Initial commit

.clang-format

+---
+Language:        Cpp
+# BasedOnStyle:  Google
+AccessModifierOffset: -1
+AlignAfterOpenBracket: Align
+AlignConsecutiveAssignments: false
+AlignConsecutiveDeclarations: false
+AlignEscapedNewlines: Left
+AlignOperands:   true
+AlignTrailingComments: true
+AllowAllParametersOfDeclarationOnNextLine: true
+AllowShortBlocksOnASingleLine: false
+AllowShortCaseLabelsOnASingleLine: false
+AllowShortFunctionsOnASingleLine: All
+AllowShortIfStatementsOnASingleLine: true
+AllowShortLoopsOnASingleLine: true
+AlwaysBreakAfterDefinitionReturnType: None
+AlwaysBreakAfterReturnType: None
+AlwaysBreakBeforeMultilineStrings: true
+AlwaysBreakTemplateDeclarations: true
+BinPackArguments: true
+BinPackParameters: true
+BraceWrapping:   
+  AfterClass:      false
+  AfterControlStatement: false
+  AfterEnum:       false
+  AfterFunction:   false
+  AfterNamespace:  false
+  AfterObjCDeclaration: false
+  AfterStruct:     false
+  AfterUnion:      false
+  AfterExternBlock: false
+  BeforeCatch:     false
+  BeforeElse:      false
+  IndentBraces:    false
+  SplitEmptyFunction: true
+  SplitEmptyRecord: true
+  SplitEmptyNamespace: true
+BreakBeforeBinaryOperators: None
+BreakBeforeBraces: Attach
+BreakBeforeInheritanceComma: false
+BreakBeforeTernaryOperators: true
+BreakConstructorInitializersBeforeComma: false
+BreakConstructorInitializers: BeforeColon
+BreakAfterJavaFieldAnnotations: false
+BreakStringLiterals: true
+ColumnLimit:     80
+CommentPragmas:  '^ IWYU pragma:'
+CompactNamespaces: false
+ConstructorInitializerAllOnOneLineOrOnePerLine: true
+ConstructorInitializerIndentWidth: 4
+ContinuationIndentWidth: 4
+Cpp11BracedListStyle: true
+DerivePointerAlignment: true
+DisableFormat:   false
+ExperimentalAutoDetectBinPacking: false
+FixNamespaceComments: true
+ForEachMacros:   
+  - foreach
+  - Q_FOREACH
+  - BOOST_FOREACH
+IncludeBlocks:   Preserve
+IncludeCategories: 
+  - Regex:           '^<ext/.*\.h>'
+    Priority:        2
+  - Regex:           '^<.*\.h>'
+    Priority:        1
+  - Regex:           '^<.*'
+    Priority:        2
+  - Regex:           '.*'
+    Priority:        3
+IncludeIsMainRegex: '([-_](test|unittest))?$'
+IndentCaseLabels: true
+IndentPPDirectives: None
+IndentWidth:     4
+IndentWrappedFunctionNames: false
+JavaScriptQuotes: Leave
+JavaScriptWrapImports: true
+KeepEmptyLinesAtTheStartOfBlocks: false
+MacroBlockBegin: ''
+MacroBlockEnd:   ''
+MaxEmptyLinesToKeep: 1
+NamespaceIndentation: None
+ObjCBlockIndentWidth: 2
+ObjCSpaceAfterProperty: false
+ObjCSpaceBeforeProtocolList: false
+PenaltyBreakAssignment: 2
+PenaltyBreakBeforeFirstCallParameter: 1
+PenaltyBreakComment: 300
+PenaltyBreakFirstLessLess: 120
+PenaltyBreakString: 1000
+PenaltyExcessCharacter: 1000000
+PenaltyReturnTypeOnItsOwnLine: 200
+PointerAlignment: Left
+RawStringFormats: 
+  - Delimiter:       pb
+    Language:        TextProto
+    BasedOnStyle:    google
+ReflowComments:  true
+SortIncludes:    true
+SortUsingDeclarations: true
+SpaceAfterCStyleCast: false
+SpaceAfterTemplateKeyword: true
+SpaceBeforeAssignmentOperators: true
+SpaceBeforeParens: ControlStatements
+SpaceInEmptyParentheses: false
+SpacesBeforeTrailingComments: 2
+SpacesInAngles:  false
+SpacesInContainerLiterals: true
+SpacesInCStyleCastParentheses: false
+SpacesInParentheses: false
+SpacesInSquareBrackets: false
+Standard:        Auto
+TabWidth:        8
+UseTab:          Never
+...
+

