simd: use new glmm_min/max in vec4 where possible

win32, simd: ensure we are on msvc when checking MSVC specific headers

.github/workflows/cmake-wasm.yml

@@ -0,0 +1,97 @@
+name: CMake WebAssembly
+
+on:
+  push:
+    branches: [ "master" ]
+  pull_request:
+    branches: [ "master" ]
+
+env:
+  wasmtime_version: v7.0.0
+  wasmer_version: v3.1.1
+
+jobs:
+  build_wasi_sdk:
+    strategy:
+      matrix:
+        BUILD_TYPE: [Release, Debug, RelWithDebInfo, MinSizeRel]
+        C_FLAGS: ['', '-msimd128']
+        wasi_sdk_version: [19, 20]
+    # The CMake configure and build commands are platform agnostic and should work equally well on Windows or Mac.
+    # You can convert this to a matrix build if you need cross-platform coverage.
+    # See: https://docs.github.com/en/free-pro-team@latest/actions/learn-github-actions/managing-complex-workflows#using-a-build-matrix
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Downloading wasi-sdk
+      run: |
+        cd ${{github.workspace}}
+        wget --no-verbose https://github.com/WebAssembly/wasi-sdk/releases/download/wasi-sdk-${{matrix.wasi_sdk_version}}/wasi-sdk-${{matrix.wasi_sdk_version}}.0-linux.tar.gz
+        tar xf wasi-sdk-${{matrix.wasi_sdk_version}}.0-linux.tar.gz
+
+    - name: Configure CMake
+      # Configure CMake in a 'build' subdirectory. `CMAKE_BUILD_TYPE` is only required if you are using a single-configuration generator such as make.
+      # See https://cmake.org/cmake/help/latest/variable/CMAKE_BUILD_TYPE.html?highlight=cmake_build_type
+      run: cmake -B ${{github.workspace}}/build -DCMAKE_BUILD_TYPE=${{matrix.BUILD_TYPE}} -DCMAKE_C_FLAGS="${{matrix.C_FLAGS}}" -DCMAKE_TOOLCHAIN_FILE=${{github.workspace}}/wasi-sdk-${{matrix.wasi_sdk_version}}.0/share/cmake/wasi-sdk.cmake -DWASI_SDK_PREFIX=${{github.workspace}}/wasi-sdk-${{matrix.wasi_sdk_version}}.0 -DCGLM_USE_TEST=ON
+
+    - name: Build
+      # Build your program with the given configuration
+      run: cmake --build ${{github.workspace}}/build --config ${{matrix.BUILD_TYPE}}
+
+    - name: Test with wasmtime
+      run: |
+        cd ${{github.workspace}}
+        ls -lh ${{github.workspace}}/build/
+        wget --no-verbose https://github.com/bytecodealliance/wasmtime/releases/download/${{env.wasmtime_version}}/wasmtime-${{env.wasmtime_version}}-x86_64-linux.tar.xz
+        tar xf wasmtime-${{env.wasmtime_version}}-x86_64-linux.tar.xz
+        ./wasmtime-${{env.wasmtime_version}}-x86_64-linux/wasmtime run --wasm-features simd ${{github.workspace}}/build/tests
+
+    - name: Test with wasmer
+      run: |
+        cd ${{github.workspace}}
+        mkdir wasmer
+        cd wasmer
+        wget --no-verbose https://github.com/wasmerio/wasmer/releases/download/${{env.wasmer_version}}/wasmer-linux-amd64.tar.gz
+        tar xf wasmer-linux-amd64.tar.gz
+        ./bin/wasmer run --enable-simd ${{github.workspace}}/build/tests
+
+  build_emsdk:
+    strategy:
+      matrix:
+        BUILD_TYPE: [Release, Debug, RelWithDebInfo, MinSizeRel]
+        C_FLAGS: ['', '-msimd128', '-msse -msse2 -msimd128', '-msse -msse2 -msse3 -msse4 -msimd128']
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+    - name: Setup emsdk
+      uses: mymindstorm/setup-emsdk@v12
+
+    - name: Verify emsdk
+      run: emcc -v
+
+    - name: Configure CMake
+      # Configure CMake in a 'build' subdirectory. `CMAKE_BUILD_TYPE` is only required if you are using a single-configuration generator such as make.
+      # See https://cmake.org/cmake/help/latest/variable/CMAKE_BUILD_TYPE.html?highlight=cmake_build_type
+      run: emcmake cmake -B ${{github.workspace}}/build -DCMAKE_BUILD_TYPE=${{matrix.BUILD_TYPE}} -DCMAKE_C_FLAGS="${{matrix.C_FLAGS}}" -DCMAKE_EXE_LINKER_FLAGS="-s STANDALONE_WASM" -DCGLM_STATIC=ON -DCGLM_USE_TEST=ON
+
+    - name: Build
+      # Build your program with the given configuration
+      run: cmake --build ${{github.workspace}}/build --config ${{matrix.BUILD_TYPE}}
+
+    - name: Test with wasmtime
+      run: |
+        cd ${{github.workspace}}
+        ls -lh ${{github.workspace}}/build/
+        wget --no-verbose https://github.com/bytecodealliance/wasmtime/releases/download/${{env.wasmtime_version}}/wasmtime-${{env.wasmtime_version}}-x86_64-linux.tar.xz
+        tar xf wasmtime-${{env.wasmtime_version}}-x86_64-linux.tar.xz
+        ./wasmtime-${{env.wasmtime_version}}-x86_64-linux/wasmtime run --wasm-features simd ${{github.workspace}}/build/tests.wasm
+    - name: Test with wasmer
+      run: |
+        cd ${{github.workspace}}
+        mkdir wasmer
+        cd wasmer
+        wget --no-verbose https://github.com/wasmerio/wasmer/releases/download/${{env.wasmer_version}}/wasmer-linux-amd64.tar.gz
+        tar xf wasmer-linux-amd64.tar.gz
+        ./bin/wasmer run --enable-simd ${{github.workspace}}/build/tests.wasm

.github/workflows/meson-wasm.yml

@@ -0,0 +1,79 @@
+name: Meson WebAssembly
+
+on:
+  push:
+    branches: [ "master" ]
+  pull_request:
+    branches: [ "master" ]
+
+env:
+  wasmtime_version: v7.0.0
+  wasmer_version: v3.1.1
+
+jobs:
+  build_emsdk:
+    strategy:
+      matrix:
+        BUILD_TYPE: [debug, debugoptimized, release, minsize]
+        C_FLAGS: ['', '-msimd128', '-msse -msse2 -msimd128', '-msse -msse2 -msse3 -msse4 -msimd128']
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+    - name: Setup emsdk
+      uses: mymindstorm/setup-emsdk@v13
+
+    - name: Verify emsdk
+      run: emcc -v
+
+    - name: Creating cross file
+      run: |
+        cat << EOF > ${{github.workspace}}/meson_cross_emsdk.txt
+        [binaries]
+        c = '`which emcc`'
+        cpp = '`which em++`'
+        ar = '`which emar`'
+        
+        [built-in options]
+        
+        c_args = ['-Wno-unused-parameter']
+        c_link_args = ['-s', 'STANDALONE_WASM']
+        cpp_args = ['-Wno-unused-parameter']
+        cpp_link_args = ['-s', 'STANDALONE_WASM']
+        
+        [host_machine]
+        
+        system = 'emscripten'
+        cpu_family = 'wasm32'
+        cpu = 'wasm32'
+        endian = 'little'
+
+        EOF
+        cat ${{github.workspace}}/meson_cross_emsdk.txt
+
+    - uses: actions/setup-python@v4
+    
+    - name: Install meson
+      run: |
+        sudo python3 -m pip install meson ninja
+
+    - name: Build with meson
+      run: |
+        meson setup build -Dbuildtype=${{matrix.BUILD_TYPE}} --cross-file ${{github.workspace}}/meson_cross_emsdk.txt --default-library=static -Dbuild_tests=true
+        meson test -C build
+
+    - name: Test with wasmtime
+      run: |
+        cd ${{github.workspace}}
+        ls -lh ${{github.workspace}}/build/
+        wget --no-verbose https://github.com/bytecodealliance/wasmtime/releases/download/${{env.wasmtime_version}}/wasmtime-${{env.wasmtime_version}}-x86_64-linux.tar.xz
+        tar xf wasmtime-${{env.wasmtime_version}}-x86_64-linux.tar.xz
+        ./wasmtime-${{env.wasmtime_version}}-x86_64-linux/wasmtime run --wasm-features simd ${{github.workspace}}/build/tests.wasm
+
+    - name: Test with wasmer
+      run: |
+        cd ${{github.workspace}}
+        mkdir wasmer
+        cd wasmer
+        wget --no-verbose https://github.com/wasmerio/wasmer/releases/download/${{env.wasmer_version}}/wasmer-linux-amd64.tar.gz
+        tar xf wasmer-linux-amd64.tar.gz
+        ./bin/wasmer run --enable-simd ${{github.workspace}}/build/tests.wasm

.readthedocs.yaml

@@ -0,0 +1,39 @@
+# Read the Docs configuration file for Sphinx projects
+# See https://docs.readthedocs.io/en/stable/config-file/v2.html for details
+
+# Required
+version: 2
+
+# Set the OS, Python version and other tools you might need
+build:
+  os: ubuntu-22.04
+  tools:
+    python: "3.12"
+    # You can also specify other tool versions:
+    # nodejs: "20"
+    # rust: "1.70"
+    # golang: "1.20"
+
+# Build documentation in the "docs/" directory with Sphinx
+sphinx:
+  configuration: docs/source/conf.py
+  # You can configure Sphinx to use a different builder, for instance use the dirhtml builder for simpler URLs
+  # builder: "dirhtml"
+  # Fail on all warnings to avoid broken references
+  # fail_on_warning: true
+
+# Optionally build your docs in additional formats such as PDF and ePub
+# formats:
+#   - pdf
+#   - epub
+
+# Optional but recommended, declare the Python requirements required
+# to build your documentation
+# See https://docs.readthedocs.io/en/stable/guides/reproducible-builds.html
+# python:
+#   install:
+#     - requirements: docs/requirements.txt
+
+python:
+  install:
+    - requirements: docs/requirements.txt

.vscode/c_cpp_properties.json

@@ -0,0 +1,20 @@
+{
+    "configurations": [
+        {
+            "name": "Mac",
+            "includePath": [
+                "${workspaceFolder}/**"
+            ],
+            "defines": [],
+            "macFrameworkPath": [
+                "/Applications/Xcode.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks"
+            ],
+            "compilerPath": "/usr/bin/clang",
+            "cStandard": "c23",
+            "cppStandard": "c++23",
+            "intelliSenseMode": "macos-clang-arm64",
+            "configurationProvider": "vector-of-bool.cmake-tools"
+        }
+    ],
+    "version": 4
+}

.vscode/settings.json

@@ -1,4 +1,31 @@
 {
   "C_Cpp.default.configurationProvider": "vector-of-bool.cmake-tools",
-  "restructuredtext.confPath": "${workspaceFolder}/docs/source"
+  "restructuredtext.confPath": "${workspaceFolder}/docs/source",
+  "workbench.colorTheme": "Default Light+ Experimental",
+  "editor.fontSize": 14,
+  "workbench.colorCustomizations": {
+	"[Default Light*]": {
+  	    "editor.background": "#fefefe",
+  	    "sideBar.background": "#fefefe",
+  	    "sideBar.foreground": "#343436",
+  	    "sideBarTitle.foreground": "#343436",
+  	    "sideBar.border": "#e2e2e4",
+  	    "statusBar.background": "#fefefe",
+  	    "titleBar.activeBackground": "#fefefe",
+  	    "tab.activeBackground": "#f4fff4aa",
+  	    "tab.inactiveBackground": "#fefefe",
+  	    "activityBar.background": "#fefefe",
+  	    "editorGroupHeader.tabsBackground": "#fefefe"
+	},
+	"[Default Dark*]": {
+		"editor.background": "#1D1D25",
+		"sideBar.background": "#1D1D25",
+		"statusBar.background": "#1D1D25",
+		"titleBar.activeBackground": "#1D1D25",
+		"tab.activeBackground": "#2C2C3A",
+		"tab.inactiveBackground": "#1D1D25",
+		"activityBar.background": "#1D1D25",
+		"editorGroupHeader.tabsBackground": "#1D1D25"
+    }
+  },
 }

CMakeLists.txt

@@ -1,6 +1,6 @@
 cmake_minimum_required(VERSION 3.8.2)
 project(cglm
-  VERSION 0.8.9
+  VERSION 0.9.3
   HOMEPAGE_URL https://github.com/recp/cglm
   DESCRIPTION "OpenGL Mathematics (glm) for C"
   LANGUAGES C
@@ -16,6 +16,11 @@ option(CGLM_STATIC "Static build" OFF)
 option(CGLM_USE_C99 "" OFF)
 option(CGLM_USE_TEST "Enable Tests" OFF)
 
+if(CMAKE_SYSTEM_NAME STREQUAL WASI)
+  set(CGLM_STATIC ON CACHE BOOL "Static option" FORCE)
+  set(CGLM_SHARED OFF CACHE BOOL "Shared option" FORCE)
+endif()
+
 if(NOT CGLM_STATIC AND CGLM_SHARED)
   set(CGLM_BUILD SHARED)
 else(CGLM_STATIC)
@@ -38,7 +43,7 @@ if(MSVC)
     string(REGEX REPLACE "/RTC(su|[1su])" "" ${flag_var} "${${flag_var}}")
   endforeach(flag_var)
 else()
-  add_compile_options(-Wall -Werror -O3)
+  add_compile_options(-Wall -O3)
 endif()
 
 get_directory_property(hasParent PARENT_DIRECTORY)
@@ -74,8 +79,14 @@ add_library(${PROJECT_NAME}
   src/vec4.c
   src/ivec4.c
   src/mat2.c
+  src/mat2x3.c
+  src/mat2x4.c
   src/mat3.c
+  src/mat3x2.c
+  src/mat3x4.c
   src/mat4.c
+  src/mat4x2.c
+  src/mat4x3.c
   src/plane.c
   src/frustum.c
   src/box.c

Makefile.am

@@ -11,8 +11,7 @@ AM_CFLAGS = -Wall \
             -std=gnu11 \
             -O3 \
             -Wstrict-aliasing=2 \
-            -fstrict-aliasing \
+            -fstrict-aliasing
-            -Werror=strict-prototypes
 
 lib_LTLIBRARIES = libcglm.la
 libcglm_la_LDFLAGS = -no-undefined -version-info 0:1:0
@@ -43,8 +42,14 @@ cglm_HEADERS = include/cglm/version.h \
                include/cglm/cam.h \
                include/cglm/io.h \
                include/cglm/mat4.h \
+               include/cglm/mat4x2.h \
+               include/cglm/mat4x3.h \
                include/cglm/mat3.h \
+               include/cglm/mat3x2.h \
+               include/cglm/mat3x4.h \
                include/cglm/mat2.h \
+               include/cglm/mat2x3.h \
+               include/cglm/mat2x4.h \
                include/cglm/affine-pre.h \
                include/cglm/affine-post.h \
                include/cglm/affine.h \
@@ -64,6 +69,7 @@ cglm_HEADERS = include/cglm/version.h \
                include/cglm/plane.h \
                include/cglm/frustum.h \
                include/cglm/box.h \
+               include/cglm/aabb2d.h \
                include/cglm/color.h \
                include/cglm/project.h \
                include/cglm/sphere.h \
@@ -95,8 +101,14 @@ cglm_clipspace_HEADERS = include/cglm/clipspace/persp.h \
 
 cglm_calldir=$(includedir)/cglm/call
 cglm_call_HEADERS = include/cglm/call/mat4.h \
+                    include/cglm/call/mat4x2.h \
+                    include/cglm/call/mat4x3.h \
                     include/cglm/call/mat3.h \
+                    include/cglm/call/mat3x2.h \
+                    include/cglm/call/mat3x4.h \
                     include/cglm/call/mat2.h \
+                    include/cglm/call/mat2x3.h \
+                    include/cglm/call/mat2x4.h \
                     include/cglm/call/vec2.h \
                     include/cglm/call/vec3.h \
                     include/cglm/call/vec4.h \
@@ -138,7 +150,8 @@ cglm_call_clipspace_HEADERS = include/cglm/call/clipspace/persp_lh_no.h \
 cglm_simddir=$(includedir)/cglm/simd
 cglm_simd_HEADERS = include/cglm/simd/intrin.h \
                     include/cglm/simd/x86.h \
-                    include/cglm/simd/arm.h
+                    include/cglm/simd/arm.h \
+                    include/cglm/simd/wasm.h
 
 cglm_simd_sse2dir=$(includedir)/cglm/simd/sse2
 cglm_simd_sse2_HEADERS = include/cglm/simd/sse2/affine.h \
@@ -157,12 +170,30 @@ cglm_simd_neon_HEADERS = include/cglm/simd/neon/affine.h \
                          include/cglm/simd/neon/mat4.h \
                          include/cglm/simd/neon/quat.h
 
+cglm_simd_wasmdir=$(includedir)/cglm/simd/wasm
+cglm_simd_wasm_HEADERS = include/cglm/simd/wasm/affine.h \
+                         include/cglm/simd/wasm/mat2.h \
+                         include/cglm/simd/wasm/mat3.h \
+                         include/cglm/simd/wasm/mat4.h \
+                         include/cglm/simd/wasm/quat.h
+
+cglm_handeddir=$(includedir)/cglm/handed
+cglm_handed_HEADERS = include/cglm/handed/euler_to_quat_lh.h \
+                      include/cglm/handed/euler_to_quat_rh.h
+
 cglm_structdir=$(includedir)/cglm/struct
 cglm_struct_HEADERS = include/cglm/struct/mat4.h \
+                      include/cglm/struct/mat4x2.h \
+                      include/cglm/struct/mat4x3.h \
                       include/cglm/struct/mat3.h \
+                      include/cglm/struct/mat3x2.h \
+                      include/cglm/struct/mat3x4.h \
                       include/cglm/struct/mat2.h \
+                      include/cglm/struct/mat2x3.h \
+                      include/cglm/struct/mat2x4.h \
                       include/cglm/struct/affine-pre.h \
                       include/cglm/struct/affine-post.h \
+                      include/cglm/struct/affine-mat.h \
                       include/cglm/struct/affine.h \
                       include/cglm/struct/affine2d.h \
                       include/cglm/struct/vec2.h \
@@ -178,6 +209,7 @@ cglm_struct_HEADERS = include/cglm/struct/mat4.h \
                       include/cglm/struct/plane.h \
                       include/cglm/struct/frustum.h \
                       include/cglm/struct/box.h \
+                      include/cglm/struct/aabb2d.h \
                       include/cglm/struct/project.h \
                       include/cglm/struct/sphere.h \
                       include/cglm/struct/color.h \
@@ -199,6 +231,10 @@ cglm_struct_clipspace_HEADERS = include/cglm/struct/clipspace/persp_lh_no.h \
                                 include/cglm/struct/clipspace/project_no.h \
                                 include/cglm/struct/clipspace/project_zo.h
 
+cglm_struct_handeddir=$(includedir)/cglm/struct/handed
+cglm_struct_handed_HEADERS = include/cglm/struct/handed/euler_to_quat_lh.h \
+                             include/cglm/struct/handed/euler_to_quat_rh.h
+
 libcglm_la_SOURCES=\
     src/euler.c \
     src/affine.c \
@@ -212,8 +248,14 @@ libcglm_la_SOURCES=\
     src/vec4.c \
     src/ivec4.c \
     src/mat2.c \
+    src/mat2x3.c \
+    src/mat2x4.c \
     src/mat3.c \
+    src/mat3x2.c \
+    src/mat3x4.c \
     src/mat4.c \
+    src/mat4x2.c \
+    src/mat4x3.c \
     src/plane.c \
     src/frustum.c \
     src/box.c \
@@ -243,11 +285,6 @@ test_tests_SOURCES=\
     test/runner.c \
     test/src/test_common.c \
     test/src/tests.c \
-    test/src/test_cam.c \
-    test/src/test_cam_lh_zo.c \
-    test/src/test_cam_rh_zo.c \
-    test/src/test_cam_lh_no.c \
-    test/src/test_cam_rh_no.c \
     test/src/test_clamp.c \
     test/src/test_euler.c \
     test/src/test_bezier.c \

