/projects/abduld-libwb
+ -->