.gitignore

+.vscode
+perf.data*
+build*
+

.gitlab-ci.yml

+stages:
+    - test
+
+test:
+    stage: test
+    script: 
+        - mkdir -p build/debug
+        - cd build/debug
+        - cmake -DCMAKE_BUILD_TYPE=Debug ../..
+        - make -j8
+        - ./test_all
+    cache:
+        key: "$CI_JOB_STAGE-$CI_COMMIT_REF_NAME"
+        paths:
+            - build/debug
+        policy: pull-push
+    tags:
+        - "fdedi"
+
+clear_cache_test:
+    stage: test
+    script:
+        - rm -rf ./build/*
+    cache:
+        key: "$CI_JOB_STAGE-$CI_COMMIT_REF_NAME"
+        paths:
+            - build/
+        policy: pull-push
+    when: manual
+    allow_failure: true

CMakeLists.txt

+cmake_minimum_required(VERSION 3.5)
+project(tpchjoinoptimized C CXX)
+
+set(CMAKE_C_COMPILER gcc)
+set(CMAKE_CXX_COMPILER g++)
+set(CMAKE_CXX_STANDARD 17)
+
+# Compiler flags for the different targets
+set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=address -fPIC -Wall -Wextra -fno-omit-frame-pointer -march=native -Wno-unknown-pragmas")
+set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -fsanitize=address -fPIC -Wall -Wextra -fno-omit-frame-pointer -march=native")
+
+find_package(Threads)
+
+include(lib/gtest.cmake)
+
+### Library for join query
+add_library(sum src/sum.cpp)
+target_include_directories(sum PUBLIC
+    $<BUILD_INTERFACE:${CMAKE_CURRENT_SOURCE_DIR}/include>
+    $<INSTALL_INTERFACE:include>
+    PRIVATE src)
+
+### Tests
+add_executable(test_all test/test_sum.cpp)
+target_link_libraries(test_all sum gtest gtest_main pthread)
+
+### main executable
+add_executable(main src/main.cpp)
+target_link_libraries(main sum)

README.md

+# FDE Exercise 3.4
+
+## Task
+
+Implement the missing code fragments in src/main.cpp and src/sum.cpp.
+See the corresponding header include/sum.hpp for documentation.
+You can use the test provided in test/test_sum.cpp to check if your implementation works
+correctly.
+
+## Build
+A configuration file is provided to build this project with CMake.
+This allows you to build the project in the terminal but also
+provides the option to use Jetbrains CLion or Microsoft Visual Studio
+and other IDEs.
+
+Building from Terminal:
+Start in the project directory.
+```
+mkdir -p build/debug
+cd build/debug
+cmake -DCMAKE_BUILD_TYPE=Debug ../..
+make
+```
+
+This creates the binaries test_all and main.
+
+Make sure your builds are not failing! <br/>
+*Left Sidebar > CI /CD > Pipelines*