README.md

@@ -242,6 +242,38 @@ add_subdirectory(external/cglm/)
 # or you can use find_package to configure cglm
 ```
 
+#### Use CMake to build for WebAssembly
+
+Since math functions like `sinf` is used, this can not be targeted at `wasm32-unknown-unknown`, one of [wasi-sdk](https://github.com/WebAssembly/wasi-sdk) or [emscripten](https://github.com/emscripten-core/emsdk) should be used.
+
+Should note that shared build is not yet supported for WebAssembly.
+
+For [simd128](https://github.com/WebAssembly/simd) support, add `-msimd128` to `CMAKE_C_FLAGS`, in command line `-DCMAKE_C_FLAGS="-msimd128"`.
+
+For tests, the cmake option `CGLM_USE_TEST` would still work, you'll need a wasi runtime for running tests, see our [ci config file](.github/workflows/cmake-wasm.yml) for a detailed example.
+
+##### Use CMake and WASI SDK to build for WebAssembly
+
+```bash
+$ cmake .. \
+  -DCMAKE_TOOLCHAIN_FILE=/path/to/wasi-sdk-19.0/share/cmake/wasi-sdk.cmake \
+  -DWASI_SDK_PREFIX=/path/to/wasi-sdk-19.0
+```
+
+Where `/path/to/wasi-sdk-19.0/` is the path to extracted [wasi sdk](https://github.com/WebAssembly/wasi-sdk).
+
+In this case it would by default make a static build.
+
+##### Use CMake and Emscripten SDK to build for WebAssembly
+
+```bash
+$ emcmake cmake .. \
+  -DCMAKE_EXE_LINKER_FLAGS="-s STANDALONE_WASM" \
+  -DCGLM_STATIC=ON
+```
+
+The `emcmake` here is the cmake wrapper for Emscripten from installed [emsdk](https://github.com/emscripten-core/emsdk).
+
 ### Meson (All platforms)
 
