Merge branch 'main' into gather-scatter

Author: ashvardanian

.vscode/settings.json

@@ -7,25 +7,34 @@
     "ashvardanian",
     "asynchrony",
     "bioinformatics",
+    "BLAS",
     "Boccara",
     "bootcamps",
     "Byrne",
+    "Cawley",
+    "cblas",
     "clflush",
+    "colsb",
     "consteval",
     "coro",
     "cppcoro",
     "CTRE",
     "CUDA",
+    "DOTPROD",
     "Dusíková",
+    "Eigen",
     "excerise",
     "fconcepts",
     "Fedor",
     "Fugaku",
+    "Giga",
     "Goodput",
+    "GOPS",
     "grandkids",
     "Hana",
     "Ibireme",
     "JeanHeyd",
+    "jemalloc",
     "Kulukundis",
     "Lelbach",
     "Lemire",
@@ -35,6 +44,7 @@
     "Maclaurin",
     "matmul",
     "Meneide",
+    "mimalloc",
     "MSVC",
     "Müller",
     "Neoverse",
@@ -44,6 +54,7 @@
     "NVCC",
     "openmp",
     "Ormrod",
+    "Peta",
     "Pikus",
     "pmf",
     "popcountll",
@@ -60,13 +71,21 @@
     "Sutter",
     "sysfs",
     "taskset",
+    "Tera",
+    "TMUL",
     "Trettner",
     "Unif",
     "unifex",
     "unsalvageable",
     "unscalable",
     "Vardanian",
+    "VNNI",
     "Weis",
+    "XCOMP",
+    "XFEATURE",
+    "XTILE",
+    "XTILECFG",
+    "XTILEDATA",
     "Yaoyuan",
     "yyjson",
     "Zverovich"
@@ -90,11 +109,13 @@
     "__tuple": "cpp",
     "__verbose_abort": "cpp",
     "algorithm": "cpp",
+    "any": "cpp",
     "array": "cpp",
     "atomic": "cpp",
     "bit": "cpp",
     "bitset": "cpp",
     "cctype": "cpp",
+    "cfenv": "cpp",
     "charconv": "cpp",
     "chrono": "cpp",
     "clocale": "cpp",
@@ -116,6 +137,7 @@
     "deque": "cpp",
     "exception": "cpp",
     "execution": "cpp",
+    "expected": "cpp",
     "format": "cpp",
     "forward_list": "cpp",
     "fstream": "cpp",
@@ -148,6 +170,7 @@
     "shared_mutex": "cpp",
     "source_location": "cpp",
     "span": "cpp",
+    "sstream": "cpp",
     "stdexcept": "cpp",
     "stop_token": "cpp",
     "streambuf": "cpp",
@@ -162,10 +185,25 @@
     "unordered_map": "cpp",
     "unordered_set": "cpp",
     "utility": "cpp",
+    "valarray": "cpp",
     "variant": "cpp",
     "vector": "cpp",
-    "sstream": "cpp",
-    "cfenv": "cpp",
-    "expected": "cpp"
+    "core": "cpp",
+    "superlusupport": "cpp",
+    "*.evaluator": "cpp",
+    "*.traits": "cpp",
+    "adolcforward": "cpp",
+    "alignedvector3": "cpp",
+    "autodiff": "cpp",
+    "bvh": "cpp",
+    "eulerangles": "cpp",
+    "fft": "cpp",
+    "kroneckerproduct": "cpp",
+    "mprealsupport": "cpp",
+    "nnls": "cpp",
+    "numericaldiff": "cpp",
+    "openglsupport": "cpp",
+    "specialfunctions": "cpp",
+    "splines": "cpp"
   }
 }

CMakeLists.txt

@@ -5,8 +5,8 @@ cmake_minimum_required(VERSION 3.16 FATAL_ERROR)
 # Project Setup
 # ------------------------------------------------------------------------------
 project(less_slow
-  VERSION 0.1.0
-  LANGUAGES C CXX
+  VERSION 0.2.0
+  LANGUAGES C CXX ASM
   DESCRIPTION "Learning how to write Less Slow code, from numerical micro-kernels and SIMD to coroutines, ranges, and polymorphic state machines"
   HOMEPAGE_URL "https://github.com/ashvardanian/less_slow.cpp")
 
@@ -32,6 +32,13 @@ endif()
 # ------------------------------------------------------------------------------
 find_package(Threads REQUIRED)
 find_package(OpenMP REQUIRED)
+find_package(BLAS REQUIRED)
+if (BLAS_FOUND)
+    message(STATUS "BLAS found: ${BLAS_LIBRARIES}")
+else ()
+    message(FATAL_ERROR "BLAS not found")
+endif ()
+
 
 set(FETCHCONTENT_QUIET OFF)
 include(FetchContent)
@@ -167,12 +174,29 @@ FetchContent_Declare(
 )
 FetchContent_MakeAvailable(YaoyuanGuoYYJSON)
 
+# Eigen is one of the few libraries not using GitHub
+FetchContent_Declare(
+  LibEigenEigen
+  GIT_REPOSITORY https://gitlab.com/libeigen/eigen.git
+  GIT_TAG master
+)
+FetchContent_MakeAvailable(LibEigenEigen)
+
 # ------------------------------------------------------------------------------
 # Main Executable
 # ------------------------------------------------------------------------------
 add_executable(less_slow less_slow.cpp)
 set_target_properties(less_slow PROPERTIES POSITION_INDEPENDENT_CODE ON)
 
+# Conditionally add the assembly file(s)
+if(CMAKE_SYSTEM_PROCESSOR MATCHES "x86_64")
+    set_source_files_properties(less_slow_amd64.S PROPERTIES LANGUAGE ASM)
+    target_sources(less_slow PRIVATE less_slow_amd64.S)
+elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "aarch64")
+    set_source_files_properties(less_slow_aarch64.S PROPERTIES LANGUAGE ASM)
+    target_sources(less_slow PRIVATE less_slow_aarch64.S)
+endif()
+
 # ------------------------------------------------------------------------------
 # Compiler Flags / Options
 # ------------------------------------------------------------------------------
@@ -226,6 +250,8 @@ target_link_libraries(less_slow
     # https://github.com/abseil/abseil-cpp/blob/master/CMake/README.md#available-abseil-cmake-public-targets
     absl::flat_hash_map
     nlohmann_json::nlohmann_json
+    Eigen3::Eigen
+    ${BLAS_LIBRARIES}
     $<$<STREQUAL:${CMAKE_SYSTEM_NAME},Linux>:TBB::tbb>
     $<$<STREQUAL:${CMAKE_SYSTEM_NAME},Linux>:OpenMP::OpenMP_CXX>
 )

VERSION

@@ -1 +1 @@
-0.1.0
+0.2.0