include/sum.hpp

+#pragma once
+
+#include <cassert>
+#include <string>
+#include <string_view>
+
+// pattern : the character to search broadcasted into a 64bit integer
+// begin : points somewhere into the partition
+// len : remaining length of the partition
+// return : the position of the first matching character or -1 otherwise
+ssize_t find_first(uint64_t pattern, const char* begin, size_t len);
+
+// begin : points somewhere into the partition
+// len : remaining length of the partition
+// return : 64-bit integer representation of the provided numeric
+int64_t read_numeric(const char* begin, size_t len);
+
+// data_start : pointer to the first byte of the file
+// partition_start: start of partition
+// partition_size_hint : size of partition in bytes (must be corrected to end on
+// a newline character).
+// result : pointer to the variable where the result should be stored in
+template <bool first_partition = false>
+void sum_extendedprice(const char* data_start, const char* partition_start,
+                       size_t partition_size_hint, int64_t* result) {
+    int64_t sum = 0;
+
+    // correct partition size
+    if (!first_partition) {
+        size_t offset = 0;
+        const size_t max_offset = partition_start - data_start;
+        while (offset < partition_size_hint) {
+            if (offset > max_offset) {
+                return;
+            } else if (*(partition_start - offset) == '\n') {
+                break;
+            } else {
+                offset++;
+            }
+        }
+        partition_start -= offset;
+        partition_size_hint += offset;
+    }
+
+    constexpr uint64_t bar_pattern = 0x7C7C7C7C7C7C7C7Cull;
+    constexpr uint64_t newline_pattern = 0x0A0A0A0A0A0A0A0Aull;
+
+    size_t i = 0;
+    int64_t bar_cnt = 0;
+    // for each line
+    while (i < partition_size_hint) {
+        auto pos = find_first(bar_pattern, partition_start + i,
+                              partition_size_hint - i);
+        bar_cnt += (pos >= 0);
+        if (bar_cnt == 5) {
+            auto bar_pos = i + pos + 1;
+            auto len = find_first(bar_pattern, partition_start + bar_pos,
+                                  partition_size_hint - bar_pos);
+            assert(len >= 1);
+            int64_t extendedprice =
+                read_numeric(partition_start + bar_pos, len);
+            sum += extendedprice;
+            bar_cnt = 0;
+
+            // skip to end of line
+            i = bar_pos + len + 1;
+            auto newline_pos = find_first(newline_pattern, partition_start + i,
+                                          partition_size_hint - i);
+            if (newline_pos < 0 || (newline_pos + i) > partition_size_hint) {
+                // undo
+                sum -= extendedprice;
+                break;
+            }
+            i += newline_pos + 1;
+        } else {
+            i += pos + 1;
+        }
+    }
+
+    *result = sum;
+}

include/util.hpp

+#include <string>
+//---------------------------------------------------------------------------
+inline std::string getDir(const std::string &file)
+/// Returns path to file in 'file'
+{
+    size_t found = file.find_last_of("/\\");
+    return (file.substr(0, found));
+}
+//---------------------------------------------------------------------------

lib/benchmark.cmake

+include(ExternalProject)
+find_package(Git REQUIRED)
+
+# Get benchmark
+ExternalProject_Add(
+    benchmark_src
+    PREFIX "vendor/benchmark"
+    GIT_REPOSITORY "https://github.com/google/benchmark.git"
+    GIT_TAG v1.2.0
+    TIMEOUT 10
+    CMAKE_ARGS
+        -DCMAKE_INSTALL_PREFIX=${CMAKE_BINARY_DIR}/vendor/benchmark
+        -DCMAKE_C_COMPILER=${CMAKE_C_COMPILER}
+        -DCMAKE_CXX_COMPILER=${CMAKE_CXX_COMPILER}
+        -DCMAKE_CXX_FLAGS=${CMAKE_CXX_FLAGS}
+        -DCMAKE_BUILD_TYPE:STRING=${CMAKE_BUILD_TYPE}
+    UPDATE_COMMAND ""
+)
+
+# Prepare benchmark
+ExternalProject_Get_Property(benchmark_src install_dir)
+set(BENCHMARK_INCLUDE_DIR ${install_dir}/include)
+set(BENCHMARK_LIBRARY_PATH ${install_dir}/lib/libbenchmark.a)
+file(MAKE_DIRECTORY ${BENCHMARK_INCLUDE_DIR})
+add_library(benchmark STATIC IMPORTED)
+set_property(TARGET benchmark PROPERTY IMPORTED_LOCATION ${BENCHMARK_LIBRARY_PATH})
+set_property(TARGET benchmark APPEND PROPERTY INTERFACE_INCLUDE_DIRECTORIES ${BENCHMARK_INCLUDE_DIR})
+
+# Dependencies
+add_dependencies(benchmark benchmark_src)