 ```bash
@@ -257,7 +289,7 @@ $ sudo ninja install # [Optional]
 c_std=c11
 buildtype=release
 default_library=shared
-enable_tests=false # to run tests: ninja test
+build_tests=true # to run tests: ninja test
 ```
 #### Use with your Meson project
 * Example:

cglm.podspec

@@ -2,7 +2,7 @@ Pod::Spec.new do |s|
 
   # Description
   s.name         = "cglm"
-  s.version      = "0.8.8"
+  s.version      = "0.9.2"
   s.summary      = "📽 Highly Optimized Graphics Math (glm) for C"
   s.description  = <<-DESC
 cglm is math library for graphics programming for C. See the documentation or README for all features.

configure.ac

@@ -7,8 +7,8 @@
 #*****************************************************************************
 
 AC_PREREQ([2.69])
-AC_INIT([cglm], [0.8.9], [info@recp.me])
+AC_INIT([cglm], [0.9.3], [info@recp.me])
-AM_INIT_AUTOMAKE([-Wall -Werror foreign subdir-objects serial-tests])
+AM_INIT_AUTOMAKE([-Wall foreign subdir-objects serial-tests])
 
 # Don't use the default cflags (-O2 -g), we set ours manually in Makefile.am.
 : ${CFLAGS=""}

docs/requirements.txt

@@ -0,0 +1,4 @@
+# Defining the exact version will make sure things don't break
+sphinx==5.3.0
+sphinx_rtd_theme==1.1.1
+readthedocs-sphinx-search==0.1.1

docs/source/aabb2d.rst

@@ -0,0 +1,173 @@
+.. default-domain:: C
+
+2d axis aligned bounding box (AABB)
+================================================================================
+
+Header: cglm/aabb2d.h
+
+Some convenient functions provided for AABB.
+
+**Definition of aabb:**
+
+cglm defines an aabb as a two dimensional array of vec2's.
+The first element is the **min** point and the second one is the **max** point.
+If you have another type e.g. struct or even another representation then you must
+convert it before and after calling a cglm aabb2d function.
+
+Table of contents (click to go):
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Functions:
+
+1. :c:func:`glm_aabb2d_copy`
+#. :c:func:`glm_aabb2d_transform`
+#. :c:func:`glm_aabb2d_merge`
+#. :c:func:`glm_aabb2d_crop`
+#. :c:func:`glm_aabb2d_crop_until`
+#. :c:func:`glm_aabb2d_invalidate`
+#. :c:func:`glm_aabb2d_isvalid`
+#. :c:func:`glm_aabb2d_size`
+#. :c:func:`glm_aabb2d_radius`
+#. :c:func:`glm_aabb2d_center`
+#. :c:func:`glm_aabb2d_aabb`
+#. :c:func:`glm_aabb2d_circle`
+#. :c:func:`glm_aabb2d_point`
+#. :c:func:`glm_aabb2d_contains`
+
+Functions documentation
+~~~~~~~~~~~~~~~~~~~~~~~
+
+.. c:function:: void  glm_aabb2d_copy(vec2 aabb[2], vec2 dest[2])
+
+    | copy all members of [aabb] to [dest]
+
+    Parameters:
+      | *[in]*  **aabb**  bounding box
+      | *[out]* **dest**  destination
+
+.. c:function:: void  glm_aabb2d_transform(vec2 aabb[2], mat3 m, vec2 dest[2])
+
+    | apply transform to Axis-Aligned Bounding Box
+
+    Parameters:
+      | *[in]*  **aabb**   bounding box
+      | *[in]*  **m**     transform matrix
+      | *[out]* **dest**  transformed bounding box
+
+.. c:function:: void  glm_aabb2d_merge(vec2 aabb1[2], vec2 aabb2[2], vec2 dest[2])
+
+    | merges two AABB bounding box and creates new one
+
+    two aabb must be in the same space
+
+    Parameters:
+      | *[in]*  **aabb1** bounding box 1
+      | *[in]*  **aabb2** bounding box 2
+      | *[out]* **dest** merged bounding box
+
+.. c:function:: void  glm_aabb2d_crop(vec2 aabb[2], vec2 cropAabb[2], vec2 dest[2])
+
+    | crops a bounding box with another one.
+
+    this could be useful for gettng a bbox which fits with view frustum and
+    object bounding boxes. In this case you crop view frustum box with objects
+    box
+
+    Parameters:
+      | *[in]*  **aabb**      bounding box 1
+      | *[in]*  **cropAabb**  crop box
+      | *[out]* **dest**     cropped bounding box
+
+.. c:function:: void  glm_aabb2d_crop_until(vec2 aabb[2], vec2 cropAabb[2], vec2 clampAabb[2], vec2 dest[2])
+
+    | crops a bounding box with another one.
+
+    this could be useful for gettng a bbox which fits with view frustum and
+    object bounding boxes. In this case you crop view frustum box with objects
+    box
+
+    Parameters:
+      | *[in]*  **aabb**      bounding box
+      | *[in]*  **cropAabb**  crop box
+      | *[in]*  **clampAabb** miniumum box
+      | *[out]* **dest**     cropped bounding box
+
+.. c:function:: void  glm_aabb2d_invalidate(vec2 aabb[2])
+
+    | invalidate AABB min and max values
+
+    | It fills *max* values with -FLT_MAX and *min* values with +FLT_MAX
+
+    Parameters:
+      | *[in, out]*   **aabb**     bounding box
+
+.. c:function:: bool  glm_aabb2d_isvalid(vec2 aabb[2])
+
+    | check if AABB is valid or not
+
+    Parameters:
+      | *[in]*   **aabb**     bounding box
+
+    Returns:
+      returns true if aabb is valid otherwise false
+
+.. c:function:: float  glm_aabb2d_size(vec2 aabb[2])
+
+    | distance between of min and max
+
+    Parameters:
+      | *[in]*   **aabb**     bounding box
+
+    Returns:
+      distance between min - max
+
+.. c:function:: float  glm_aabb2d_radius(vec2 aabb[2])
+
+    | radius of sphere which surrounds AABB
+
+    Parameters:
+      | *[in]*   **aabb**     bounding box
+
+.. c:function:: void  glm_aabb2d_center(vec2 aabb[2], vec2 dest)
+
+    | computes center point of AABB
+
+    Parameters:
+      | *[in]*    **aabb**      bounding box
+      | *[out]*   **dest**     center of bounding box
+
+.. c:function:: bool  glm_aabb2d_aabb(vec2 aabb[2], vec2 other[2])
+
+    | check if two AABB intersects
+
+    Parameters:
+      | *[in]*    **aabb**     bounding box
+      | *[out]*   **other**   other bounding box
+
+.. c:function:: bool  glm_aabb2d_circle(vec2 aabb[2], vec3 c)
+
+    | check if AABB intersects with sphere
+
+    | https://github.com/erich666/GraphicsGems/blob/master/gems/BoxSphere.c
+    | Solid Box - Solid Sphere test.
+
+    Parameters:
+      | *[in]*    **aabb**     solid bounding box
+      | *[out]*   **c**        solid circle
+
+.. c:function:: bool  glm_aabb2d_point(vec2 aabb[2], vec2 point)
+
+    | check if point is inside of AABB
+
+    Parameters:
+      | *[in]*    **aabb**     bounding box
+      | *[out]*   **point**   point
+
+.. c:function:: bool  glm_aabb2d_contains(vec2 aabb[2], vec2 other[2])
+
+    | check if AABB contains other AABB
+
+    Parameters:
+      | *[in]*    **aabb**     bounding box
+      | *[out]*   **other**   other bounding box
+

docs/source/api.rst

@@ -1,59 +1,28 @@
-API documentation
+📚 API documentation
 ================================
 
-Some functions may exist twice,
+**cglm** provides a few APIs for similar functions.
-once for their namespace and once for global namespace
-to make easier to write very common functions
 
-For instance, in general we use :code:`glm_vec3_dot` to get dot product
+* 📦 **Inline API**: All functions are inline. You can include **cglm/cglm.h** header
-of two **vec3**. Now we can also do this with :code:`glm_dot`,
+  to use this API. This is the default API. `glm_` is namespace/prefix for this API. 
-same for *_cross* and so on...
+* 📦 **Call API**: All functions are not inline. You need to build *cglm* and link it
+  to your project. You can include **cglm/call.h** header to use this API. `glmc_` is namespace/prefix for this API. 
 
-The original function stays where it is, the function in global namespace
+And also there are also sub categories:
-of same name is just an alias, so there is no call version of those functions.
-e.g there is no func like :code:`glmc_dot` because *glm_dot* is just alias for
-:code:`glm_vec3_dot`
 
-By including **cglm/cglm.h** header you will include all inline version
+* 📦 **Array API**: Types are raw arrays and functions takes array as argument. You can include **cglm/cglm.h** header
-of functions. Since functions in this header[s] are inline you don't need to
+  to use this API. This is the default API. `glm_` is namespace/prefix for this API. 
-build or link *cglm* against your project.
+* 📦 **Struct API**: Types are union/struct and functions takes struct as argument and return structs if needed. You can include **cglm/struct.h** header
+  to use this API. This also includes **cglm/cglm.h** header.`glms_` is namespace/prefix for this API but your can omit or change it, see struct api docs.
+* 📦 **SIMD API**: SIMD functions and helpers. `glmm_` is namespace/prefix for this API. 
 
-But by including **cglm/call.h** header you will include all *non-inline*
+📌 Since struct api and call api are built top of inline array api, follow inline array api docs for individual functions. 
-version of functions. You need to build *cglm* and link it.
-Follow the :doc:`build` documentation for this
 
 .. toctree::
    :maxdepth: 1
-   :caption: API categories:
+   :caption: API documentations:
 
-   affine
+   api_inline_array
-   affine-mat
+   api_struct
-   affine2d
+   api_call
-   cam
+   api_simd
-   frustum
-   box
-   quat
-   euler
-   mat2
-   mat3
-   mat4
-   vec2
-   vec2-ext
-   vec3
-   vec3-ext
-   vec4
-   vec4-ext
-   ivec2
-   ivec3
-   ivec4
-   color
-   plane
-   project
-   util
-   io
-   call
-   sphere
-   curve
-   bezier
-   version
-   ray

docs/source/api_call.rst

@@ -0,0 +1,11 @@
+Call API
+================================
+
+Call API is pre-built API for making calls from library. It is built on top of the array api. **glmc_** is the namespace for the call api. 
+**c** stands for call. 
+
+You need to built cglm to use call api. See build instructions (:doc:`build`) for more details.
+
+The functions except namespace **glmc_** are same as inline api. See ( :doc:`api_inline_array` ) for more details.
+
+📌 In the future we can define option to forward inline functions or struct api to call api.

docs/source/api_inline_array.rst

@@ -0,0 +1,76 @@
+Array API - Inline (Default)
+========================================
+
+This is the default API of *cglm*. All functions are forced to be inlined 
+and struct api, call api uses this inline api to share implementation.
+
+📌 Call api is also array api but it is not inlined. 
+In the future there may be option to forward struct api to call api instead of inline api to reduce binary size if needed.
+
+📌 **USE this API docs for similar functions in struct and call api** 
+
+📌 In struct api you can omit namespace e.g :code:`glms_vec3_dot` can be called as :code:`vec3_dot` in struct api, see :doc:`struct-api` to configure struct api for more details. 
+📌 In struct api functions can return struct/union
+📌 In struct api you can access items like **.x**, **.y**, **.z**, **.w**, **.r**, **.g**, **.b**, **.a**, **.m00**, **m01**... 
+
+Some functions may exist twice, once for their namespace and once for global namespace
+to make easier to write very common functions
+
+For instance, in general we use :code:`glm_vec3_dot` to get dot product
+of two **vec3**. Now we can also do this with :code:`glm_dot`,
+same for *_cross* and so on...
+
+The original function stays where it is, the function in global namespace
+of same name is just an alias, so there is no call version of those functions.
+e.g there is no func like :code:`glmc_dot` because *glm_dot* is just alias for
+:code:`glm_vec3_dot`
+
+By including **cglm/cglm.h** header you will include all inline version
+of functions. Since functions in this header[s] are inline you don't need to
+build or link *cglm* against your project.
+
+But by including **cglm/call.h** header you will include all *non-inline*
+version of functions. You need to build *cglm* and link it.
+Follow the :doc:`build` documentation for this
+
+.. toctree::
+   :maxdepth: 1
+   :caption: API categories:
+
+   affine
+   affine-mat
+   affine2d
+   cam
+   frustum
+   box
+   quat
+   euler
+   mat2
+   mat2x3
+   mat2x4
+   mat3
+   mat3x2
+   mat3x4
+   mat4
+   mat4x2
+   mat4x3
+   vec2
+   vec2-ext
+   vec3
+   vec3-ext
+   vec4
+   vec4-ext
+   ivec2
+   ivec3
+   ivec4
+   color
+   plane
+   project
+   util
+   io
+   call
+   sphere
+   curve
+   bezier
+   version
+   ray

docs/source/api_simd.rst

@@ -0,0 +1,12 @@
+SIMD API
+================================
+
+SIMD api is special api for SIMD operations. **glmm_** prefix is used for SIMD operations in cglm. It is used in many places in cglm.
+You can use it for your own SIMD operations too. In the future the api may be extended by time.
+
+Supported SIMD architectures ( may vary by time )
+
+* SSE / SSE2
+* AVX
+* NEON
+* WASM 128

docs/source/api_struct.rst

@@ -0,0 +1,98 @@
+Struct API 
+================================
+
+Struct API is alternative API to array api to use **cglm** with improved type safety and easy to use. 
+Since struct api is built top of array api, every struct API is not documented here. 
+See array api documentation for more information for individual functions.
+
+By default struct api adds `s` suffix to every type name e.g. vec3s, mat4s, versors etc. 
+Also struct api `s` suffix to namespace e.g. `glms_vec3_add`, `glms_mat4_mul` etc.
+
+By starting v0.9.0, struct api namespace is configurable. We can omit **glms_** namespace or 
+even change it with custom name to move existing api integrations to **cglm** more easliy...
+We can also add **s** to functin names if we want e.g. `glms_vec3_add()` -> `vec3_add()` or `vec3s_add()`.
+
+By including **cglm/struct.h** header you will include all struct api. It will also include **cglm/cglm.h** too. 
+Since struct apis are inline you don't need to build or link *cglm* against 
+your project unless if you want to use pre-built call-api too.
+
+Struct API is built top of array api. So you can mix them. 
+Use **.raw** union member to access raw array data to use it with array api. 
+
+Unlike array api ([0], [1], [0][0] ...), it is also possible to use struct api 
+with **.x**, **.y**, **.z**, **.w**, **.r**, **.g**, **.b**, **.a**, **.m00**, **m01**... 
+accessors to access individual elements/properties of vectors and matrices.
+
+Struct API usage:
+-----------------
+
+.. code-block:: c
+
+   #include <cglm/struct.h>
+   
+   mat4s m1  = glms_mat4_identity(); /* init...  */
+   mat4s m2  = glms_mat4_identity(); /* init...  */
+   mat4s m3  = glms_mat4_mul(glms_mat4_mul(m1, m2), glms_mat4_mul(m3, m4));
+   
+   vec3s v1  = { 1.0f, 0.0f, 0.0f };
+   vec3s v2  = { 0.0f, 1.0f, 0.0f };
+   vec4s v4  = { 0.0f, 1.0f, 0.0f, 0.0f };
+   vec4  v5a = { 0.0f, 1.0f, 0.0f, 0.0f };
+   
+   mat4s m4 = glms_rotate(m3, M_PI_2, 
+                          glms_vec3_cross(glms_vec3_add(v1, v6) 
+                                          glms_vec3_add(v1, v7)));
+
+   v1.x = 1.0f; v1.y = 0.0f; v1.z = 0.0f;
+   // or
+   v1.raw[0] = 1.0f; v1.raw[1] = 0.0f; v1.raw[2] = 0.0f;
+
+   /* use struct api with array api (mix them). */
+   /* use .raw to access array and use it with array api */
+
+   glm_vec4_add(m4.col[0].raw, v5a, m4.col[0].raw);
+   glm_mat4_mulv(m4.raw, v4.raw, v5a);
+
+or omit `glms_` namespace completely (see options below):
+
+.. code-block:: c
+
+   #define CGLM_OMIT_NS_FROM_STRUCT_API
+
+   #include <cglm/struct.h>
+
+   mat4s m1  = mat4_identity(); /* init...  */
+   mat4s m2  = mat4_identity(); /* init...  */
+   mat4s m3  = mat4_mul(mat4_mul(m1, m2), mat4_mul(m3, m4));
+   
+   vec3s v1  = { 1.0f, 0.0f, 0.0f };
+   vec3s v2  = { 0.0f, 1.0f, 0.0f };
+   vec4s v4  = { 0.0f, 1.0f, 0.0f, 0.0f };
+   vec4  v5a = { 0.0f, 1.0f, 0.0f, 0.0f };
+   
+   mat4s m4 = glms_rotate(m3, M_PI_2, 
+                          vec3_cross(vec3_add(v1, v6) 
+                                     vec3_add(v1, v7)));
+   
+   v1.x = 1.0f; v1.y = 0.0f; v1.z = 0.0f;
+   // or
+   v1.raw[0] = 1.0f; v1.raw[1] = 0.0f; v1.raw[2] = 0.0f;
+
+   /* use struct api with array api (mix them) */
+   glm_vec4_add(m4.col[0].raw, v5a, m4.col[0].raw);
+   glm_mat4_mulv(m4.raw, v4.raw, v5a);
+   
+Configuring the Struct API:
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+To configure the Struct API namespace, you can define the following macros before including the cglm/struct.h header:
+
+- **CGLM_OMIT_NS_FROM_STRUCT_API**: omits CGLM_STRUCT_API_NS (`glms_`) namespace completely if there is sub namespace e.g `mat4_`, `vec4_` ... DEFAULT is not defined
+- **CGLM_STRUCT_API_NS**:           define name space for struct api, DEFAULT is **glms**
+- **CGLM_STRUCT_API_NAME_SUFFIX**:  define name suffix, DEFAULT is **empty** e.g defining it as #define CGLM_STRUCT_API_NAME_SUFFIX  s will add s suffix to mat4_mul -> mat4s_mul
+
+
+Detailed documentation for Struct API:
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Since struct api if built top of array api, see array api functions for more information about individual functions.

docs/source/cam.rst

@@ -101,7 +101,7 @@ Functions documentation
     Parameters:
       | *[in]*  **box**      AABB
       | *[in]*  **padding**  padding
-      | *[out]* **d**        result matrix
+      | *[out]* **dest**        result matrix
 
 .. c:function:: void  glm_ortho_aabb_pz(vec3 box[2], float padding, mat4 dest)
 
@@ -113,7 +113,7 @@ Functions documentation
     Parameters:
       | *[in]*  **box**      AABB
       | *[in]*  **padding**  padding for near and far
-      | *[out]* **d**        result matrix
+      | *[out]* **dest**        result matrix
 
     Returns:
       square of norm / magnitude

docs/source/conf.py

@@ -62,16 +62,16 @@ author = u'Recep Aslantas'
 # built documents.
 #
 # The short X.Y version.
-version = u'0.8.9'
+version = u'0.9.2'
 # The full version, including alpha/beta/rc tags.
-release = u'0.8.9'
+release = u'0.9.2'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.
 #
 # This is also used if you do content translation via gettext catalogs.
 # Usually you set "language" from the command line for these cases.
-language = None
+language = 'en'
 
 # List of patterns, relative to source directory, that match files and
 # directories to ignore when looking for source files.
@@ -111,7 +111,7 @@ html_theme_options = {
 # Add any paths that contain custom static files (such as style sheets) here,
 # relative to this directory. They are copied after the builtin static files,
 # so a file named "default.css" will overwrite the builtin "default.css".
-html_static_path = ['_static']
+# html_static_path = ['_static']
 
 
 # -- Options for HTMLHelp output ------------------------------------------

docs/source/features.rst

@@ -1,7 +1,7 @@
 Features
 ================================================================================
 
-* **scalar** and **simd** (sse, avx, neon...) optimizations
+* **scalar** and **simd** (sse, avx, neon, wasm...) optimizations
 * option to use different clipspaces e.g. Left Handed, Zero-to-One... (currrently right handed negative-one is default)
 * array api and struct api, you can use arrays or structs.
 * general purpose matrix operations (mat4, mat3)
@@ -18,6 +18,7 @@ Features
 * inline or pre-compiled function call
 * frustum (extract view frustum planes, corners...)
 * bounding box (AABB in Frustum (culling), crop, merge...)
+* 2d bounding box (crop, merge...)
 * bounding sphere
 * project, unproject
 * easing functions

docs/source/index.rst

@@ -28,7 +28,7 @@ considered to be supported as optional.
    opengl
 
 .. toctree::
-   :maxdepth: 2
+   :maxdepth: 3
    :caption: API:
 
    api

docs/source/io.rst

@@ -53,8 +53,11 @@ Functions:
 1. :c:func:`glm_mat4_print`
 #. :c:func:`glm_mat3_print`
 #. :c:func:`glm_vec4_print`
+#. :c:func:`glm_ivec4_print`
 #. :c:func:`glm_vec3_print`
 #. :c:func:`glm_ivec3_print`
+#. :c:func:`glm_vec2_print`
+#. :c:func:`glm_ivec2_print`
 #. :c:func:`glm_versor_print`
 #. :c:func:`glm_aabb_print`
 
@@ -63,7 +66,7 @@ Functions documentation
 
 .. c:function:: void  glm_mat4_print(mat4 matrix, FILE * __restrict ostream)
 
-    | print mat4 to given stream
+    | print matrix to given stream
 
     Parameters:
       | *[in]*  **matrix**   matrix
@@ -71,7 +74,7 @@ Functions documentation
 
 .. c:function:: void  glm_mat3_print(mat3 matrix, FILE * __restrict ostream)
 
-    | print mat3 to given stream
+    | print matrix to given stream
 
     Parameters:
       | *[in]*  **matrix**   matrix
@@ -79,7 +82,15 @@ Functions documentation
 
 .. c:function:: void  glm_vec4_print(vec4 vec, FILE * __restrict ostream)
 
-    | print vec4 to given stream
+    | print vector to given stream
+
+    Parameters:
+      | *[in]*  **vec**      vector
+      | *[in]*  **ostream**  FILE to write
+
+.. c:function:: void  glm_ivec4_print(ivec4 vec, FILE * __restrict ostream)
+
+    | print vector to given stream
 
     Parameters:
       | *[in]*  **vec**      vector
@@ -87,7 +98,7 @@ Functions documentation
 
 .. c:function:: void  glm_vec3_print(vec3 vec, FILE * __restrict ostream)
 
-    | print vec3 to given stream
+    | print vector to given stream
 
     Parameters:
       | *[in]*  **vec**      vector
@@ -95,12 +106,29 @@ Functions documentation
 
 .. c:function:: void  glm_ivec3_print(ivec3 vec, FILE * __restrict ostream)
 
-    | print ivec3 to given stream
+    | print vector to given stream
 
     Parameters:
       | *[in]*  **vec**      vector
       | *[in]*  **ostream**  FILE to write
 
+.. c:function:: void  glm_vec2_print(vec2 vec, FILE * __restrict ostream)
+
+    | print vector to given stream
+
+    Parameters:
+      | *[in]*  **vec**      vector
+      | *[in]*  **ostream**  FILE to write
+
+.. c:function:: void  glm_ivec2_print(ivec2 vec, FILE * __restrict ostream)
+
+    | print vector to given stream
+
+    Parameters:
+      | *[in]*  **vec**      vector
+      | *[in]*  **ostream**  FILE to write
+
+
 .. c:function:: void  glm_versor_print(versor vec, FILE * __restrict ostream)
 
     | print quaternion to given stream

docs/source/ivec2.rst

@@ -143,6 +143,31 @@ Functions documentation
     Returns:
         distance
 
+.. c:function:: void  glm_ivec2_fill(ivec2 v, int val)
+
+    fill a vector with specified value
+
+    Parameters:
+      | *[out]*  **v**    vector
+      | *[in]*   **val**  value
+
+.. c:function:: bool  glm_ivec2_eq(ivec2 v, int val)
+
+    check if vector is equal to value
+
+    Parameters:
+      | *[in]*  **v**    vector
+      | *[in]*  **val**  value
+
+.. c:function:: bool  glm_ivec2_eqv(ivec2 v1, ivec2 v2)
+
+    check if vector is equal to another vector
+
+    Parameters:
+      | *[in]*  **vec**   vector 1
+      | *[in]*  **vec**   vector 2
+
+
 .. c:function:: void glm_ivec2_maxv(ivec2 a, ivec2 b, ivec2 dest)
 
     set each member of dest to greater of vector a and b

docs/source/ivec3.rst

@@ -29,6 +29,9 @@ Functions:
 #. :c:func:`glm_ivec3_scale`
 #. :c:func:`glm_ivec3_distance2`
 #. :c:func:`glm_ivec3_distance`
+#. :c:func:`glm_ivec3_fill`
+#. :c:func:`glm_ivec3_eq`
+#. :c:func:`glm_ivec3_eqv`
 #. :c:func:`glm_ivec3_maxv`
 #. :c:func:`glm_ivec3_minv`
 #. :c:func:`glm_ivec3_clamp`
@@ -143,6 +146,30 @@ Functions documentation
     Returns:
         distance
 
+.. c:function:: void  glm_ivec3_fill(ivec3 v, int val)
+
+    fill a vector with specified value
+
+    Parameters:
+      | *[out]*  **v**    vector
+      | *[in]*   **val**  value
+
+.. c:function:: bool  glm_ivec3_eq(ivec3 v, int val)
+
+    check if vector is equal to value
+
+    Parameters:
+      | *[in]*  **v**    vector
+      | *[in]*  **val**  value
+
+.. c:function:: bool  glm_ivec3_eqv(ivec3 v1, ivec3 v2)
+
+    check if vector is equal to another vector
+
+    Parameters:
+      | *[in]*  **vec**   vector 1
+      | *[in]*  **vec**   vector 2
+
 .. c:function:: void glm_ivec3_maxv(ivec3 a, ivec3 b, ivec3 dest)
 
     set each member of dest to greater of vector a and b

docs/source/mat2.rst

@@ -32,6 +32,7 @@ Functions:
 #. :c:func:`glm_mat2_swap_col`
 #. :c:func:`glm_mat2_swap_row`
 #. :c:func:`glm_mat2_rmc`
+#. :c:func:`glm_mat2_make`
 
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
@@ -64,11 +65,12 @@ Functions documentation
     make given matrix zero
 
     Parameters:
-      | *[in,out]* **mat**  matrix to
+      | *[in,out]* **mat**  matrix
 
 .. c:function:: void glm_mat2_mul(mat2 m1, mat2 m2, mat2 dest)
 
     multiply m1 and m2 to dest
+
     m1, m2 and dest matrices can be same matrix, it is possible to write this:
 
     .. code-block:: c
@@ -100,7 +102,7 @@ Functions documentation
 
 .. c:function:: void glm_mat2_mulv(mat2 m, vec2 v, vec2 dest)
 
-    multiply mat4 with vec4 (column vector) and store in dest vector
+    multiply mat2 with vec2 (column vector) and store in dest vector
 
     Parameters:
       | *[in]*  **mat**   mat2 (left)
@@ -112,8 +114,8 @@ Functions documentation
     multiply matrix with scalar
 
     Parameters:
-      | *[in, out]* **mat**   matrix
+      | *[in, out]* **m**  matrix
-      | *[in]*      **dest**  scalar
+      | *[in]*      **s**  scalar
 
 .. c:function:: float  glm_mat2_det(mat2 mat)
 
@@ -177,3 +179,13 @@ Functions documentation
 
     Returns:
         scalar value e.g. Matrix1x1
+
+.. c:function:: void glm_mat2_make(float * __restrict src, mat2 dest)
+
+    Create mat2 matrix from pointer
+
+    | NOTE: **@src** must contain at least 4 elements.
+
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination matrix2x2

docs/source/mat2x3.rst

@@ -0,0 +1,90 @@
+.. default-domain:: C
+
+mat2x3
+======
+
+Header: cglm/mat2x3.h
+
+Table of contents (click to go):
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Macros:
+
+1. GLM_MAT2X3_ZERO_INIT
+#. GLM_MAT2X3_ZERO
+
+Functions:
+
+1. :c:func:`glm_mat2x3_copy`
+#. :c:func:`glm_mat2x3_zero`
+#. :c:func:`glm_mat2x3_make`
+#. :c:func:`glm_mat2x3_mul`
+#. :c:func:`glm_mat2x3_mulv`
+#. :c:func:`glm_mat2x3_transpose`
+#. :c:func:`glm_mat2x3_scale`
+
+Functions documentation
+~~~~~~~~~~~~~~~~~~~~~~~
+
+.. c:function:: void glm_mat2x3_copy(mat2x3 mat, mat2x3 dest)
+
+    copy mat2x3 to another one (dest).
+
+    Parameters:
+      | *[in]*  **mat**   source
+      | *[out]* **dest**  destination
+
+.. c:function:: void glm_mat2x3_zero(mat2x3 mat)
+
+    make given matrix zero
+
+    Parameters:
+      | *[in,out]* **mat**  matrix
+
+.. c:function:: void glm_mat2x3_make(float * __restrict src, mat2x3 dest)
+
+    Create mat2x3 matrix from pointer
+
+    | NOTE: **@src** must contain at least 6 elements.
+
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination matrix2x3
+
+.. c:function:: void glm_mat2x3_mul(mat2x3 m1, mat3x2 m2, mat2 dest)
+
+    multiply m1 and m2 to dest
+
+    .. code-block:: c
+
+       glm_mat2x3_mul(mat2x3, mat3x2, mat2);
+
+    Parameters:
+      | *[in]*  **m1**    left matrix (mat2x3)
+      | *[in]*  **m2**    right matrix (mat3x2)
+      | *[out]* **dest**  destination matrix (mat2)
+
+.. c:function:: void glm_mat2x3_mulv(mat2x3 m, vec3 v, vec2 dest)
+
+    multiply mat2x3 with vec3 (column vector) and store in dest vector
+
+    Parameters:
+      | *[in]*  **m**     mat2x3 (left)
+      | *[in]*  **v**     vec3 (right, column vector)
+      | *[out]* **dest**  destination (result, column vector)
+
+.. c:function:: void glm_mat2x3_transpose(mat2x3 m, mat3x2 dest)
+
+    transpose matrix and store in dest
+
+    Parameters:
+      | *[in]*  **m**     matrix
+      | *[out]* **dest**  destination
+
+.. c:function:: void  glm_mat2x3_scale(mat2x3 m, float s)
+
+    multiply matrix with scalar
+
+    Parameters:
+      | *[in, out]* **m** matrix
+      | *[in]*      **s** scalar

docs/source/mat2x4.rst

@@ -0,0 +1,90 @@
+.. default-domain:: C
+
+mat2x4
+======
+
+Header: cglm/mat2x4.h
+
+Table of contents (click to go):
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Macros:
+
+1. GLM_MAT2X4_ZERO_INIT
+#. GLM_MAT2X4_ZERO
+
+Functions:
+
+1. :c:func:`glm_mat2x4_copy`
+#. :c:func:`glm_mat2x4_zero`
+#. :c:func:`glm_mat2x4_make`
+#. :c:func:`glm_mat2x4_mul`
+#. :c:func:`glm_mat2x4_mulv`
+#. :c:func:`glm_mat2x4_transpose`
+#. :c:func:`glm_mat2x4_scale`
+
+Functions documentation
+~~~~~~~~~~~~~~~~~~~~~~~
+
+.. c:function:: void glm_mat2x4_copy(mat2x4 mat, mat2x4 dest)
+
+    copy mat2x4 to another one (dest).
+
+    Parameters:
+      | *[in]*  **mat**   source
+      | *[out]* **dest**  destination
+
+.. c:function:: void glm_mat2x4_zero(mat2x4 mat)
+
+    make given matrix zero
+
+    Parameters:
+      | *[in,out]* **mat**  matrix
+
+.. c:function:: void glm_mat2x4_make(float * __restrict src, mat2x4 dest)
+
+    Create mat2x4 matrix from pointer
+
+    | NOTE: **@src** must contain at least 8 elements.
+
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination matrix2x4
+
+.. c:function:: void glm_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat2 dest)
+
+    multiply m1 and m2 to dest
+
+    .. code-block:: c
+
+       glm_mat2x4_mul(mat2x4, mat4x2, mat2);
+
+    Parameters:
+      | *[in]*  **m1**    left matrix (mat2x4)
+      | *[in]*  **m2**    right matrix (mat4x2)
+      | *[out]* **dest**  destination matrix (mat2)
+
+.. c:function:: void glm_mat2x4_mulv(mat2x4 m, vec4 v, vec2 dest)
+
+    multiply mat2x4 with vec4 (column vector) and store in dest vector
+
+    Parameters:
+      | *[in]*  **m**     mat2x4 (left)
+      | *[in]*  **v**     vec4 (right, column vector)
+      | *[out]* **dest**  destination (result, column vector)
+
+.. c:function:: void glm_mat2x4_transpose(mat2x4 m, mat4x2 dest)
+
+    transpose matrix and store in dest
+
+    Parameters:
+      | *[in]*  **m**     matrix
+      | *[out]* **dest**  destination
+
+.. c:function:: void  glm_mat2x4_scale(mat2x4 m, float s)
+
+    multiply matrix with scalar
+
+    Parameters:
+      | *[in, out]* **m** matrix
+      | *[in]*      **s** scalar

docs/source/mat3.rst

@@ -34,6 +34,7 @@ Functions:
 #. :c:func:`glm_mat3_swap_col`
 #. :c:func:`glm_mat3_swap_row`
 #. :c:func:`glm_mat3_rmc`
+#. :c:func:`glm_mat3_make`
 
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
@@ -71,6 +72,7 @@ Functions documentation
 .. c:function:: void  glm_mat3_mul(mat3 m1, mat3 m2, mat3 dest)
 
     multiply m1 and m2 to dest
+
     m1, m2 and dest matrices can be same matrix, it is possible to write this:
 
     .. code-block:: c
@@ -102,10 +104,10 @@ Functions documentation
 
 .. c:function:: void  glm_mat3_mulv(mat3 m, vec3 v, vec3 dest)
 
-    multiply mat4 with vec4 (column vector) and store in dest vector
+    multiply mat3 with vec3 (column vector) and store in dest vector
 
     Parameters:
-      | *[in]*  **mat**   mat3 (left)
+      | *[in]*  **m**     mat3 (left)
       | *[in]*  **v**     vec3 (right, column vector)
       | *[out]* **dest**  destination (result, column vector)
 
@@ -122,8 +124,8 @@ Functions documentation
     multiply matrix with scalar
 
     Parameters:
-      | *[in, out]* **mat**   matrix
+      | *[in, out]* **m** matrix
-      | *[in]*      **dest**  scalar
+      | *[in]*      **s** scalar
 
 .. c:function:: float  glm_mat3_det(mat3 mat)
 
@@ -187,3 +189,13 @@ Functions documentation
 
     Returns:
         scalar value e.g. Matrix1x1
+
+.. c:function:: void glm_mat3_make(float * __restrict src, mat3 dest)
+
+    Create mat3 matrix from pointer
+
+    | NOTE: **@src** must contain at least 9 elements.
+
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination matrix3x3

docs/source/mat3x2.rst

@@ -0,0 +1,89 @@
+.. default-domain:: C
+
+mat3x2
+======
+
+Header: cglm/mat3x2.h
+
+Table of contents (click to go):
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Macros:
+
+1. GLM_MAT3X2_ZERO_INIT
+#. GLM_MAT3X2_ZERO
+
+Functions:
+
+1. :c:func:`glm_mat3x2_copy`
+#. :c:func:`glm_mat3x2_zero`
+#. :c:func:`glm_mat3x2_make`
+#. :c:func:`glm_mat3x2_mul`
+#. :c:func:`glm_mat3x2_mulv`
+#. :c:func:`glm_mat3x2_transpose`
+#. :c:func:`glm_mat3x2_scale`
+
+Functions documentation
+~~~~~~~~~~~~~~~~~~~~~~~
+
+.. c:function:: void glm_mat3x2_copy(mat3x2 mat, mat3x2 dest)
+
+    copy mat3x2 to another one (dest).
+
+    Parameters:
+      | *[in]*  **mat**   source
+      | *[out]* **dest**  destination
+
+.. c:function:: void glm_mat3x2_zero(mat3x2 mat)
+
+    make given matrix zero
+
+    Parameters:
+      | *[in,out]* **mat**  matrix
+
+.. c:function:: void glm_mat3x2_make(float * __restrict src, mat3x2 dest)
+
+    Create mat3x2 matrix from pointer
+
+    | NOTE: **@src** must contain at least 6 elements.
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination matrix3x2
+
+.. c:function:: void glm_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat3 dest)
+
+    multiply m1 and m2 to dest
+
+    .. code-block:: c
+
+       glm_mat3x2_mul(mat3x2, mat2x3, mat3);
+
+    Parameters:
+      | *[in]*  **m1**    left matrix (mat3x2)
+      | *[in]*  **m2**    right matrix (mat2x3)
+      | *[out]* **dest**  destination matrix (mat3)
+
+.. c:function:: void glm_mat3x2_mulv(mat3x2 m, vec2 v, vec3 dest)
+
+    multiply mat3x2 with vec2 (column vector) and store in dest vector
+
+    Parameters:
+      | *[in]*  **m**     mat3x2 (left)
+      | *[in]*  **v**     vec3 (right, column vector)
+      | *[out]* **dest**  destination (result, column vector)
+
+.. c:function:: void glm_mat3x2_transpose(mat3x2 m, mat2x3 dest)
+
+    transpose matrix and store in dest
+
+    Parameters:
+      | *[in]*  **m**     matrix
+      | *[out]* **dest**  destination
+
+.. c:function:: void  glm_mat3x2_scale(mat3x2 m, float s)
+
+    multiply matrix with scalar
+
+    Parameters:
+      | *[in, out]* **m** matrix
+      | *[in]*      **s** scalar

docs/source/mat3x4.rst

@@ -0,0 +1,89 @@
+.. default-domain:: C
+
+mat3x4
+======
+
+Header: cglm/mat3x4.h
+
+Table of contents (click to go):
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Macros:
+
+1. GLM_MAT3X4_ZERO_INIT
+#. GLM_MAT3X4_ZERO
+
+Functions:
+
+1. :c:func:`glm_mat3x4_copy`
+#. :c:func:`glm_mat3x4_zero`
+#. :c:func:`glm_mat3x4_make`
+#. :c:func:`glm_mat3x4_mul`
+#. :c:func:`glm_mat3x4_mulv`
+#. :c:func:`glm_mat3x4_transpose`
+#. :c:func:`glm_mat3x4_scale`
+
+Functions documentation
+~~~~~~~~~~~~~~~~~~~~~~~
+
+.. c:function:: void glm_mat3x4_copy(mat3x4 mat, mat3x4 dest)
+
+    copy mat3x4 to another one (dest).
+
+    Parameters:
+      | *[in]*  **mat**   source
+      | *[out]* **dest**  destination
+
+.. c:function:: void glm_mat3x4_zero(mat3x4 mat)
+
+    make given matrix zero
+
+    Parameters:
+      | *[in,out]* **mat**  matrix
+
+.. c:function:: void glm_mat3x4_make(float * __restrict src, mat3x4 dest)
+
+    Create mat3x4 matrix from pointer
+
+    | NOTE: **@src** must contain at least 12 elements.
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination matrix3x4
+
+.. c:function:: void glm_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat3 dest)
+
+    multiply m1 and m2 to dest
+
+    .. code-block:: c
+
+       glm_mat3x4_mul(mat3x4, mat4x3, mat3);
+
+    Parameters:
+      | *[in]*  **m1**    left matrix (mat3x4)
+      | *[in]*  **m2**    right matrix (mat4x3)
+      | *[out]* **dest**  destination matrix (mat3)
+
+.. c:function:: void glm_mat3x4_mulv(mat3x4 m, vec4 v, vec3 dest)
+
+    multiply mat3x4 with vec4 (column vector) and store in dest vector
+
+    Parameters:
+      | *[in]*  **m**     mat3x4 (left)
+      | *[in]*  **v**     vec4 (right, column vector)
+      | *[out]* **dest**  destination (result, column vector)
+
+.. c:function:: void glm_mat3x4_transpose(mat3x4 m, mat4x3 dest)
+
+    transpose matrix and store in dest
+
+    Parameters:
+      | *[in]*  **m**     matrix
+      | *[out]* **dest**  destination
+
+.. c:function:: void  glm_mat3x4_scale(mat3x4 m, float s)
+
+    multiply matrix with scalar
+
+    Parameters:
+      | *[in, out]* **m** matrix
+      | *[in]*      **s** scalar

docs/source/mat4.rst

@@ -47,6 +47,7 @@ Functions:
 #. :c:func:`glm_mat4_swap_col`
 #. :c:func:`glm_mat4_swap_row`
 #. :c:func:`glm_mat4_rmc`
+#. :c:func:`glm_mat4_make`
 
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
@@ -118,6 +119,7 @@ Functions documentation
 .. c:function:: void  glm_mat4_mul(mat4 m1, mat4 m2, mat4 dest)
 
     multiply m1 and m2 to dest
+
     m1, m2 and dest matrices can be same matrix, it is possible to write this:
 
     .. code-block:: c
@@ -156,7 +158,6 @@ Functions documentation
     Parameters:
       | *[in]*  **m**     mat4 (left)
       | *[in]*  **v**     vec4 (right, column vector)
-      | *[in]*  **last**  4th item to make it vec4
       | *[out]* **dest**  vec4 (result, column vector)
 
 .. c:function:: void  glm_mat4_mulv3(mat4 m, vec3 v, float last, vec3 dest)
@@ -302,3 +303,13 @@ Functions documentation
 
     Returns:
         scalar value e.g. Matrix1x1
+
+.. c:function:: void glm_mat4_make(float * __restrict src, mat4 dest)
+
+    Create mat4 matrix from pointer
+
+    | NOTE: **@src** must contain at least 16 elements.
+
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination matrix4x4

docs/source/mat4x2.rst

@@ -0,0 +1,89 @@
+.. default-domain:: C
+
+mat4x2
+======
+
+Header: cglm/mat4x2.h
+
+Table of contents (click to go):
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Macros:
+
+1. GLM_MAT4X2_ZERO_INIT
+#. GLM_MAT4X2_ZERO
+
+Functions:
+
+1. :c:func:`glm_mat4x2_copy`
+#. :c:func:`glm_mat4x2_zero`
+#. :c:func:`glm_mat4x2_make`
+#. :c:func:`glm_mat4x2_mul`
+#. :c:func:`glm_mat4x2_mulv`
+#. :c:func:`glm_mat4x2_transpose`
+#. :c:func:`glm_mat4x2_scale`
+
+Functions documentation
+~~~~~~~~~~~~~~~~~~~~~~~
+
+.. c:function:: void glm_mat4x2_copy(mat4x2 mat, mat4x2 dest)
+
+    copy mat4x2 to another one (dest).
+
+    Parameters:
+      | *[in]*  **mat**   source
+      | *[out]* **dest**  destination
+
+.. c:function:: void glm_mat4x2_zero(mat4x2 mat)
+
+    make given matrix zero
+
+    Parameters:
+      | *[in,out]* **mat**  matrix
+
+.. c:function:: void glm_mat4x2_make(float * __restrict src, mat4x2 dest)
+
+    Create mat4x2 matrix from pointer
+
+    | NOTE: **@src** must contain at least 8 elements.
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination matrix4x2
+
+.. c:function:: void glm_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat4 dest)
+
+    multiply m1 and m2 to dest
+
+    .. code-block:: c
+
+       glm_mat4x2_mul(mat4x2, mat2x4, mat4);
+
+    Parameters:
+      | *[in]*  **m1**    left matrix (mat4x2)
+      | *[in]*  **m2**    right matrix (mat2x4)
+      | *[out]* **dest**  destination matrix (mat4)
+
+.. c:function:: void glm_mat4x2_mulv(mat4x2 m, vec2 v, vec4 dest)
+
+    multiply mat4x2 with vec2 (column vector) and store in dest vector
+
+    Parameters:
+      | *[in]*  **m**     mat4x2 (left)
+      | *[in]*  **v**     vec2 (right, column vector)
+      | *[out]* **dest**  destination (result, column vector)
+
+.. c:function:: void glm_mat4x2_transpose(mat4x2 m, mat2x4 dest)
+
+    transpose matrix and store in dest
+
+    Parameters:
+      | *[in]*  **m**     matrix
+      | *[out]* **dest**  destination
+
+.. c:function:: void  glm_mat4x2_scale(mat4x2 m, float s)
+
+    multiply matrix with scalar
+
+    Parameters:
+      | *[in, out]* **m** matrix
+      | *[in]*      **s** scalar

docs/source/mat4x3.rst

@@ -0,0 +1,89 @@
+.. default-domain:: C
+
+mat4x3
+======
+
+Header: cglm/mat4x3.h
+
+Table of contents (click to go):
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Macros:
+
+1. GLM_MAT4X3_ZERO_INIT
+#. GLM_MAT4X3_ZERO
+
+Functions:
+
+1. :c:func:`glm_mat4x3_copy`
+#. :c:func:`glm_mat4x3_zero`
+#. :c:func:`glm_mat4x3_make`
+#. :c:func:`glm_mat4x3_mul`
+#. :c:func:`glm_mat4x3_mulv`
+#. :c:func:`glm_mat4x3_transpose`
+#. :c:func:`glm_mat4x3_scale`
+
+Functions documentation
+~~~~~~~~~~~~~~~~~~~~~~~
+
+.. c:function:: void glm_mat4x3_copy(mat4x3 mat, mat4x3 dest)
+
+    copy mat4x3 to another one (dest).
+
+    Parameters:
+      | *[in]*  **mat**   source
+      | *[out]* **dest**  destination
+
+.. c:function:: void glm_mat4x3_zero(mat4x3 mat)
+
+    make given matrix zero
+
+    Parameters:
+      | *[in,out]* **mat**  matrix
+
+.. c:function:: void glm_mat4x3_make(float * __restrict src, mat4x3 dest)
+
+    Create mat4x3 matrix from pointer
+
+    | NOTE: **@src** must contain at least 12 elements.
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination matrix4x3
+
+.. c:function:: void glm_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat4 dest)
+
+    multiply m1 and m2 to dest
+
+    .. code-block:: c
+
+       glm_mat4x3_mul(mat4x3, mat3x4, mat4);
+
+    Parameters:
+      | *[in]*  **m1**    left matrix (mat4x3)
+      | *[in]*  **m2**    right matrix (mat3x4)
+      | *[out]* **dest**  destination matrix (mat4)
+
+.. c:function:: void glm_mat4x3_mulv(mat4x3 m, vec3 v, vec4 dest)
+
+    multiply mat4x3 with vec3 (column vector) and store in dest vector
+
+    Parameters:
+      | *[in]*  **m**     mat4x3 (left)
+      | *[in]*  **v**     vec3 (right, column vector)