lib/gtest.cmake

+include(ExternalProject)
+find_package(Git REQUIRED)
+find_package(Threads REQUIRED)
+
+# Get googletest
+ExternalProject_Add(
+    googletest
+    PREFIX "vendor/gtm"
+    GIT_REPOSITORY "https://github.com/google/googletest.git"
+    GIT_TAG release-1.8.0
+    TIMEOUT 10
+    CONFIGURE_COMMAND ""
+    BUILD_COMMAND ""
+    INSTALL_COMMAND ""
+    UPDATE_COMMAND ""
+)
+
+# Build gtest
+ExternalProject_Add(
+    gtest_src
+    PREFIX "vendor/gtm"
+    SOURCE_DIR "vendor/gtm/src/googletest/googletest"
+    INSTALL_DIR "vendor/gtm/gtest"
+    CMAKE_ARGS
+        -DCMAKE_INSTALL_PREFIX=${CMAKE_BINARY_DIR}/vendor/gtm/gtest
+        -DCMAKE_C_COMPILER=${CMAKE_C_COMPILER}
+        -DCMAKE_CXX_COMPILER=${CMAKE_CXX_COMPILER}
+        -DCMAKE_CXX_FLAGS=${CMAKE_CXX_FLAGS}
+    DOWNLOAD_COMMAND ""
+    UPDATE_COMMAND ""
+)
+
+# Build gmock
+ExternalProject_Add(
+    gmock_src
+    PREFIX "vendor/gtm"
+    SOURCE_DIR "vendor/gtm/src/googletest/googlemock"
+    INSTALL_DIR "vendor/gtm/gmock"
+    CMAKE_ARGS
+        -DCMAKE_INSTALL_PREFIX=${CMAKE_BINARY_DIR}/vendor/gtm/gmock
+        -DCMAKE_C_COMPILER=${CMAKE_C_COMPILER}
+        -DCMAKE_CXX_COMPILER=${CMAKE_CXX_COMPILER}
+        -DCMAKE_CXX_FLAGS=${CMAKE_CXX_FLAGS}
+    DOWNLOAD_COMMAND ""
+    UPDATE_COMMAND ""
+)
+
+# Prepare gtest
+ExternalProject_Get_Property(gtest_src install_dir)
+set(GTEST_INCLUDE_DIR ${install_dir}/include )
+set(GTEST_LIBRARY_PATH ${install_dir}/lib/libgtest.a)
+file(MAKE_DIRECTORY ${GTEST_INCLUDE_DIR})
+
+add_library(gtest STATIC IMPORTED)
+set_property(TARGET gtest PROPERTY IMPORTED_LOCATION ${GTEST_LIBRARY_PATH})
+set_property(TARGET gtest APPEND PROPERTY INTERFACE_INCLUDE_DIRECTORIES ${GTEST_INCLUDE_DIR})
+
+add_library(gtest_main STATIC IMPORTED)
+set(GTEST_MAIN_LIBRARY_PATH ${install_dir}/lib/libgtest_main.a)
+set_property(TARGET gtest_main PROPERTY IMPORTED_LOCATION ${GTEST_MAIN_LIBRARY_PATH})
+set_property(TARGET gtest_main APPEND PROPERTY INTERFACE_INCLUDE_DIRECTORIES ${GTEST_INCLUDE_DIR})
+
+# Prepare gmock
+ExternalProject_Get_Property(gmock_src install_dir)
+set(GMOCK_INCLUDE_DIR ${install_dir}/include)
+set(GMOCK_LIBRARY_PATH ${install_dir}/lib/libgmock.a)
+file(MAKE_DIRECTORY ${GMOCK_INCLUDE_DIR})
+add_library(gmock STATIC IMPORTED)
+set_property(TARGET gmock PROPERTY IMPORTED_LOCATION ${GMOCK_LIBRARY_PATH})
+set_property(TARGET gmock APPEND PROPERTY INTERFACE_INCLUDE_DIRECTORIES ${GMOCK_INCLUDE_DIR})
+
+# Dependencies
+add_dependencies(gtest_src googletest)
+add_dependencies(gmock_src googletest)
+add_dependencies(gtest gtest_src)
+add_dependencies(gtest_main gtest_src)
+add_dependencies(gmock gmock_src)

src/main.cpp

+#include <fcntl.h>
+#include <sys/mman.h>
+#include <unistd.h>
+#include <cassert>
+#include <chrono>
+#include <iostream>
+#include <memory>
+#include <numeric>
+#include <string>
+#include <vector>
+
+#include "sum.hpp"
+
+using namespace std::chrono;
+
+int main(int argc, char* argv[]) {
+    if (argc < 2) {
+        std::cout << "Usage: " << argv[0] << " <lineitem.tbl>" << std::endl;
+        return 1;
+    }
+
+    const auto start = high_resolution_clock::now();
+
+    //-- TODO exercise 3.4
+    // your code goes here
+    //--
+
+    char* data_start = static_cast<char*>(data);
+    int64_t price_sum;
+    sum_extendedprice<true>(data_start, data_start, size, &price_sum);
+    std::cout << price_sum << std::endl;
+
+    const auto duration =
+        duration_cast<milliseconds>(high_resolution_clock::now() - start)
+            .count();
+    std::cout << "duration: " << duration << " ms" << std::endl;
+
+    //-- TODO exercise 3.4
+    // cleanup
+    //--
+
+    return 0;
+}