+      | *[out]* **dest**  destination (result, column vector)
+
+.. c:function:: void glm_mat4x3_transpose(mat4x3 m, mat3x4 dest)
+
+    transpose matrix and store in dest
+
+    Parameters:
+      | *[in]*  **m**     matrix
+      | *[out]* **dest**  destination
+
+.. c:function:: void  glm_mat4x3_scale(mat4x3 m, float s)
+
+    multiply matrix with scalar
+
+    Parameters:
+      | *[in, out]* **m** matrix
+      | *[in]*      **s** scalar

docs/source/opt.rst

@@ -1,6 +1,6 @@
 .. default-domain:: C
 
-Options
+🛠️ Options
 ===============================================================================
 
 A few options are provided via macros.
@@ -90,6 +90,16 @@ You have to extra options for dot product: **CGLM_SSE4_DOT** and **CGLM_SSE3_DOT
 
 otherwise cglm will use custom cglm's hadd functions which are optimized too.
 
+Struct API Options
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+To configure the Struct API namespace, you can define the following macros before including the cglm/struct.h header:
+
+- **CGLM_OMIT_NS_FROM_STRUCT_API**: omits CGLM_STRUCT_API_NS (`glms_`) namespace completely if there is sub namespace e.g `mat4_`, `vec4_` ... DEFAULT is not defined
+- **CGLM_STRUCT_API_NS**:           define name space for struct api, DEFAULT is **glms**
+- **CGLM_STRUCT_API_NAME_SUFFIX**:  define name suffix, DEFAULT is **empty** e.g defining it as #define CGLM_STRUCT_API_NAME_SUFFIX  s will add s suffix to mat4_mul -> mat4s_mul
+
+
 Print Options
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 

docs/source/quat.rst

@@ -62,6 +62,7 @@ Functions:
 #. :c:func:`glm_quat_rotate`
 #. :c:func:`glm_quat_rotate_at`
 #. :c:func:`glm_quat_rotate_atm`
+#. :c:func:`glm_quat_make`
 
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
@@ -420,3 +421,13 @@ Functions documentation
       | *[in, out]*  **m**      existing transform matrix to rotate
       | *[in]*       **q**      quaternion
       | *[in]*       **pivot**  pivot
+
+.. c:function:: void glm_quat_make(float * __restrict src, versor dest)
+
+    Create quaternion from pointer
+
+    | NOTE: **@src** must contain at least 4 elements. cglm store quaternions as [x, y, z, w].
+
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination quaternion

docs/source/troubleshooting.rst

@@ -3,30 +3,30 @@
 Troubleshooting
 ================================================================================
 
-It is possible that sometimes you may get crashes or wrong results.
+It is possible that you may sometimes get crashes or wrong results.
 Follow these topics
 
 Memory Allocation:
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-Again, **cglm** doesn't alloc any memory on heap.
+Recall that **cglm** does not alloc any memory on the heap.
-cglm functions works like memcpy; it copies data from src,
+cglm functions work like memcpy; they copy data from src,
-makes calculations then copy the result to dest.
+make calculations, then copy the result to dest.
 
 You are responsible for allocation of **src** and **dest** parameters.
 
 Alignment:
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-**vec4** and **mat4** types requires 16 byte alignment.
+**vec4** and **mat4** types require 16 byte alignment.
-These types are marked with align attribute to let compiler know about this
+These types are marked with the align attribute to let the compiler know about this
 requirement.
 
-But since MSVC (Windows) throws the error:
+Since MSVC (Windows) throws this error:
 
 **"formal parameter with requested alignment of 16 won't be aligned"**
 
-The alignment attribute has been commented for MSVC
+The alignment attribute has been commented out for MSVC
 
 .. code-block:: c
 
@@ -37,61 +37,61 @@ The alignment attribute has been commented for MSVC
    #endif.
 
 So MSVC may not know about alignment requirements when creating variables.
-The interesting thing is that, if I remember correctly Visual Studio 2017
+The interesting thing is that, if I remember correctly, Visual Studio 2017
 doesn't throw the above error. So we may uncomment that line for Visual Studio 2017,
 you may do it yourself.
 
 **This MSVC issue is still in TODOs.**
 
-**UPDATE:** By starting v0.4.5 cglm provides an option to disable alignment requirement.
+**UPDATE:** Starting with v0.4.5, cglm provides an option to disable the alignment requirement.
-Also alignment is disabled for older msvc verisons as default. Now alignment is only required in Visual Studio 2017 version 15.6+ if CGLM_ALL_UNALIGNED macro is not defined.
+Also, alignment is disabled for older msvc versions by default. Now alignment is only required in Visual Studio 2017 version 15.6+ if the CGLM_ALL_UNALIGNED macro is not defined.
 
 Crashes, Invalid Memory Access:
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-Probably you are trying to write to invalid memory location.
+Most likely, you are trying to write to an invalid memory location.
 
-You may used wrong function for what you want to do.
+You may have used a wrong function for what you want to do.
 
-For instance you may called **glm_vec4_** functions for **vec3** data type.
+For example, you may have called a **glm_vec4_** function for a **vec3** data type.
-It will try to write 32 byte but since **vec3** is 24 byte it should throw
+It will try to write 32 bytes, but since **vec3** is 24 bytes, it should throw
-memory access error or exit the app without saying anything.
+a memory access error or exit the app without saying anything.
 
 **UPDATE - IMPORTANT:** 
 
   | On MSVC or some other compilers, if alignment is enabled (default) then double check alignment requirements if you got a crash.
 
-  | If you send GLM_VEC4_ONE or similar macros directly to a function, it may be crashed.
+  | If you send GLM_VEC4_ONE or similar macros directly to a function, it may crash.
-  | Because compiler may not apply alignment as defined on **typedef** to that macro while passing it (on stack) to a function.
+  | Because the compiler may not apply alignment as defined on **typedef** to that macro while passing it (on stack) to a function.
 
 Wrong Results:
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-Again, you may used wrong function.
+Again, you may have used a wrong function.
 
-For instance if you use **glm_normalize()** or **glm_vec3_normalize()** for **vec4**,
+For instance if you use **glm_normalize()** or **glm_vec3_normalize()** for a **vec4**,
-it will assume that passed param is **vec3** and will normalize it for **vec3**.
+it will assume that the passed param is a **vec3**, and will normalize it for **vec3**.
-Since you need to **vec4** to be normalized in your case, you will get wrong results.
+Since you need a **vec4** to be normalized in your case, you will get wrong results.
 
-Accessing vec4 type with vec3 functions is valid, you will not get any error, exception or crash.
+Accessing a vec4 type with vec3 functions is valid, you will not get any error, exception or crash.
 You only get wrong results if you don't know what you are doing!
 
-So be carefull, when your IDE (Xcode, Visual Studio ...) tried to autocomplete function names, READ IT :)
+So be careful, when your IDE (Xcode, Visual Studio ...) tries to autocomplete function names, READ IT :)
 
-**Also implementation may be wrong please let us know by creating an issue on Github.**
+**Also implementation may be wrong, please let us know by creating an issue on Github.**
 
 BAD_ACCESS : Thread 1: EXC_BAD_ACCESS (code=EXC_I386_GPFLT) or Similar Errors/Crashes
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-This is similar issue with alignment. For instance if you compiled **cglm** with 
+This is a similar issue with alignment. For instance if you compiled **cglm** with 
 AVX (**-mavx**, intentionally or not) and if you use **cglm** in an environment that doesn't 
 support AVX (or if AVX is disabled intentionally) e.g. environment that max support SSE2/3/4, 
 then you probably get **BAD ACCESS** or similar...
 
 Because if you compile **cglm** with AVX it aligns **mat4** with 32 byte boundary, 
-and your project aligns that as 16 byte boundary...
+and your project aligns that as a 16 byte boundary...
 
-Check alignment, supported vector extension or simd in **cglm** and linked projects...
+Check alignment, supported vector extension, or simd in **cglm** and linked projects...
 
 Other Issues?
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

docs/source/vec2.rst

@@ -45,12 +45,14 @@ Functions:
 #. :c:func:`glm_vec2_normalize`
 #. :c:func:`glm_vec2_normalize_to`
 #. :c:func:`glm_vec2_rotate`
+#. :c:func:`glm_vec2_center`
 #. :c:func:`glm_vec2_distance2`
 #. :c:func:`glm_vec2_distance`
 #. :c:func:`glm_vec2_maxv`
 #. :c:func:`glm_vec2_minv`
 #. :c:func:`glm_vec2_clamp`
 #. :c:func:`glm_vec2_lerp`
+#. :c:func:`glm_vec2_make`
 
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
@@ -313,6 +315,15 @@ Functions documentation
       | *[in]*  **axis**   axis vector
       | *[out]* **dest**   destination
 
+.. c:function:: void  glm_vec2_center(vec2 v1, vec2 v2, vec2 dest)
+
+    find center point of two vector
+
+    Parameters:
+      | *[in]*  **v1**    vector1
+      | *[in]*  **v2**    vector2
+      | *[out]* **dest**  center point
+
 .. c:function:: float glm_vec2_distance2(vec2 v1, vec2 v2)
 
     squared distance between two vectors
@@ -373,3 +384,12 @@ Functions documentation
       | *[in]*  **to**     to value
       | *[in]*  **t**      interpolant (amount) clamped between 0 and 1
       | *[out]* **dest**   destination
+
+.. c:function:: void glm_vec2_make(float * __restrict src, vec2 dest)
+
+    Create two dimensional vector from pointer
+
+    | NOTE: **@src** must contain at least 2 elements.
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination vector

docs/source/vec3-ext.rst

@@ -50,6 +50,14 @@ Functions documentation
       | *[in]*  **val**   value
       | *[out]* **dest**  destination
 
+.. c:function:: void  glm_vec3_fill(vec3 v, float val)
+
+    fill a vector with specified value
+
+    Parameters:
+      | *[out]*  **v**    vector
+      | *[in]*   **val**  value
+
 .. c:function:: bool  glm_vec3_eq(vec3 v, float val)
 
     check if vector is equal to value (without epsilon)

docs/source/vec3.rst

@@ -79,6 +79,7 @@ Functions:
 #. :c:func:`glm_vec3_ortho`
 #. :c:func:`glm_vec3_clamp`
 #. :c:func:`glm_vec3_lerp`
+#. :c:func:`glm_vec3_make`
 
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
@@ -501,3 +502,13 @@ Functions documentation
       | *[in]*  **to**     to value
       | *[in]*  **t**      interpolant (amount) clamped between 0 and 1
       | *[out]* **dest**   destination
+
+.. c:function:: void glm_vec3_make(float * __restrict src, vec3 dest)
+
+    Create three dimensional vector from pointer
+
+    | NOTE: **@src** must contain at least 3 elements.
+
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination vector

docs/source/vec4.rst

@@ -59,6 +59,7 @@ Functions:
 #. :c:func:`glm_vec4_clamp`
 #. :c:func:`glm_vec4_lerp`
 #. :c:func:`glm_vec4_cubic`
+#. :c:func:`glm_vec4_make`
 
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
@@ -406,3 +407,12 @@ Functions documentation
     Parameters:
       | *[in]*  **s**      parameter
       | *[out]* **dest**   destination
+
+.. c:function:: void glm_vec4_make(float * __restrict src, vec4 dest)
+
+    Create four dimensional vector from pointer
+
+    | NOTE: **@src** must contain at least 4 elements.
+    Parameters:
+      | *[in]*  **src**  pointer to an array of floats
+      | *[out]* **dest** destination vector

include/cglm/aabb2d.h

@@ -0,0 +1,245 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglm_aabb2d_h
+#define cglm_aabb2d_h
+
+#include "common.h"
+#include "vec2.h"
+#include "vec4.h"
+#include "util.h"
+
+/*!
+ * @brief copy all members of [aabb] to [dest]
+ *
+ * @param[in]  aabb source
+ * @param[out] dest destination
+ */
+CGLM_INLINE
+void
+glm_aabb2d_copy(vec2 aabb[2], vec2 dest[2]) {
+  glm_vec2_copy(aabb[0], dest[0]);
+  glm_vec2_copy(aabb[1], dest[1]);
+}
+
+/*!
+ * @brief apply transform to Axis-Aligned Bounding aabb
+ *
+ * @param[in]  aabb  bounding aabb
+ * @param[in]  m    transform matrix
+ * @param[out] dest transformed bounding aabb
+ */
+CGLM_INLINE
+void
+glm_aabb2d_transform(vec2 aabb[2], mat3 m, vec2 dest[2]) {
+  vec2 v[2], xa, xb, ya, yb;
+
+  glm_vec2_scale(m[0], aabb[0][0], xa);
+  glm_vec2_scale(m[0], aabb[1][0], xb);
+
+  glm_vec2_scale(m[1], aabb[0][1], ya);
+  glm_vec2_scale(m[1], aabb[1][1], yb);
+
+  /* translation + min(xa, xb) + min(ya, yb) */
+  glm_vec2(m[2], v[0]);
+  glm_vec2_minadd(xa, xb, v[0]);
+  glm_vec2_minadd(ya, yb, v[0]);
+
+  /* translation + max(xa, xb) + max(ya, yb) */
+  glm_vec2(m[2], v[1]);
+  glm_vec2_maxadd(xa, xb, v[1]);
+  glm_vec2_maxadd(ya, yb, v[1]);
+
+  glm_vec2_copy(v[0], dest[0]);
+  glm_vec2_copy(v[1], dest[1]);
+}
+
+/*!
+ * @brief merges two AABB bounding aabb and creates new one
+ *
+ * two aabb must be in same space, if one of aabb is in different space then
+ * you should consider to convert it's space by glm_aabb_space
+ *
+ * @param[in]  aabb1 bounding aabb 1
+ * @param[in]  aabb2 bounding aabb 2
+ * @param[out] dest merged bounding aabb
+ */
+CGLM_INLINE
+void
+glm_aabb2d_merge(vec2 aabb1[2], vec2 aabb2[2], vec2 dest[2]) {
+  dest[0][0] = glm_min(aabb1[0][0], aabb2[0][0]);
+  dest[0][1] = glm_min(aabb1[0][1], aabb2[0][1]);
+
+  dest[1][0] = glm_max(aabb1[1][0], aabb2[1][0]);
+  dest[1][1] = glm_max(aabb1[1][1], aabb2[1][1]);
+}
+
+/*!
+ * @brief crops a bounding aabb with another one.
+ *
+ * this could be useful for gettng a baabb which fits with view frustum and
+ * object bounding aabbes. In this case you crop view frustum aabb with objects
+ * aabb
+ *
+ * @param[in]  aabb     bounding aabb 1
+ * @param[in]  cropAabb crop aabb
+ * @param[out] dest     cropped bounding aabb
+ */
+CGLM_INLINE
+void
+glm_aabb2d_crop(vec2 aabb[2], vec2 cropAabb[2], vec2 dest[2]) {
+  dest[0][0] = glm_max(aabb[0][0], cropAabb[0][0]);
+  dest[0][1] = glm_max(aabb[0][1], cropAabb[0][1]);
+
+  dest[1][0] = glm_min(aabb[1][0], cropAabb[1][0]);
+  dest[1][1] = glm_min(aabb[1][1], cropAabb[1][1]);
+}
+
+/*!
+ * @brief crops a bounding aabb with another one.
+ *
+ * this could be useful for gettng a baabb which fits with view frustum and
+ * object bounding aabbes. In this case you crop view frustum aabb with objects
+ * aabb
+ *
+ * @param[in]  aabb      bounding aabb
+ * @param[in]  cropAabb  crop aabb
+ * @param[in]  clampAabb miniumum aabb
+ * @param[out] dest      cropped bounding aabb
+ */
+CGLM_INLINE
+void
+glm_aabb2d_crop_until(vec2 aabb[2],
+                      vec2 cropAabb[2],
+                      vec2 clampAabb[2],
+                      vec2 dest[2]) {
+  glm_aabb2d_crop(aabb, cropAabb, dest);
+  glm_aabb2d_merge(clampAabb, dest, dest);
+}
+
+/*!
+ * @brief invalidate AABB min and max values
+ *
+ * @param[in, out]  aabb bounding aabb
+ */
+CGLM_INLINE
+void
+glm_aabb2d_invalidate(vec2 aabb[2]) {
+  glm_vec2_fill(aabb[0], FLT_MAX);
+  glm_vec2_fill(aabb[1], -FLT_MAX);
+}
+
+/*!
+ * @brief check if AABB is valid or not
+ *
+ * @param[in]  aabb bounding aabb
+ */
+CGLM_INLINE
+bool
+glm_aabb2d_isvalid(vec2 aabb[2]) {
+  return glm_vec2_max(aabb[0]) != FLT_MAX
+         && glm_vec2_min(aabb[1]) != -FLT_MAX;
+}
+
+/*!
+ * @brief distance between of min and max
+ *
+ * @param[in]  aabb bounding aabb
+ */
+CGLM_INLINE
+float
+glm_aabb2d_size(vec2 aabb[2]) {
+  return glm_vec2_distance(aabb[0], aabb[1]);
+}
+
+/*!
+ * @brief radius of sphere which surrounds AABB
+ *
+ * @param[in]  aabb bounding aabb
+ */
+CGLM_INLINE
+float
+glm_aabb2d_radius(vec2 aabb[2]) {
+  return glm_aabb2d_size(aabb) * 0.5f;
+}
+
+/*!
+ * @brief computes center point of AABB
+ *
+ * @param[in]   aabb  bounding aabb
+ * @param[out]  dest center of bounding aabb
+ */
+CGLM_INLINE
+void
+glm_aabb2d_center(vec2 aabb[2], vec2 dest) {
+  glm_vec2_center(aabb[0], aabb[1], dest);
+}
+
+/*!
+ * @brief check if two AABB intersects
+ *
+ * @param[in]   aabb    bounding aabb
+ * @param[in]   other  other bounding aabb
+ */
+CGLM_INLINE
+bool
+glm_aabb2d_aabb(vec2 aabb[2], vec2 other[2]) {
+  return (aabb[0][0] <= other[1][0] && aabb[1][0] >= other[0][0])
+      && (aabb[0][1] <= other[1][1] && aabb[1][1] >= other[0][1]);
+}
+
+/*!
+ * @brief check if AABB intersects with a circle
+ *
+ * Circle Representation in cglm: [center.x, center.y, radii]
+ *
+ * @param[in]   aabb   solid bounding aabb
+ * @param[in]   c      solid circle
+ */
+CGLM_INLINE
+bool
+glm_aabb2d_circle(vec2 aabb[2], vec3 c) {
+  float dmin;
+  int   a, b;
+
+  a = (c[0] < aabb[0][0]) + (c[0] > aabb[1][0]);
+  b = (c[1] < aabb[0][1]) + (c[1] > aabb[1][1]);
+
+  dmin  = glm_pow2((c[0] - aabb[!(a - 1)][0]) * (a != 0))
+        + glm_pow2((c[1] - aabb[!(b - 1)][1]) * (b != 0));
+
+  return dmin <= glm_pow2(c[2]);
+}
+
+/*!
+ * @brief check if point is inside of AABB
+ *
+ * @param[in]   aabb    bounding aabb
+ * @param[in]   point  point
+ */
+CGLM_INLINE
+bool
+glm_aabb2d_point(vec2 aabb[2], vec2 point) {
+  return (point[0] >= aabb[0][0] && point[0] <= aabb[1][0])
+      && (point[1] >= aabb[0][1] && point[1] <= aabb[1][1]);
+}
+
+/*!
+ * @brief check if AABB contains other AABB
+ *
+ * @param[in]   aabb    bounding aabb
+ * @param[in]   other  other bounding aabb
+ */
+CGLM_INLINE
+bool
+glm_aabb2d_contains(vec2 aabb[2], vec2 other[2]) {
+  return (aabb[0][0] <= other[0][0] && aabb[1][0] >= other[1][0])
+      && (aabb[0][1] <= other[0][1] && aabb[1][1] >= other[1][1]);
+}
+
+#endif /* cglm_aabb2d_h */
+

include/cglm/affine-mat.h

@@ -8,6 +8,7 @@
 /*
  Functions:
    CGLM_INLINE void glm_mul(mat4 m1, mat4 m2, mat4 dest);
+   CGLM_INLINE void glm_mul_rot(mat4 m1, mat4 m2, mat4 dest);
    CGLM_INLINE void glm_inv_tr(mat4 mat);
  */
 
@@ -30,6 +31,10 @@
 #  include "simd/neon/affine.h"
 #endif
 
+#ifdef CGLM_SIMD_WASM
+#  include "simd/wasm/affine.h"
+#endif
+
 /*!
  * @brief this is similar to glm_mat4_mul but specialized to affine transform
  *
@@ -49,7 +54,9 @@
 CGLM_INLINE
 void
 glm_mul(mat4 m1, mat4 m2, mat4 dest) {
-#ifdef __AVX__
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_mul_wasm(m1, m2, dest);
+#elif defined(__AVX__)
   glm_mul_avx(m1, m2, dest);
 #elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_mul_sse2(m1, m2, dest);
@@ -107,7 +114,9 @@ glm_mul(mat4 m1, mat4 m2, mat4 dest) {
 CGLM_INLINE
 void
 glm_mul_rot(mat4 m1, mat4 m2, mat4 dest) {
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_mul_rot_wasm(m1, m2, dest);
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_mul_rot_sse2(m1, m2, dest);
 #elif defined(CGLM_NEON_FP)
   glm_mul_rot_neon(m1, m2, dest);
@@ -156,7 +165,9 @@ glm_mul_rot(mat4 m1, mat4 m2, mat4 dest) {
 CGLM_INLINE
 void
 glm_inv_tr(mat4 mat) {
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_inv_tr_wasm(mat);
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_inv_tr_sse2(mat);
 #elif defined(CGLM_NEON_FP)
   glm_inv_tr_neon(mat);

include/cglm/bezier.h

@@ -23,7 +23,7 @@
 #define GLM_HERMITE_MAT ((mat4)GLM_HERMITE_MAT_INIT)
 
 #define CGLM_DECASTEL_EPS   1e-9f
-#define CGLM_DECASTEL_MAX   1000.0f
+#define CGLM_DECASTEL_MAX   1000
 #define CGLM_DECASTEL_SMALL 1e-20f
 
 /*!

include/cglm/call.h

@@ -19,14 +19,21 @@ extern "C" {
 #include "call/ivec3.h"
 #include "call/ivec4.h"
 #include "call/mat2.h"
+#include "call/mat2x3.h"
+#include "call/mat2x4.h"
 #include "call/mat3.h"
+#include "call/mat3x2.h"
+#include "call/mat3x4.h"
 #include "call/mat4.h"
+#include "call/mat4x2.h"
+#include "call/mat4x3.h"
 #include "call/affine.h"
 #include "call/cam.h"
 #include "call/quat.h"
 #include "call/euler.h"
 #include "call/plane.h"
 #include "call/frustum.h"
+#include "call/aabb2d.h"
 #include "call/box.h"
 #include "call/io.h"
 #include "call/project.h"

include/cglm/call/aabb2d.h

@@ -0,0 +1,80 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglmc_aabb2d_h
+#define cglmc_aabb2d_h
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include "../cglm.h"
+
+CGLM_EXPORT
+void
+glmc_aabb2d_copy(vec2 aabb[2], vec2 dest[2]);
+
+CGLM_EXPORT
+void
+glmc_aabb2d_transform(vec2 aabb[2], mat3 m, vec2 dest[2]);
+
+CGLM_EXPORT
+void
+glmc_aabb2d_merge(vec2 aabb1[2], vec2 aabb2[2], vec2 dest[2]);
+
+CGLM_EXPORT
+void
+glmc_aabb2d_crop(vec2 aabb[2], vec2 cropAabb[2], vec2 dest[2]);
+
+CGLM_EXPORT
+void
+glmc_aabb2d_crop_until(vec2 aabb[2],
+                     vec2 cropAabb[2],
+                     vec2 clampAabb[2],
+                     vec2 dest[2]);
+
+CGLM_EXPORT
+void
+glmc_aabb2d_invalidate(vec2 aabb[2]);
+
+CGLM_EXPORT
+bool
+glmc_aabb2d_isvalid(vec2 aabb[2]);
+
+CGLM_EXPORT
+float
+glmc_aabb2d_size(vec2 aabb[2]);
+
+CGLM_EXPORT
+float
+glmc_aabb2d_radius(vec2 aabb[2]);
+
+CGLM_EXPORT
+void
+glmc_aabb2d_center(vec2 aabb[2], vec2 dest);
+
+CGLM_EXPORT
+bool
+glmc_aabb2d_aabb(vec2 aabb[2], vec2 other[2]);
+
+CGLM_EXPORT
+bool
+glmc_aabb2d_point(vec2 aabb[2], vec2 point);
+
+CGLM_EXPORT
+bool
+glmc_aabb2d_contains(vec2 aabb[2], vec2 other[2]);
+
+CGLM_EXPORT
+bool
+glmc_aabb2d_circle(vec2 aabb[2], vec3 s);
+
+#ifdef __cplusplus
+}
+#endif
+#endif /* cglmc_aabb2d_h */
+
+

include/cglm/call/euler.h

@@ -49,6 +49,31 @@ CGLM_EXPORT
 void
 glmc_euler_by_order(vec3 angles, glm_euler_seq axis, mat4 dest);
 
+CGLM_EXPORT
+void
+glmc_euler_xyz_quat(vec3 angles, versor dest);
+
+CGLM_EXPORT
+void
+glmc_euler_xzy_quat(vec3 angles, versor dest);
+
+CGLM_EXPORT
+void
+glmc_euler_yxz_quat(vec3 angles, versor dest);
+
+CGLM_EXPORT
+void
+glmc_euler_yzx_quat(vec3 angles, versor dest);
+
+CGLM_EXPORT
+void
+glmc_euler_zxy_quat(vec3 angles, versor dest);
+
+CGLM_EXPORT
+void
+glmc_euler_zyx_quat(vec3 angles, versor dest);
+
+
 #ifdef __cplusplus
 }
 #endif

include/cglm/call/ivec2.h

@@ -53,6 +53,70 @@ CGLM_EXPORT
 void
 glmc_ivec2_scale(ivec2 v, int s, ivec2 dest);
 
+CGLM_EXPORT 
+void 
+glmc_ivec2_addadd(ivec2 a, ivec2 b, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_addadds(ivec2 a, int s, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_subadd(ivec2 a, ivec2 b, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_subadds(ivec2 a, int s, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_muladd(ivec2 a, ivec2 b, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_muladds(ivec2 a, int s, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_maxadd(ivec2 a, ivec2 b, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_minadd(ivec2 a, ivec2 b, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_subsub(ivec2 a, ivec2 b, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_subsubs(ivec2 a, int s, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_addsub(ivec2 a, ivec2 b, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_addsubs(ivec2 a, int s, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_mulsub(ivec2 a, ivec2 b, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_mulsubs(ivec2 a, int s, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_maxsub(ivec2 a, ivec2 b, ivec2 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec2_minsub(ivec2 a, ivec2 b, ivec2 dest);
+
 CGLM_EXPORT
 int
 glmc_ivec2_distance2(ivec2 a, ivec2 b);
@@ -61,6 +125,18 @@ CGLM_EXPORT
 float
 glmc_ivec2_distance(ivec2 a, ivec2 b);
 
+CGLM_EXPORT
+void
+glmc_ivec2_fill(ivec2 v, int val);
+
+CGLM_EXPORT
+bool
+glmc_ivec2_eq(ivec2 v, int val);
+
+CGLM_EXPORT
+bool
+glmc_ivec2_eqv(ivec2 a, ivec2 b);
+
 CGLM_EXPORT
 void
 glmc_ivec2_maxv(ivec2 a, ivec2 b, ivec2 dest);

include/cglm/call/ivec3.h

@@ -53,6 +53,70 @@ CGLM_EXPORT
 void
 glmc_ivec3_scale(ivec3 v, int s, ivec3 dest);
 
+CGLM_EXPORT 
+void 
+glmc_ivec3_addadd(ivec3 a, ivec3 b, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_addadds(ivec3 a, int s, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_subadd(ivec3 a, ivec3 b, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_subadds(ivec3 a, int s, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_muladd(ivec3 a, ivec3 b, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_muladds(ivec3 a, int s, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_maxadd(ivec3 a, ivec3 b, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_minadd(ivec3 a, ivec3 b, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_subsub(ivec3 a, ivec3 b, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_subsubs(ivec3 a, int s, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_addsub(ivec3 a, ivec3 b, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_addsubs(ivec3 a, int s, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_mulsub(ivec3 a, ivec3 b, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_mulsubs(ivec3 a, int s, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_maxsub(ivec3 a, ivec3 b, ivec3 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec3_minsub(ivec3 a, ivec3 b, ivec3 dest);
+
 CGLM_EXPORT
 int
 glmc_ivec3_distance2(ivec3 a, ivec3 b);
@@ -61,6 +125,18 @@ CGLM_EXPORT
 float
 glmc_ivec3_distance(ivec3 a, ivec3 b);
 
+CGLM_EXPORT
+void
+glmc_ivec3_fill(ivec3 v, int val);
+
+CGLM_EXPORT
+bool
+glmc_ivec3_eq(ivec3 v, int val);
+
+CGLM_EXPORT
+bool
+glmc_ivec3_eqv(ivec3 a, ivec3 b);
+
 CGLM_EXPORT
 void
 glmc_ivec3_maxv(ivec3 a, ivec3 b, ivec3 dest);

include/cglm/call/ivec4.h

@@ -53,6 +53,70 @@ CGLM_EXPORT
 void
 glmc_ivec4_scale(ivec4 v, int s, ivec4 dest);
 
+CGLM_EXPORT 
+void 
+glmc_ivec4_addadd(ivec4 a, ivec4 b, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_addadds(ivec4 a, int s, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_subadd(ivec4 a, ivec4 b, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_subadds(ivec4 a, int s, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_muladd(ivec4 a, ivec4 b, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_muladds(ivec4 a, int s, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_maxadd(ivec4 a, ivec4 b, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_minadd(ivec4 a, ivec4 b, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_subsub(ivec4 a, ivec4 b, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_subsubs(ivec4 a, int s, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_addsub(ivec4 a, ivec4 b, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_addsubs(ivec4 a, int s, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_mulsub(ivec4 a, ivec4 b, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_mulsubs(ivec4 a, int s, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_maxsub(ivec4 a, ivec4 b, ivec4 dest);
+
+CGLM_EXPORT 
+void 
+glmc_ivec4_minsub(ivec4 a, ivec4 b, ivec4 dest);
+
 CGLM_EXPORT
 int
 glmc_ivec4_distance2(ivec4 a, ivec4 b);

include/cglm/call/mat2.h

@@ -73,6 +73,10 @@ CGLM_EXPORT
 float
 glmc_mat2_rmc(vec2 r, mat2 m, vec2 c);
 
+CGLM_EXPORT
+void
+glmc_mat2_make(float * __restrict src, mat2 dest);
+
 #ifdef __cplusplus
 }
 #endif

include/cglm/call/mat2x3.h

@@ -0,0 +1,47 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglmc_mat2x3_h
+#define cglmc_mat2x3_h
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include "../cglm.h"
+
+CGLM_EXPORT
+void
+glmc_mat2x3_copy(mat2x3 mat, mat2x3 dest);
+
+CGLM_EXPORT
+void
+glmc_mat2x3_zero(mat2x3 mat);
+
+CGLM_EXPORT
+void
+glmc_mat2x3_make(float * __restrict src, mat2x3 dest);
+
+CGLM_EXPORT
+void
+glmc_mat2x3_mul(mat2x3 m1, mat3x2 m2, mat2 dest);
+
+CGLM_EXPORT
+void
+glmc_mat2x3_mulv(mat2x3 m, vec3 v, vec2 dest);
+
+CGLM_EXPORT
+void
+glmc_mat2x3_transpose(mat2x3 m, mat3x2 dest);
+
+CGLM_EXPORT
+void
+glmc_mat2x3_scale(mat2x3 m, float s);
+
+#ifdef __cplusplus
+}
+#endif
+#endif /* cglmc_mat2x3_h */

include/cglm/call/mat2x4.h

@@ -0,0 +1,47 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglmc_mat2x4_h
+#define cglmc_mat2x4_h
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include "../cglm.h"
+
+CGLM_EXPORT
+void
+glmc_mat2x4_copy(mat2x4 mat, mat2x4 dest);
+
+CGLM_EXPORT
+void
+glmc_mat2x4_zero(mat2x4 mat);
+
+CGLM_EXPORT
+void
+glmc_mat2x4_make(float * __restrict src, mat2x4 dest);
+
+CGLM_EXPORT
+void
+glmc_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat2 dest);
+
+CGLM_EXPORT
+void
+glmc_mat2x4_mulv(mat2x4 m, vec4 v, vec2 dest);
+
+CGLM_EXPORT
+void
+glmc_mat2x4_transpose(mat2x4 m, mat4x2 dest);
+
+CGLM_EXPORT
+void
+glmc_mat2x4_scale(mat2x4 m, float s);
+
+#ifdef __cplusplus
+}
+#endif
+#endif /* cglmc_mat2x4_h */

include/cglm/call/mat3.h

@@ -80,6 +80,10 @@ CGLM_EXPORT
 float
 glmc_mat3_rmc(vec3 r, mat3 m, vec3 c);
 
+CGLM_EXPORT
+void
+glmc_mat3_make(float * __restrict src, mat3 dest);
+
 #ifdef __cplusplus
 }
 #endif

include/cglm/call/mat3x2.h

@@ -0,0 +1,47 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglmc_mat3x2_h
+#define cglmc_mat3x2_h
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include "../cglm.h"
+
+CGLM_EXPORT
+void
+glmc_mat3x2_copy(mat3x2 mat, mat3x2 dest);
+
+CGLM_EXPORT
+void
+glmc_mat3x2_zero(mat3x2 mat);
+
+CGLM_EXPORT
+void
+glmc_mat3x2_make(float * __restrict src, mat3x2 dest);
+
+CGLM_EXPORT
+void
+glmc_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat3 dest);
+
+CGLM_EXPORT
+void
+glmc_mat3x2_mulv(mat3x2 m, vec2 v, vec3 dest);
+
+CGLM_EXPORT
+void
+glmc_mat3x2_transpose(mat3x2 m, mat2x3 dest);
+
+CGLM_EXPORT
+void
+glmc_mat3x2_scale(mat3x2 m, float s);
+
+#ifdef __cplusplus
+}
+#endif
+#endif /* cglmc_mat3x2_h */

include/cglm/call/mat3x4.h

@@ -0,0 +1,47 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglmc_mat3x4_h
+#define cglmc_mat3x4_h
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include "../cglm.h"
+
+CGLM_EXPORT
+void
+glmc_mat3x4_copy(mat3x4 mat, mat3x4 dest);
+
+CGLM_EXPORT
+void
+glmc_mat3x4_zero(mat3x4 mat);
+
+CGLM_EXPORT
+void
+glmc_mat3x4_make(float * __restrict src, mat3x4 dest);
+
+CGLM_EXPORT
+void
+glmc_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat3 dest);
+
+CGLM_EXPORT
+void
+glmc_mat3x4_mulv(mat3x4 m, vec4 v, vec3 dest);
+
+CGLM_EXPORT
+void
+glmc_mat3x4_transpose(mat3x4 m, mat4x3 dest);
+
+CGLM_EXPORT
+void
+glmc_mat3x4_scale(mat3x4 m, float s);
+
+#ifdef __cplusplus
+}
+#endif
+#endif /* cglmc_mat3x4_h */

include/cglm/call/mat4.h

@@ -121,6 +121,10 @@ CGLM_EXPORT
 float
 glmc_mat4_rmc(vec4 r, mat4 m, vec4 c);
 
+CGLM_EXPORT
+void
+glmc_mat4_make(float * __restrict src, mat4 dest);
+
 #ifdef __cplusplus
 }
 #endif

include/cglm/call/mat4x2.h

@@ -0,0 +1,47 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglmc_mat4x2_h
+#define cglmc_mat4x2_h
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include "../cglm.h"
+
+CGLM_EXPORT
+void
+glmc_mat4x2_copy(mat4x2 mat, mat4x2 dest);
+
+CGLM_EXPORT
+void
+glmc_mat4x2_zero(mat4x2 mat);
+
+CGLM_EXPORT
+void
+glmc_mat4x2_make(float * __restrict src, mat4x2 dest);
+
+CGLM_EXPORT
+void
+glmc_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat4 dest);
+
+CGLM_EXPORT
+void
+glmc_mat4x2_mulv(mat4x2 m, vec2 v, vec4 dest);
+
+CGLM_EXPORT
+void
+glmc_mat4x2_transpose(mat4x2 m, mat2x4 dest);
+
+CGLM_EXPORT
+void
+glmc_mat4x2_scale(mat4x2 m, float s);
+
+#ifdef __cplusplus
+}
+#endif
+#endif /* cglmc_mat4x2_h */

include/cglm/call/mat4x3.h

@@ -0,0 +1,47 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglmc_mat4x3_h
+#define cglmc_mat4x3_h
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#include "../cglm.h"
+
+CGLM_EXPORT
+void
+glmc_mat4x3_copy(mat4x3 mat, mat4x3 dest);
+
+CGLM_EXPORT
+void
+glmc_mat4x3_zero(mat4x3 mat);
+
+CGLM_EXPORT
+void
+glmc_mat4x3_make(float * __restrict src, mat4x3 dest);
+
+CGLM_EXPORT
+void
+glmc_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat4 dest);
+
+CGLM_EXPORT
+void
+glmc_mat4x3_mulv(mat4x3 m, vec3 v, vec4 dest);
+
+CGLM_EXPORT
+void
+glmc_mat4x3_transpose(mat4x3 m, mat3x4 dest);
+
+CGLM_EXPORT
+void
+glmc_mat4x3_scale(mat4x3 m, float s);
+
+#ifdef __cplusplus
+}
+#endif
+#endif /* cglmc_mat4x3_h */

include/cglm/call/quat.h

@@ -161,6 +161,10 @@ CGLM_EXPORT
 void
 glmc_quat_rotate_atm(mat4 m, versor q, vec3 pivot);
 
+CGLM_EXPORT
+void
+glmc_quat_make(float * __restrict src, versor dest);
+
 #ifdef __cplusplus
 }
 #endif

include/cglm/call/vec2.h

@@ -105,6 +105,30 @@ CGLM_EXPORT
 void
 glmc_vec2_minadd(vec2 a, vec2 b, vec2 dest);
 
+CGLM_EXPORT
+void
+glmc_vec2_subsub(vec2 a, vec2 b, vec2 dest);
+
+CGLM_EXPORT
+void
+glmc_vec2_addsub(vec2 a, vec2 b, vec2 dest);
+
+CGLM_EXPORT
+void
+glmc_vec2_mulsub(vec2 a, vec2 b, vec2 dest);
+
+CGLM_EXPORT
+void
+glmc_vec2_mulsubs(vec2 a, float s, vec2 dest);
+
+CGLM_EXPORT
+void
+glmc_vec2_maxsub(vec2 a, vec2 b, vec2 dest);
+
+CGLM_EXPORT
+void
+glmc_vec2_minsub(vec2 a, vec2 b, vec2 dest);
+
 CGLM_EXPORT
 void
 glmc_vec2_negate_to(vec2 v, vec2 dest);
@@ -125,6 +149,10 @@ CGLM_EXPORT
 void
 glmc_vec2_rotate(vec2 v, float angle, vec2 dest);
 
+CGLM_EXPORT
+void
+glmc_vec2_center(vec2 a, vec2 b, vec2 dest);
+
 CGLM_EXPORT
 float
 glmc_vec2_distance2(vec2 a, vec2 b);
@@ -165,6 +193,10 @@ CGLM_EXPORT
 void
 glmc_vec2_complex_conjugate(vec2 a, vec2 dest);
 
+CGLM_EXPORT
+void
+glmc_vec2_make(float * __restrict src, vec2 dest);
+
 #ifdef __cplusplus
 }
 #endif

include/cglm/call/vec3.h

@@ -132,6 +132,30 @@ CGLM_EXPORT
 void
 glmc_vec3_minadd(vec3 a, vec3 b, vec3 dest);
 
+CGLM_EXPORT
+void
+glmc_vec3_subsub(vec3 a, vec3 b, vec3 dest);
+
+CGLM_EXPORT
+void
+glmc_vec3_addsub(vec3 a, vec3 b, vec3 dest);
+
+CGLM_EXPORT
+void
+glmc_vec3_mulsub(vec3 a, vec3 b, vec3 dest);
+
+CGLM_EXPORT
+void
+glmc_vec3_mulsubs(vec3 a, float s, vec3 dest);
+
+CGLM_EXPORT
+void
+glmc_vec3_maxsub(vec3 a, vec3 b, vec3 dest);
+
+CGLM_EXPORT
+void
+glmc_vec3_minsub(vec3 a, vec3 b, vec3 dest);
+
 CGLM_EXPORT
 void
 glmc_vec3_negate(vec3 v);
@@ -306,6 +330,10 @@ CGLM_EXPORT
 void
 glmc_vec3_sqrt(vec3 v, vec3 dest);
 
+CGLM_EXPORT
+void
+glmc_vec3_make(float * __restrict src, vec3 dest);
+
 #ifdef __cplusplus
 }
 #endif

include/cglm/call/vec4.h

@@ -133,6 +133,30 @@ CGLM_EXPORT
 void
 glmc_vec4_minadd(vec4 a, vec4 b, vec4 dest);
 
+CGLM_EXPORT
+void
+glmc_vec4_subsub(vec4 a, vec4 b, vec4 dest);
+
+CGLM_EXPORT
+void
+glmc_vec4_addsub(vec4 a, vec4 b, vec4 dest);
+
+CGLM_EXPORT
+void
+glmc_vec4_mulsub(vec4 a, vec4 b, vec4 dest);
+
+CGLM_EXPORT
+void
+glmc_vec4_mulsubs(vec4 a, float s, vec4 dest);
+
+CGLM_EXPORT
+void
+glmc_vec4_maxsub(vec4 a, vec4 b, vec4 dest);
+
+CGLM_EXPORT
+void
+glmc_vec4_minsub(vec4 a, vec4 b, vec4 dest);
+
 CGLM_EXPORT
 void
 glmc_vec4_negate(vec4 v);
@@ -283,6 +307,10 @@ CGLM_EXPORT
 void
 glmc_vec4_sqrt(vec4 v, vec4 dest);
 
+CGLM_EXPORT
+void
+glmc_vec4_make(float * __restrict src, vec4 dest);
+
 #ifdef __cplusplus
 }
 #endif

include/cglm/cglm.h

@@ -16,14 +16,21 @@
 #include "ivec3.h"
 #include "ivec4.h"
 #include "mat4.h"
+#include "mat4x2.h"
+#include "mat4x3.h"
 #include "mat3.h"
+#include "mat3x2.h"
+#include "mat3x4.h"
 #include "mat2.h"
+#include "mat2x3.h"
+#include "mat2x4.h"
 #include "affine.h"
 #include "cam.h"
 #include "frustum.h"
 #include "quat.h"
 #include "euler.h"
 #include "plane.h"
+#include "aabb2d.h"
 #include "box.h"
 #include "color.h"
 #include "util.h"

include/cglm/common.h

@@ -82,4 +82,34 @@
 #  endif
 #endif
 
+/* struct API configurator */
+/* TODO: move struct/common.h? */
+/* WARN: dont use concant helpers outside cglm headers, because they may be changed */
+
+#define CGLM_MACRO_CONCAT_HELPER(A, B, C, D, E, ...) A ## B ## C ## D ## E ## __VA_ARGS__
+#define CGLM_MACRO_CONCAT(A, B, C, D, E, ...) CGLM_MACRO_CONCAT_HELPER(A, B, C, D, E,__VA_ARGS__)
+
+#ifndef CGLM_OMIT_NS_FROM_STRUCT_API
+#  ifndef CGLM_STRUCT_API_NS
+#    define CGLM_STRUCT_API_NS glms
+#  endif
+#  ifndef CGLM_STRUCT_API_NS_SEPERATOR
+#    define CGLM_STRUCT_API_NS_SEPERATOR _
+#  endif
+#else
+#  define CGLM_STRUCT_API_NS
+#  define CGLM_STRUCT_API_NS_SEPERATOR
+#endif
+
+#ifndef CGLM_STRUCT_API_NAME_SUFFIX
+#  define CGLM_STRUCT_API_NAME_SUFFIX
+#endif
+
+#define CGLM_STRUCTAPI(A, ...) CGLM_MACRO_CONCAT(CGLM_STRUCT_API_NS,             \
+                                                 CGLM_STRUCT_API_NS_SEPERATOR,   \
+                                                 A,                              \
+                                                 CGLM_STRUCT_API_NAME_SUFFIX,    \
+                                                 _,                              \
+                                                 __VA_ARGS__)
+
 #endif /* cglm_common_h */

include/cglm/euler.h

@@ -30,6 +30,12 @@
    CGLM_INLINE void glm_euler_by_order(vec3         angles,
                                        glm_euler_seq ord,
                                        mat4         dest);
+   CGLM_INLINE void glm_euler_xyz_quat(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_xzy_quat(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_yxz_quat(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_yzx_quat(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_zxy_quat(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_zyx_quat(vec3 angles, versor dest);
  */
 
 #ifndef cglm_euler_h
@@ -37,6 +43,47 @@
 
 #include "common.h"
 
+#ifdef CGLM_FORCE_LEFT_HANDED
+#  include "handed/euler_to_quat_lh.h"
+#else
+#  include "handed/euler_to_quat_rh.h"
+#endif
+
+
+#ifndef CGLM_CLIPSPACE_INCLUDE_ALL
+#  if CGLM_CONFIG_CLIP_CONTROL == CGLM_CLIP_CONTROL_LH_ZO
+#    include "clipspace/ortho_lh_zo.h"
+#    include "clipspace/persp_lh_zo.h"
+#    include "clipspace/view_lh_zo.h"
+#  elif CGLM_CONFIG_CLIP_CONTROL == CGLM_CLIP_CONTROL_LH_NO
+#    include "clipspace/ortho_lh_no.h"
+#    include "clipspace/persp_lh_no.h"
+#    include "clipspace/view_lh_no.h"
+#  elif CGLM_CONFIG_CLIP_CONTROL == CGLM_CLIP_CONTROL_RH_ZO
+#    include "clipspace/ortho_rh_zo.h"
+#    include "clipspace/persp_rh_zo.h"
+#    include "clipspace/view_rh_zo.h"
+#  elif CGLM_CONFIG_CLIP_CONTROL == CGLM_CLIP_CONTROL_RH_NO
+#    include "clipspace/ortho_rh_no.h"
+#    include "clipspace/persp_rh_no.h"
+#    include "clipspace/view_rh_no.h"
+#  endif
+#else
+#  include "clipspace/ortho_lh_zo.h"
+#  include "clipspace/persp_lh_zo.h"
+#  include "clipspace/ortho_lh_no.h"
+#  include "clipspace/persp_lh_no.h"
+#  include "clipspace/ortho_rh_zo.h"
+#  include "clipspace/persp_rh_zo.h"
+#  include "clipspace/ortho_rh_no.h"
+#  include "clipspace/persp_rh_no.h"
+#  include "clipspace/view_lh_zo.h"
+#  include "clipspace/view_lh_no.h"
+#  include "clipspace/view_rh_zo.h"
+#  include "clipspace/view_rh_no.h"
+#endif
+
+
 /*!
  * if you have axis order like vec3 orderVec = [0, 1, 2] or [0, 2, 1]...
  * vector then you can convert it to this enum by doing this:
@@ -188,7 +235,6 @@ glm_euler_xzy(vec3 angles, mat4 dest) {
   dest[3][3] =  1.0f;
 }
 
-
 /*!
  * @brief build rotation matrix from euler angles
  *
@@ -448,4 +494,108 @@ glm_euler_by_order(vec3 angles, glm_euler_seq ord, mat4 dest) {
   dest[3][3] = 1.0f;
 }
 
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in x y z order (roll pitch yaw)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_xyz_quat(vec3 angles, versor dest) {
+#ifdef CGLM_FORCE_LEFT_HANDED
+  glm_euler_xyz_quat_lh(angles, dest);
+#else
+  glm_euler_xyz_quat_rh(angles, dest);
+#endif
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in x z y order (roll yaw pitch)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_xzy_quat(vec3 angles, versor dest) {
+#ifdef CGLM_FORCE_LEFT_HANDED
+  glm_euler_xzy_quat_lh(angles, dest);
+#else
+  glm_euler_xzy_quat_rh(angles, dest);
+#endif
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in y x z order (pitch roll yaw)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_yxz_quat(vec3 angles, versor dest) {
+#ifdef CGLM_FORCE_LEFT_HANDED
+  glm_euler_yxz_quat_lh(angles, dest);
+#else
+  glm_euler_yxz_quat_rh(angles, dest);
+#endif
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in y z x order (pitch yaw roll)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_yzx_quat(vec3 angles, versor dest) {
+#ifdef CGLM_FORCE_LEFT_HANDED
+  glm_euler_yzx_quat_lh(angles, dest);
+#else
+  glm_euler_yzx_quat_rh(angles, dest);
+#endif
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in z x y order (yaw roll pitch)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_zxy_quat(vec3 angles, versor dest) {
+#ifdef CGLM_FORCE_LEFT_HANDED
+  glm_euler_zxy_quat_lh(angles, dest);
+#else
+  glm_euler_zxy_quat_rh(angles, dest);
+#endif
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in z y x order (yaw pitch roll)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_zyx_quat(vec3 angles, versor dest) {
+#ifdef CGLM_FORCE_LEFT_HANDED
+  glm_euler_zyx_quat_lh(angles, dest);
+#else
+  glm_euler_zyx_quat_rh(angles, dest);
+#endif
+}
+
+