src/sum.cpp

+#include <cassert>
+#include <memory>
+#include <numeric>
+#include <string>
+#include <string_view>
+
+int64_t ToInt(std::string_view s) {
+    int64_t result = 0;
+    for (auto c : s) result = result * 10 + (c - '0');
+    return result;
+}
+
+// pattern : the character to search broadcasted into a 64-bit integer
+// block : memory block in which to search
+// return : 64-bit integer with all the matches as seen in the lecture
+inline uint64_t get_matches(uint64_t pattern, uint64_t block) {
+    //-- TODO exercise 3.4 part 2
+    // your code goes here
+    //--
+}
+
+// pattern : the character to search broadcasted into a 64bit integer
+// begin : points somewhere into the partition
+// len : remaining length of the partition
+// return : the position of the first matching character or -1 otherwise
+ssize_t find_first(uint64_t pattern, const char* begin, size_t len) {
+    // locate the position of the following character (you'll have to use the
+    // pattern argument directly in the second part of the exercise)
+    const char to_search = pattern & 0xff;
+
+    // Hint: You may assume that reads from 'begin' within [len, len + 8) yield
+    // zero
+    //-- TODO exercise 3.4
+    // your code goes here
+    //--
+    return -1;
+}
+
+// begin : points somewhere into the partition
+// len : remaining length of the partition
+// return : 64-bit integer representation of the provided numeric
+int64_t read_numeric(const char* begin, size_t len) {
+    constexpr uint64_t period_pattern = 0x2E2E2E2E2E2E2E2Eull;
+    std::string_view numeric_view(begin, len);
+    ssize_t dot_position = find_first(period_pattern, begin, len);
+    assert(dot_position > 0);
+    auto part1 = numeric_view.substr(0, dot_position);
+    auto part2 = numeric_view.substr(dot_position + 1);
+    int64_t numeric = ToInt(part1) * 100 + ToInt(part2);
+    return numeric;
+}

test/test_sum.cpp

+#include <gtest/gtest.h>
+#include <cstring>
+#include <fstream>
+#include "sum.hpp"
+#include "util.hpp"
+
+//---------------------------------------------------------------------------
+TEST(SumTest, FindFirst)
+/// test find_first()
+{
+    constexpr uint64_t bar_pattern = 0x7C7C7C7C7C7C7C7Cull;
+    auto pos = find_first(bar_pattern, "|        ", 1);
+    ASSERT_EQ(pos, 0);
+    pos = find_first(bar_pattern, "   |        ", 4);
+    ASSERT_EQ(pos, 3);
+    pos = find_first(bar_pattern, "       |        ", 8);
+    ASSERT_EQ(pos, 7);
+    pos = find_first(bar_pattern, "           |        ", 12);
+    ASSERT_EQ(pos, 11);
+    pos = find_first(bar_pattern, "               |        ", 16);
+    ASSERT_EQ(pos, 15);
+}
+//---------------------------------------------------------------------------
+TEST(SumTest, ReadNumeric)
+/// test find_first()
+{
+    auto value = read_numeric("29312.32        ", 8);
+    ASSERT_EQ(value, 2931232);
+}
+//---------------------------------------------------------------------------
+TEST(SumTest, SumExtendedPrice)
+/// test sum_extendedprice()
+{
+    auto lineitem_file = getDir(__FILE__) + "/data/tpch/sf0_001/lineitem.tbl";
+    std::ifstream in(lineitem_file);
+    std::string contents((std::istreambuf_iterator<char>(in)),
+                         std::istreambuf_iterator<char>());
+    size_t full_size = contents.size() + 8;
+    char* buffer = new char[full_size]();
+    std::unique_ptr<char[]> guard(buffer);
+    std::memcpy(buffer, contents.c_str(), contents.size());
+
+    int64_t result;
+    sum_extendedprice<true>(buffer, buffer, contents.size(), &result);
+    ASSERT_EQ(result, 15277439838ul);
+}
+//---------------------------------------------------------------------------