 #endif /* cglm_euler_h */

include/cglm/handed/euler_to_quat_lh.h

@@ -0,0 +1,167 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+/*
+ Functions:
+   CGLM_INLINE void glm_euler_xyz_quat_lh(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_xzy_quat_lh(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_yxz_quat_lh(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_yzx_quat_lh(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_zxy_quat_lh(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_zyx_quat_lh(vec3 angles, versor dest);
+ */
+
+/*
+ Things to note:
+ The only difference between euler to quat rh vs lh is that the zsin part is negative
+ */
+
+#ifndef cglm_euler_to_quat_lh_h
+#define cglm_euler_to_quat_lh_h
+
+#include "../common.h"
+
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in x y z order in left hand (roll pitch yaw)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_xyz_quat_lh(vec3 angles, versor dest) {
+  float xc, yc, zc,
+        xs, ys, zs;
+
+  xs =  sinf(angles[0] * 0.5f); xc = cosf(angles[0] * 0.5f);
+  ys =  sinf(angles[1] * 0.5f); yc = cosf(angles[1] * 0.5f);
+  zs = -sinf(angles[2] * 0.5f); zc = cosf(angles[2] * 0.5f);
+
+  dest[0] = xc * ys * zs + xs * yc * zc;
+  dest[1] = xc * ys * zc - xs * yc * zs;
+  dest[2] = xc * yc * zs + xs * ys * zc;
+  dest[3] = xc * yc * zc - xs * ys * zs;
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in x z y order in left hand (roll yaw pitch)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_xzy_quat_lh(vec3 angles, versor dest) {
+  float xc, yc, zc,
+        xs, ys, zs;
+
+  xs =  sinf(angles[0] * 0.5f); xc = cosf(angles[0] * 0.5f);
+  ys =  sinf(angles[1] * 0.5f); yc = cosf(angles[1] * 0.5f);
+  zs = -sinf(angles[2] * 0.5f); zc = cosf(angles[2] * 0.5f);
+
+  dest[0] = -xc * zs * ys + xs * zc * yc;
+  dest[1] =  xc * zc * ys - xs * zs * yc;
+  dest[2] =  xc * zs * yc + xs * zc * ys;
+  dest[3] =  xc * zc * yc + xs * zs * ys;  
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in y x z order in left hand (pitch roll yaw)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_yxz_quat_lh(vec3 angles, versor dest) {
+  float xc, yc, zc,
+        xs, ys, zs;
+
+  xs =  sinf(angles[0] * 0.5f); xc = cosf(angles[0] * 0.5f);
+  ys =  sinf(angles[1] * 0.5f); yc = cosf(angles[1] * 0.5f);
+  zs = -sinf(angles[2] * 0.5f); zc = cosf(angles[2] * 0.5f);
+
+  dest[0] =  yc * xs * zc + ys * xc * zs;
+  dest[1] = -yc * xs * zs + ys * xc * zc;
+  dest[2] =  yc * xc * zs - ys * xs * zc;
+  dest[3] =  yc * xc * zc + ys * xs * zs;
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in y z x order in left hand (pitch yaw roll)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_yzx_quat_lh(vec3 angles, versor dest) {
+  float xc, yc, zc,
+        xs, ys, zs;
+
+  xs =  sinf(angles[0] * 0.5f); xc = cosf(angles[0] * 0.5f);
+  ys =  sinf(angles[1] * 0.5f); yc = cosf(angles[1] * 0.5f);
+  zs = -sinf(angles[2] * 0.5f); zc = cosf(angles[2] * 0.5f);
+
+  dest[0] = yc * zc * xs + ys * zs * xc;
+  dest[1] = yc * zs * xs + ys * zc * xc;
+  dest[2] = yc * zs * xc - ys * zc * xs;
+  dest[3] = yc * zc * xc - ys * zs * xs;
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in z x y order in left hand (yaw roll pitch)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_zxy_quat_lh(vec3 angles, versor dest) {
+  float xc, yc, zc,
+        xs, ys, zs;
+
+  xs =  sinf(angles[0] * 0.5f); xc = cosf(angles[0] * 0.5f);
+  ys =  sinf(angles[1] * 0.5f); yc = cosf(angles[1] * 0.5f);
+  zs = -sinf(angles[2] * 0.5f); zc = cosf(angles[2] * 0.5f);
+
+  dest[0] = zc * xs * yc - zs * xc * ys;
+  dest[1] = zc * xc * ys + zs * xs * yc;
+  dest[2] = zc * xs * ys + zs * xc * yc;
+  dest[3] = zc * xc * yc - zs * xs * ys;
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in z y x order in left hand (yaw pitch roll)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_zyx_quat_lh(vec3 angles, versor dest) {
+  float xc, yc, zc,
+        xs, ys, zs;
+
+  xs =  sinf(angles[0] * 0.5f); xc = cosf(angles[0] * 0.5f);
+  ys =  sinf(angles[1] * 0.5f); yc = cosf(angles[1] * 0.5f);
+  zs = -sinf(angles[2] * 0.5f); zc = cosf(angles[2] * 0.5f);
+
+  dest[0] =  zc * yc * xs - zs * ys * xc;
+  dest[1] =  zc * ys * xc + zs * yc * xs;
+  dest[2] = -zc * ys * xs + zs * yc * xc;
+  dest[3] =  zc * yc * xc + zs * ys * xs;
+}
+
+#endif /*cglm_euler_to_quat_lh_h*/

include/cglm/handed/euler_to_quat_rh.h

@@ -0,0 +1,170 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+/*
+ Functions:
+   CGLM_INLINE void glm_euler_xyz_quat_rh(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_xzy_quat_rh(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_yxz_quat_rh(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_yzx_quat_rh(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_zxy_quat_rh(vec3 angles, versor dest);
+   CGLM_INLINE void glm_euler_zyx_quat_rh(vec3 angles, versor dest);
+ */
+
+/*
+ Things to note:
+ The only difference between euler to quat rh vs lh is that the zsin part is negative
+ */
+
+#ifndef cglm_euler_to_quat_rh_h
+#define cglm_euler_to_quat_rh_h
+
+#include "../common.h"
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in x y z order in right hand (roll pitch yaw)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_xyz_quat_rh(vec3 angles, versor dest) {
+  float xc, yc, zc,
+        xs, ys, zs;
+
+  xs = sinf(angles[0] * 0.5f); xc = cosf(angles[0] * 0.5f);
+  ys = sinf(angles[1] * 0.5f); yc = cosf(angles[1] * 0.5f);
+  zs = sinf(angles[2] * 0.5f); zc = cosf(angles[2] * 0.5f);
+
+  dest[0] = xc * ys * zs + xs * yc * zc;
+  dest[1] = xc * ys * zc - xs * yc * zs;
+  dest[2] = xc * yc * zs + xs * ys * zc;
+  dest[3] = xc * yc * zc - xs * ys * zs;
+
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in x z y order in right hand (roll yaw pitch)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_xzy_quat_rh(vec3 angles, versor dest) {
+  float xc, yc, zc,
+        xs, ys, zs;
+
+  xs = sinf(angles[0] * 0.5f); xc = cosf(angles[0] * 0.5f);
+  ys = sinf(angles[1] * 0.5f); yc = cosf(angles[1] * 0.5f);
+  zs = sinf(angles[2] * 0.5f); zc = cosf(angles[2] * 0.5f);
+
+  dest[0] = -xc * zs * ys + xs * zc * yc;
+  dest[1] =  xc * zc * ys - xs * zs * yc;
+  dest[2] =  xc * zs * yc + xs * zc * ys;
+  dest[3] =  xc * zc * yc + xs * zs * ys;  
+
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in y x z order in right hand (pitch roll yaw)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_yxz_quat_rh(vec3 angles, versor dest) {
+  float xc, yc, zc,
+        xs, ys, zs;
+
+  xs = sinf(angles[0] * 0.5f); xc = cosf(angles[0] * 0.5f);
+  ys = sinf(angles[1] * 0.5f); yc = cosf(angles[1] * 0.5f);
+  zs = sinf(angles[2] * 0.5f); zc = cosf(angles[2] * 0.5f);
+
+  dest[0] =  yc * xs * zc + ys * xc * zs;
+  dest[1] = -yc * xs * zs + ys * xc * zc;
+  dest[2] =  yc * xc * zs - ys * xs * zc;
+  dest[3] =  yc * xc * zc + ys * xs * zs;
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in y z x order in right hand (pitch yaw roll)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_yzx_quat_rh(vec3 angles, versor dest) {
+  float xc, yc, zc,
+        xs, ys, zs;
+
+  xs = sinf(angles[0] * 0.5f); xc = cosf(angles[0] * 0.5f);
+  ys = sinf(angles[1] * 0.5f); yc = cosf(angles[1] * 0.5f);
+  zs = sinf(angles[2] * 0.5f); zc = cosf(angles[2] * 0.5f);
+
+  dest[0] = yc * zc * xs + ys * zs * xc;
+  dest[1] = yc * zs * xs + ys * zc * xc;
+  dest[2] = yc * zs * xc - ys * zc * xs;
+  dest[3] = yc * zc * xc - ys * zs * xs;
+
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in z x y order in right hand (yaw roll pitch)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_zxy_quat_rh(vec3 angles, versor dest) {
+  float xc, yc, zc,
+        xs, ys, zs;
+
+  xs = sinf(angles[0] * 0.5f); xc = cosf(angles[0] * 0.5f);
+  ys = sinf(angles[1] * 0.5f); yc = cosf(angles[1] * 0.5f);
+  zs = sinf(angles[2] * 0.5f); zc = cosf(angles[2] * 0.5f);
+
+  dest[0] = zc * xs * yc - zs * xc * ys;
+  dest[1] = zc * xc * ys + zs * xs * yc;
+  dest[2] = zc * xs * ys + zs * xc * yc;
+  dest[3] = zc * xc * yc - zs * xs * ys;
+}
+
+/*!
+ * @brief creates NEW quaternion using rotation angles and does
+ *        rotations in z y x order in right hand (yaw pitch roll)
+ * 
+ * @param[in]   angles angles x y z (radians)
+ * @param[out]  dest   quaternion
+ */
+CGLM_INLINE
+void
+glm_euler_zyx_quat_rh(vec3 angles, versor dest) {
+  float xc, yc, zc,
+        xs, ys, zs;
+
+  xs = sinf(angles[0] * 0.5f); xc = cosf(angles[0] * 0.5f);
+  ys = sinf(angles[1] * 0.5f); yc = cosf(angles[1] * 0.5f);
+  zs = sinf(angles[2] * 0.5f); zc = cosf(angles[2] * 0.5f);
+
+  dest[0] =  zc * yc * xs - zs * ys * xc;
+  dest[1] =  zc * ys * xc + zs * yc * xs;
+  dest[2] = -zc * ys * xs + zs * yc * xc;
+  dest[3] =  zc * yc * xc + zs * ys * xs;
+}
+
+
+#endif /*cglm_euler_to_quat_rh_h*/

include/cglm/io.h

@@ -10,9 +10,13 @@
    CGLM_INLINE void glm_mat4_print(mat4 matrix, FILE *ostream);
    CGLM_INLINE void glm_mat3_print(mat3 matrix, FILE *ostream);
    CGLM_INLINE void glm_vec4_print(vec4 vec, FILE *ostream);
+   CGLM_INLINE void glm_ivec4_print(ivec4 vec, FILE *ostream);
    CGLM_INLINE void glm_vec3_print(vec3 vec, FILE *ostream);
    CGLM_INLINE void glm_ivec3_print(ivec3 vec, FILE *ostream);
+   CGLM_INLINE void glm_vec2_print(vec2 vec, FILE *ostream);
+   CGLM_INLINE void glm_ivec2_print(ivec2 vec, FILE *ostream);
    CGLM_INLINE void glm_versor_print(versor vec, FILE *ostream);
+   CGLM_INLINE void glm_arch_print(FILE *ostream);
  */
 
 /*
@@ -52,13 +56,64 @@
 #  define CGLM_PRINT_MAX_TO_SHORT 1e5f
 #endif
 
-#ifndef CGLM_PRINT_COLOR
+#ifndef GLM_TESTS_NO_COLORFUL_OUTPUT
-#  define CGLM_PRINT_COLOR        "\033[36m"
+#  ifndef CGLM_PRINT_COLOR
+#    define CGLM_PRINT_COLOR        "\033[36m"
+#  endif
+#  ifndef CGLM_PRINT_COLOR_RESET
+#    define CGLM_PRINT_COLOR_RESET  "\033[0m"
+#  endif
+#else
+#  ifndef CGLM_PRINT_COLOR
+#    define CGLM_PRINT_COLOR
+#  endif
+#  ifndef CGLM_PRINT_COLOR_RESET
+#    define CGLM_PRINT_COLOR_RESET
+#  endif
 #endif
 
-#ifndef CGLM_PRINT_COLOR_RESET
+/*!
-#  define CGLM_PRINT_COLOR_RESET  "\033[0m"
+ * @brief prints current SIMD path in general
+ *
+ * @param[in] ostream    stream to print e.g. stdout, stderr, FILE ...
+ */
+CGLM_INLINE
+void
+glm_arch_print(FILE* __restrict ostream) {
+  fprintf(ostream, CGLM_PRINT_COLOR "arch: "
+#if defined(CGLM_SIMD_WASM)
+  "wasm SIMD128"
+#elif defined(CGLM_SIMD_x86)
+  "x86 SSE* "
+#  ifdef __AVX__
+  " AVX"
+#  endif
+#elif defined(CGLM_SIMD_ARM)
+  "arm"
+#  ifndef __ARM_NEON_FP
+    " NEON_FP"
+#  endif
+#  ifdef CGLM_ARM64
+    " ARM64"
+#  endif
+#else
+  "uncommon"
 #endif
+  CGLM_PRINT_COLOR_RESET);
+}
+
+/*!
+ * @brief prints current SIMD path in general
+ *
+ * @param[in] ostream    stream to print e.g. stdout, stderr, FILE ...
+ */
+CGLM_INLINE
+void
+glm_arch_print_name(FILE* __restrict ostream) {
+  fprintf(ostream, CGLM_PRINT_COLOR "\ncglm ");
+  glm_arch_print(ostream);
+  fprintf(ostream, "\n\n" CGLM_PRINT_COLOR_RESET);
+}
 
 CGLM_INLINE
 void
@@ -209,6 +264,24 @@ glm_vec4_print(vec4              vec,
 #undef m
 }
 
+CGLM_INLINE
+void
+glm_ivec4_print(ivec4             vec,
+                FILE * __restrict ostream) {
+  int i;
+
+#define m 4
+
+  fprintf(ostream, "Vector (int%d): " CGLM_PRINT_COLOR "\n  (", m);
+
+  for (i = 0; i < m; i++)
+    fprintf(ostream, " % d", vec[i]);
+
+  fprintf(ostream, "  )" CGLM_PRINT_COLOR_RESET "\n\n");
+
+#undef m
+}
+
 CGLM_INLINE
 void
 glm_vec3_print(vec3              vec,
@@ -271,6 +344,24 @@ glm_vec2_print(vec2              vec,
 #undef m
 }
 
+CGLM_INLINE
+void
+glm_ivec2_print(ivec2             vec,
+                FILE * __restrict ostream) {
+  int i;
+
+#define m 2
+
+  fprintf(ostream, "Vector (int%d): " CGLM_PRINT_COLOR "\n  (", m);
+
+  for (i = 0; i < m; i++)
+    fprintf(ostream, " % d", vec[i]);
+
+  fprintf(ostream, "  )" CGLM_PRINT_COLOR_RESET "\n\n");
+
+#undef m
+}
+
 CGLM_INLINE
 void
 glm_versor_print(versor            vec,
@@ -335,11 +426,15 @@ glm_aabb_print(vec3                    bbox[2],
 #define glm_mat3_print(v, s) (void)v; (void)s;
 #define glm_mat2_print(v, s) (void)v; (void)s;
 #define glm_vec4_print(v, s) (void)v; (void)s;
+#define glm_ivec4_print(v, s) (void)v; (void)s;
 #define glm_vec3_print(v, s) (void)v; (void)s;
 #define glm_ivec3_print(v, s) (void)v; (void)s;
 #define glm_vec2_print(v, s) (void)v; (void)s;
+#define glm_ivec2_print(v, s) (void)v; (void)s;
 #define glm_versor_print(v, s) (void)v; (void)s;
 #define glm_aabb_print(v, t, s) (void)v; (void)t; (void)s;
+#define glm_arch_print(s) (void)s;
+#define glm_arch_print_name(s) (void)s;
 
 #endif
 #endif /* cglm_io_h */

include/cglm/ivec2.h

@@ -23,8 +23,27 @@
   CGLM_INLINE void glm_ivec2_subs(ivec2 v, int s, ivec2 dest)
   CGLM_INLINE void glm_ivec2_mul(ivec2 a, ivec2 b, ivec2 dest)
   CGLM_INLINE void glm_ivec2_scale(ivec2 v, int s, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_addadd(ivec2 a, ivec2 b, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_addadds(ivec2 a, int s, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_subadd(ivec2 a, ivec2 b, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_subadds(ivec2 a, int s, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_muladd(ivec2 a, ivec2 b, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_muladds(ivec2 a, int s, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_maxadd(ivec2 a, ivec2 b, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_minadd(ivec2 a, ivec2 b, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_subsub(ivec2 a, ivec2 b, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_subsubs(ivec2 a, int s, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_addsub(ivec2 a, ivec2 b, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_addsubs(ivec2 a, int s, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_mulsub(ivec2 a, ivec2 b, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_mulsubs(ivec2 a, int s, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_maxsub(ivec2 a, ivec2 b, ivec2 dest)
+  CGLM_INLINE void glm_ivec2_minsub(ivec2 a, ivec2 b, ivec2 dest)
   CGLM_INLINE int glm_ivec2_distance2(ivec2 a, ivec2 b)
   CGLM_INLINE float glm_ivec2_distance(ivec2 a, ivec2 b)
+  CGLM_INLINE void glm_ivec2_fill(ivec2 v, int val);
+  CGLM_INLINE bool glm_ivec2_eq(ivec2 v, int val);
+  CGLM_INLINE bool glm_ivec2_eqv(ivec2 a, ivec2 b);
   CGLM_INLINE void glm_ivec2_maxv(ivec2 a, ivec2 b, ivec2 dest)
   CGLM_INLINE void glm_ivec2_minv(ivec2 a, ivec2 b, ivec2 dest)
   CGLM_INLINE void glm_ivec2_clamp(ivec2 v, int minVal, int maxVal)
@@ -35,6 +54,7 @@
 #define cglm_ivec2_h
 
 #include "common.h"
+#include "util.h"
 
 #define GLM_IVEC2_ONE_INIT   {1, 1}
 #define GLM_IVEC2_ZERO_INIT  {0, 0}
@@ -174,6 +194,262 @@ glm_ivec2_scale(ivec2 v, int s, ivec2 dest) {
   dest[1] = v[1] * s;
 }
 
+/*!
+ * @brief add vector [a] with vector [b] and add result to vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += (a + b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_addadd(ivec2 a, ivec2 b, ivec2 dest) {
+  dest[0] += a[0] + b[0];
+  dest[1] += a[1] + b[1];
+}
+
+/*!
+ * @brief add scalar [s] onto vector [a] and add result to vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest += (a + s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_addadds(ivec2 a, int s, ivec2 dest) {
+  dest[0] += a[0] + s;
+  dest[1] += a[1] + s;
+}
+
+/*!
+ * @brief subtract vector [a] from vector [b] and add result to [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += (a - b)  
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_subadd(ivec2 a, ivec2 b, ivec2 dest) {
+  dest[0] += a[0] - b[0];
+  dest[1] += a[1] - b[1];
+}
+
+/*!
+ * @brief subtract scalar [s] from vector [a] and add result to [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first
+ * @param[in]  s    scalar
+ * @param[out] dest dest += (a - s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_subadds(ivec2 a, int s, ivec2 dest) {
+  dest[0] += a[0] - s;
+  dest[1] += a[1] - s;
+}
+
+/*!
+ * @brief multiply vector [a] with vector [b] and add result to [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += (a * b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_muladd(ivec2 a, ivec2 b, ivec2 dest) {
+  dest[0] += a[0] * b[0];
+  dest[1] += a[1] * b[1];
+}
+
+/*!
+ * @brief multiply vector [a] with scalar [s] and add result to [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest += (a * s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_muladds(ivec2 a, int s, ivec2 dest) {
+  dest[0] += a[0] * s;
+  dest[1] += a[1] * s;
+}
+
+/*!
+ * @brief add maximum of vector [a] and vector [b] to vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += max(a, b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_maxadd(ivec2 a, ivec2 b, ivec2 dest) {
+  dest[0] += glm_imax(a[0], b[0]);
+  dest[1] += glm_imax(a[1], b[1]);
+}
+
+/*!
+ * @brief add minimum of vector [a] and vector [b] to vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += min(a, b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_minadd(ivec2 a, ivec2 b, ivec2 dest) {
+  dest[0] += glm_imin(a[0], b[0]);
+  dest[1] += glm_imin(a[1], b[1]);
+}
+
+/*!
+ * @brief subtract vector [a] from vector [b] and subtract result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest -= (a - b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_subsub(ivec2 a, ivec2 b, ivec2 dest) {
+  dest[0] -= a[0] - b[0];
+  dest[1] -= a[1] - b[1];
+}
+
+/*!
+ * @brief subtract scalar [s] from vector [a] and subtract result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest -= (a - s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_subsubs(ivec2 a, int s, ivec2 dest) {
+  dest[0] -= a[0] - s;
+  dest[1] -= a[1] - s;
+}
+
+/*!
+ * @brief add vector [a] to vector [b] and subtract the result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  b    scalar
+ * @param[out] dest dest -= (a + b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_addsub(ivec2 a, ivec2 b, ivec2 dest) {
+  dest[0] -= a[0] + b[0];
+  dest[1] -= a[1] + b[1];
+}
+
+/*!
+ * @brief add scalar [s] to vector [a] and subtract the result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest -= (a + b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_addsubs(ivec2 a, int s, ivec2 dest) {
+  dest[0] -= a[0] + s;
+  dest[1] -= a[1] + s;
+}
+
+/*!
+ * @brief multiply vector [a] and vector [b] and subtract the result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  b    scalar
+ * @param[out] dest dest -= (a * b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_mulsub(ivec2 a, ivec2 b, ivec2 dest) {
+  dest[0] -= a[0] * b[0];
+  dest[1] -= a[1] * b[1];
+}
+
+/*!
+ * @brief multiply vector [a] with scalar [s] and subtract the result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest -= (a * s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_mulsubs(ivec2 a, int s, ivec2 dest) {
+  dest[0] -= a[0] * s;
+  dest[1] -= a[1] * s;
+}
+
+/*!
+ * @brief subtract maximum of vector [a] and vector [b] from vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest -= max(a, b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_maxsub(ivec2 a, ivec2 b, ivec2 dest) {
+  dest[0] -= glm_imax(a[0], b[0]);
+  dest[1] -= glm_imax(a[1], b[1]);
+}
+
+/*!
+ * @brief subtract minimum of vector [a] and vector [b] from vector [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest -= min(a, b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec2_minsub(ivec2 a, ivec2 b, ivec2 dest) {
+  dest[0] -= glm_imin(a[0], b[0]);
+  dest[1] -= glm_imin(a[1], b[1]);
+}
+
 /*!
  * @brief squared distance between two vectors
  *
@@ -203,6 +479,44 @@ glm_ivec2_distance(ivec2 a, ivec2 b) {
   return sqrtf((float)glm_ivec2_distance2(a, b));
 }
 
+
+/*!
+ * @brief fill a vector with specified value
+ *
+ * @param[out] v   dest
+ * @param[in]  val value
+ */
+CGLM_INLINE
+void
+glm_ivec2_fill(ivec2 v, int val) {
+  v[0] = v[1] = val;
+}
+
+/*!
+ * @brief check if vector is equal to value
+ *
+ * @param[in] v   vector
+ * @param[in] val value
+ */
+CGLM_INLINE
+bool
+glm_ivec2_eq(ivec2 v, int val) {
+  return v[0] == val && v[0] == v[1];
+}
+
+/*!
+ * @brief check if vector is equal to another
+ *
+ * @param[in] a vector
+ * @param[in] b vector
+ */
+CGLM_INLINE
+bool
+glm_ivec2_eqv(ivec2 a, ivec2 b) {
+  return a[0] == b[0]
+         && a[1] == b[1];
+}
+
 /*!
  * @brief set each member of dest to greater of vector a and b
  *

include/cglm/ivec3.h

@@ -23,8 +23,27 @@
   CGLM_INLINE void glm_ivec3_subs(ivec3 v, int s, ivec3 dest)
   CGLM_INLINE void glm_ivec3_mul(ivec3 a, ivec3 b, ivec3 dest)
   CGLM_INLINE void glm_ivec3_scale(ivec3 v, int s, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_addadd(ivec3 a, ivec3 b, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_addadds(ivec3 a, int s, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_subadd(ivec3 a, ivec3 b, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_subadds(ivec3 a, int s, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_muladd(ivec3 a, ivec3 b, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_muladds(ivec3 a, int s, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_maxadd(ivec3 a, ivec3 b, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_minadd(ivec3 a, ivec3 b, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_subsub(ivec3 a, ivec3 b, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_subsubs(ivec3 a, int s, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_addsub(ivec3 a, ivec3 b, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_addsubs(ivec3 a, int s, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_mulsub(ivec3 a, ivec3 b, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_mulsubs(ivec3 a, int s, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_maxsub(ivec3 a, ivec3 b, ivec3 dest)
+  CGLM_INLINE void glm_ivec3_minsub(ivec3 a, ivec3 b, ivec3 dest)
   CGLM_INLINE int glm_ivec3_distance2(ivec3 a, ivec3 b)
   CGLM_INLINE float glm_ivec3_distance(ivec3 a, ivec3 b)
+  CGLM_INLINE void glm_ivec3_fill(ivec3 v, int val);
+  CGLM_INLINE bool glm_ivec3_eq(ivec3 v, int val);
+  CGLM_INLINE bool glm_ivec3_eqv(ivec3 a, ivec3 b);
   CGLM_INLINE void glm_ivec3_maxv(ivec3 a, ivec3 b, ivec3 dest)
   CGLM_INLINE void glm_ivec3_minv(ivec3 a, ivec3 b, ivec3 dest)
   CGLM_INLINE void glm_ivec3_clamp(ivec3 v, int minVal, int maxVal)
@@ -35,6 +54,7 @@
 #define cglm_ivec3_h
 
 #include "common.h"
+#include "util.h"
 
 #define GLM_IVEC3_ONE_INIT   {1, 1, 1}
 #define GLM_IVEC3_ZERO_INIT  {0, 0, 0}
@@ -182,6 +202,278 @@ glm_ivec3_scale(ivec3 v, int s, ivec3 dest) {
   dest[2] = v[2] * s;
 }
 
+/*!
+ * @brief add vector [a] with vector [b] and add result to vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += (a + b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_addadd(ivec3 a, ivec3 b, ivec3 dest) {
+  dest[0] += a[0] + b[0];
+  dest[1] += a[1] + b[1];
+  dest[2] += a[2] + b[2];
+}
+
+/*!
+ * @brief add scalar [s] onto vector [a] and add result to vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest += (a + s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_addadds(ivec3 a, int s, ivec3 dest) {
+  dest[0] += a[0] + s;
+  dest[1] += a[1] + s;
+  dest[2] += a[2] + s;
+}
+
+/*!
+ * @brief subtract vector [a] from vector [b] and add result to [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += (a - b)  
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_subadd(ivec3 a, ivec3 b, ivec3 dest) {
+  dest[0] += a[0] - b[0];
+  dest[1] += a[1] - b[1];
+  dest[2] += a[2] - b[2];
+}
+
+/*!
+ * @brief subtract scalar [s] from vector [a] and add result to [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first
+ * @param[in]  s    scalar
+ * @param[out] dest dest += (a - s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_subadds(ivec3 a, int s, ivec3 dest) {
+  dest[0] += a[0] - s;
+  dest[1] += a[1] - s;
+  dest[2] += a[2] - s;
+}
+
+/*!
+ * @brief multiply vector [a] with vector [b] and add result to [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += (a * b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_muladd(ivec3 a, ivec3 b, ivec3 dest) {
+  dest[0] += a[0] * b[0];
+  dest[1] += a[1] * b[1];
+  dest[2] += a[2] * b[2];
+}
+
+/*!
+ * @brief multiply vector [a] with scalar [s] and add result to [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest += (a * s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_muladds(ivec3 a, int s, ivec3 dest) {
+  dest[0] += a[0] * s;
+  dest[1] += a[1] * s;
+  dest[2] += a[2] * s;
+}
+
+/*!
+ * @brief add maximum of vector [a] and vector [b] to vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += max(a, b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_maxadd(ivec3 a, ivec3 b, ivec3 dest) {
+  dest[0] += glm_imax(a[0], b[0]);
+  dest[1] += glm_imax(a[1], b[1]);
+  dest[2] += glm_imax(a[2], b[2]);
+}
+
+/*!
+ * @brief add minimum of vector [a] and vector [b] to vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += min(a, b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_minadd(ivec3 a, ivec3 b, ivec3 dest) {
+  dest[0] += glm_imin(a[0], b[0]);
+  dest[1] += glm_imin(a[1], b[1]);
+  dest[2] += glm_imin(a[2], b[2]);
+}
+
+/*!
+ * @brief subtract vector [a] from vector [b] and subtract result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest -= (a - b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_subsub(ivec3 a, ivec3 b, ivec3 dest) {
+  dest[0] -= a[0] - b[0];
+  dest[1] -= a[1] - b[1];
+  dest[2] -= a[2] - b[2];
+}
+
+/*!
+ * @brief subtract scalar [s] from vector [a] and subtract result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest -= (a - s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_subsubs(ivec3 a, int s, ivec3 dest) {
+  dest[0] -= a[0] - s;
+  dest[1] -= a[1] - s;
+  dest[2] -= a[2] - s;
+}
+
+/*!
+ * @brief add vector [a] to vector [b] and subtract the result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  b    scalar
+ * @param[out] dest dest -= (a + b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_addsub(ivec3 a, ivec3 b, ivec3 dest) {
+  dest[0] -= a[0] + b[0];
+  dest[1] -= a[1] + b[1];
+  dest[2] -= a[2] + b[2];
+}
+
+/*!
+ * @brief add scalar [s] to vector [a] and subtract the result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest -= (a + b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_addsubs(ivec3 a, int s, ivec3 dest) {
+  dest[0] -= a[0] + s;
+  dest[1] -= a[1] + s;
+  dest[2] -= a[2] + s;
+}
+
+/*!
+ * @brief multiply vector [a] and vector [b] and subtract the result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  b    scalar
+ * @param[out] dest dest -= (a * b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_mulsub(ivec3 a, ivec3 b, ivec3 dest) {
+  dest[0] -= a[0] * b[0];
+  dest[1] -= a[1] * b[1];
+  dest[2] -= a[2] * b[2];
+}
+
+/*!
+ * @brief multiply vector [a] with scalar [s] and subtract the result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest -= (a * s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_mulsubs(ivec3 a, int s, ivec3 dest) {
+  dest[0] -= a[0] * s;
+  dest[1] -= a[1] * s;
+  dest[2] -= a[2] * s;
+}
+
+/*!
+ * @brief subtract maximum of vector [a] and vector [b] from vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest -= max(a, b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_maxsub(ivec3 a, ivec3 b, ivec3 dest) {
+  dest[0] -= glm_imax(a[0], b[0]);
+  dest[1] -= glm_imax(a[1], b[1]);
+  dest[2] -= glm_imax(a[2], b[2]);
+}
+
+/*!
+ * @brief subtract minimum of vector [a] and vector [b] from vector [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest -= min(a, b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec3_minsub(ivec3 a, ivec3 b, ivec3 dest) {
+  dest[0] -= glm_imin(a[0], b[0]);
+  dest[1] -= glm_imin(a[1], b[1]);
+  dest[2] -= glm_imin(a[2], b[2]);
+}
+
 /*!
  * @brief squared distance between two vectors
  *
@@ -212,6 +504,44 @@ glm_ivec3_distance(ivec3 a, ivec3 b) {
   return sqrtf((float)glm_ivec3_distance2(a, b));
 }
 
+/*!
+ * @brief fill a vector with specified value
+ *
+ * @param[out] v   dest
+ * @param[in]  val value
+ */
+CGLM_INLINE
+void
+glm_ivec3_fill(ivec3 v, int val) {
+  v[0] = v[1] = v[2] = val;
+}
+
+/*!
+ * @brief check if vector is equal to value
+ *
+ * @param[in] v   vector
+ * @param[in] val value
+ */
+CGLM_INLINE
+bool
+glm_ivec3_eq(ivec3 v, int val) {
+  return v[0] == val && v[0] == v[1] && v[0] == v[2];
+}
+
+/*!
+ * @brief check if vector is equal to another
+ *
+ * @param[in] a vector
+ * @param[in] b vector
+ */
+CGLM_INLINE
+bool
+glm_ivec3_eqv(ivec3 a, ivec3 b) {
+  return a[0] == b[0]
+         && a[1] == b[1]
+         && a[2] == b[2];
+}
+
 /*!
  * @brief set each member of dest to greater of vector a and b
  *

include/cglm/ivec4.h

@@ -23,6 +23,22 @@
   CGLM_INLINE void glm_ivec4_subs(ivec4 v, int s, ivec4 dest)
   CGLM_INLINE void glm_ivec4_mul(ivec4 a, ivec4 b, ivec4 dest)
   CGLM_INLINE void glm_ivec4_scale(ivec4 v, int s, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_addadd(ivec4 a, ivec4 b, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_addadds(ivec4 a, int s, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_subadd(ivec4 a, ivec4 b, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_subadds(ivec4 a, int s, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_muladd(ivec4 a, ivec4 b, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_muladds(ivec4 a, int s, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_maxadd(ivec4 a, ivec4 b, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_minadd(ivec4 a, ivec4 b, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_subsub(ivec4 a, ivec4 b, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_subsubs(ivec4 a, int s, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_addsub(ivec4 a, ivec4 b, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_addsubs(ivec4 a, int s, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_mulsub(ivec4 a, ivec4 b, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_mulsubs(ivec4 a, int s, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_maxsub(ivec4 a, ivec4 b, ivec4 dest)
+  CGLM_INLINE void glm_ivec4_minsub(ivec4 a, ivec4 b, ivec4 dest)
   CGLM_INLINE int glm_ivec4_distance2(ivec4 a, ivec4 b)
   CGLM_INLINE float glm_ivec4_distance(ivec4 a, ivec4 b)
   CGLM_INLINE void glm_ivec4_maxv(ivec4 a, ivec4 b, ivec4 dest)
@@ -35,6 +51,7 @@
 #define cglm_ivec4_h
 
 #include "common.h"
+#include "util.h"
 
 #define GLM_IVEC4_ONE_INIT   {1, 1, 1, 1}
 #define GLM_IVEC4_ZERO_INIT  {0, 0, 0, 0}
@@ -191,6 +208,294 @@ glm_ivec4_scale(ivec4 v, int s, ivec4 dest) {
   dest[3] = v[3] * s;
 }
 
+/*!
+ * @brief add vector [a] with vector [b] and add result to vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += (a + b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_addadd(ivec4 a, ivec4 b, ivec4 dest) {
+  dest[0] += a[0] + b[0];
+  dest[1] += a[1] + b[1];
+  dest[2] += a[2] + b[2];
+  dest[3] += a[3] + b[3];
+}
+
+/*!
+ * @brief add scalar [s] onto vector [a] and add result to vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest += (a + s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_addadds(ivec4 a, int s, ivec4 dest) {
+  dest[0] += a[0] + s;
+  dest[1] += a[1] + s;
+  dest[2] += a[2] + s;
+  dest[3] += a[3] + s;
+}
+
+/*!
+ * @brief subtract vector [a] from vector [b] and add result to [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += (a - b)  
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_subadd(ivec4 a, ivec4 b, ivec4 dest) {
+  dest[0] += a[0] - b[0];
+  dest[1] += a[1] - b[1];
+  dest[2] += a[2] - b[2];
+  dest[3] += a[3] - b[3];
+}
+
+/*!
+ * @brief subtract scalar [s] from vector [a] and add result to [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first
+ * @param[in]  s    scalar
+ * @param[out] dest dest += (a - s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_subadds(ivec4 a, int s, ivec4 dest) {
+  dest[0] += a[0] - s;
+  dest[1] += a[1] - s;
+  dest[2] += a[2] - s;
+  dest[3] += a[3] - s;
+}
+
+/*!
+ * @brief multiply vector [a] with vector [b] and add result to [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += (a * b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_muladd(ivec4 a, ivec4 b, ivec4 dest) {
+  dest[0] += a[0] * b[0];
+  dest[1] += a[1] * b[1];
+  dest[2] += a[2] * b[2];
+  dest[3] += a[3] * b[3];
+}
+
+/*!
+ * @brief multiply vector [a] with scalar [s] and add result to [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest += (a * s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_muladds(ivec4 a, int s, ivec4 dest) {
+  dest[0] += a[0] * s;
+  dest[1] += a[1] * s;
+  dest[2] += a[2] * s;
+  dest[3] += a[3] * s;
+}
+
+/*!
+ * @brief add maximum of vector [a] and vector [b] to vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += max(a, b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_maxadd(ivec4 a, ivec4 b, ivec4 dest) {
+  dest[0] += glm_imax(a[0], b[0]);
+  dest[1] += glm_imax(a[1], b[1]);
+  dest[2] += glm_imax(a[2], b[2]);
+  dest[3] += glm_imax(a[3], b[3]);
+}
+
+/*!
+ * @brief add minimum of vector [a] and vector [b] to vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest += min(a, b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_minadd(ivec4 a, ivec4 b, ivec4 dest) {
+  dest[0] += glm_imin(a[0], b[0]);
+  dest[1] += glm_imin(a[1], b[1]);
+  dest[2] += glm_imin(a[2], b[2]);
+  dest[3] += glm_imin(a[3], b[3]);
+}
+
+/*!
+ * @brief subtract vector [a] from vector [b] and subtract result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest -= (a - b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_subsub(ivec4 a, ivec4 b, ivec4 dest) {
+  dest[0] -= a[0] - b[0];
+  dest[1] -= a[1] - b[1];
+  dest[2] -= a[2] - b[2];
+  dest[3] -= a[3] - b[3];
+}
+
+/*!
+ * @brief subtract scalar [s] from vector [a] and subtract result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest -= (a - s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_subsubs(ivec4 a, int s, ivec4 dest) {
+  dest[0] -= a[0] - s;
+  dest[1] -= a[1] - s;
+  dest[2] -= a[2] - s;
+  dest[3] -= a[3] - s;
+}
+
+/*!
+ * @brief add vector [a] to vector [b] and subtract the result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  b    scalar
+ * @param[out] dest dest -= (a + b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_addsub(ivec4 a, ivec4 b, ivec4 dest) {
+  dest[0] -= a[0] + b[0];
+  dest[1] -= a[1] + b[1];
+  dest[2] -= a[2] + b[2];
+  dest[3] -= a[3] + b[3];
+}
+
+/*!
+ * @brief add scalar [s] to vector [a] and subtract the result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest -= (a + b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_addsubs(ivec4 a, int s, ivec4 dest) {
+  dest[0] -= a[0] + s;
+  dest[1] -= a[1] + s;
+  dest[2] -= a[2] + s;
+  dest[3] -= a[3] + s;
+}
+
+/*!
+ * @brief multiply vector [a] and vector [b] and subtract the result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  b    scalar
+ * @param[out] dest dest -= (a * b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_mulsub(ivec4 a, ivec4 b, ivec4 dest) {
+  dest[0] -= a[0] * b[0];
+  dest[1] -= a[1] * b[1];
+  dest[2] -= a[2] * b[2];
+  dest[3] -= a[3] * b[3];
+}
+
+/*!
+ * @brief multiply vector [a] with scalar [s] and subtract the result from [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    vector
+ * @param[in]  s    scalar
+ * @param[out] dest dest -= (a * s)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_mulsubs(ivec4 a, int s, ivec4 dest) {
+  dest[0] -= a[0] * s;
+  dest[1] -= a[1] * s;
+  dest[2] -= a[2] * s;
+  dest[3] -= a[3] * s;
+}
+
+/*!
+ * @brief subtract maximum of vector [a] and vector [b] from vector [dest]
+ *
+ * applies += operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest -= max(a, b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_maxsub(ivec4 a, ivec4 b, ivec4 dest) {
+  dest[0] -= glm_imax(a[0], b[0]);
+  dest[1] -= glm_imax(a[1], b[1]);
+  dest[2] -= glm_imax(a[2], b[2]);
+  dest[3] -= glm_imax(a[3], b[3]);
+}
+
+/*!
+ * @brief subtract minimum of vector [a] and vector [b] from vector [dest]
+ *
+ * applies -= operator so dest must be initialized
+ *
+ * @param[in]  a    first vector
+ * @param[in]  b    second vector
+ * @param[out] dest dest -= min(a, b)
+ */
+CGLM_INLINE 
+void 
+glm_ivec4_minsub(ivec4 a, ivec4 b, ivec4 dest) {
+  dest[0] -= glm_imin(a[0], b[0]);
+  dest[1] -= glm_imin(a[1], b[1]);
+  dest[2] -= glm_imin(a[2], b[2]);
+  dest[3] -= glm_imin(a[3], b[3]);
+}
+
 /*!
  * @brief squared distance between two vectors
  *

include/cglm/mat2.h

@@ -28,6 +28,7 @@
    CGLM_INLINE void  glm_mat2_swap_col(mat2 mat, int col1, int col2)
    CGLM_INLINE void  glm_mat2_swap_row(mat2 mat, int row1, int row2)
    CGLM_INLINE float glm_mat2_rmc(vec2 r, mat2 m, vec2 c)
+   CGLM_INLINE void  glm_mat2_make(float * restrict src, mat2 dest)
  */
 
 #ifndef cglm_mat2_h
@@ -44,6 +45,10 @@
 #  include "simd/neon/mat2.h"
 #endif
 
+#ifdef CGLM_SIMD_WASM
+#  include "simd/wasm/mat2.h"
+#endif
+
 #define GLM_MAT2_IDENTITY_INIT  {{1.0f, 0.0f}, {0.0f, 1.0f}}
 #define GLM_MAT2_ZERO_INIT      {{0.0f, 0.0f}, {0.0f, 0.0f}}
 
@@ -132,7 +137,9 @@ glm_mat2_zero(mat2 mat) {
 CGLM_INLINE
 void
 glm_mat2_mul(mat2 m1, mat2 m2, mat2 dest) {
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_mat2_mul_wasm(m1, m2, dest);
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_mat2_mul_sse2(m1, m2, dest);
 #elif defined(CGLM_NEON_FP)
   glm_mat2_mul_neon(m1, m2, dest);
@@ -160,7 +167,9 @@ glm_mat2_mul(mat2 m1, mat2 m2, mat2 dest) {
 CGLM_INLINE
 void
 glm_mat2_transpose_to(mat2 m, mat2 dest) {
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_mat2_transp_wasm(m, dest);
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_mat2_transp_sse2(m, dest);
 #else
   dest[0][0] = m[0][0];
@@ -222,7 +231,10 @@ glm_mat2_trace(mat2 m) {
 CGLM_INLINE
 void
 glm_mat2_scale(mat2 m, float s) {
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glmm_store(m[0], wasm_f32x4_mul(wasm_v128_load(m[0]),
+                                  wasm_f32x4_splat(s)));
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   glmm_store(m[0], _mm_mul_ps(_mm_loadu_ps(m[0]), _mm_set1_ps(s)));
 #elif defined(CGLM_NEON_FP)
   vst1q_f32(m[0], vmulq_f32(vld1q_f32(m[0]), vdupq_n_f32(s)));
@@ -334,4 +346,19 @@ glm_mat2_rmc(vec2 r, mat2 m, vec2 c) {
   return glm_vec2_dot(r, tmp);
 }
 
+/*!
+ * @brief Create mat2 matrix from pointer
+ *
+ * @param[in]  src  pointer to an array of floats
+ * @param[out] dest matrix
+ */
+CGLM_INLINE
+void
+glm_mat2_make(float * __restrict src, mat2 dest) {
+  dest[0][0] = src[0];
+  dest[0][1] = src[1];
+  dest[1][0] = src[2];
+  dest[1][1] = src[3];
+}
+
 #endif /* cglm_mat2_h */

include/cglm/mat2x3.h

@@ -0,0 +1,154 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+/*
+ Macros:
+   GLM_MAT2X3_ZERO_INIT
+   GLM_MAT2X3_ZERO
+
+ Functions:
+   CGLM_INLINE void glm_mat2x3_copy(mat2x3 mat, mat2x3 dest);
+   CGLM_INLINE void glm_mat2x3_zero(mat2x3 mat);
+   CGLM_INLINE void glm_mat2x3_make(float * __restrict src, mat2x3 dest);
+   CGLM_INLINE void glm_mat2x3_mul(mat2x3 m1, mat3x2 m2, mat2 dest);
+   CGLM_INLINE void glm_mat2x3_mulv(mat2x3 m, vec3 v, vec2 dest);
+   CGLM_INLINE void glm_mat2x3_transpose(mat2x3 m, mat3x2 dest);
+   CGLM_INLINE void glm_mat2x3_scale(mat2x3 m, float s);
+ */
+
+#ifndef cglm_mat2x3_h
+#define cglm_mat2x3_h
+
+#include "common.h"
+
+#define GLM_MAT2X3_ZERO_INIT {{0.0f, 0.0f, 0.0f}, {0.0f, 0.0f, 0.0f}}
+
+/* for C only */
+#define GLM_MAT2X3_ZERO GLM_MAT2X3_ZERO_INIT
+
+/*!
+ * @brief copy all members of [mat] to [dest]
+ *
+ * @param[in]  mat  source
+ * @param[out] dest destination
+ */
+CGLM_INLINE
+void
+glm_mat2x3_copy(mat2x3 mat, mat2x3 dest) {
+  dest[0][0] = mat[0][0];
+  dest[0][1] = mat[0][1];
+  dest[0][2] = mat[0][2];
+
+  dest[1][0] = mat[1][0];
+  dest[1][1] = mat[1][1];
+  dest[1][2] = mat[1][2];
+}
+
+/*!
+ * @brief make given matrix zero.
+ *
+ * @param[in, out]  mat  matrix
+ */
+CGLM_INLINE
+void
+glm_mat2x3_zero(mat2x3 mat) {
+  CGLM_ALIGN_MAT mat2x3 t = GLM_MAT2X3_ZERO_INIT;
+  glm_mat2x3_copy(t, mat);
+}
+
+/*!
+ * @brief Create mat2x3 matrix from pointer
+ *
+ * @param[in]  src  pointer to an array of floats
+ * @param[out] dest matrix
+ */
+CGLM_INLINE
+void
+glm_mat2x3_make(float * __restrict src, mat2x3 dest) {
+  dest[0][0] = src[0];
+  dest[0][1] = src[1];
+  dest[0][2] = src[2];
+
+  dest[1][0] = src[3];
+  dest[1][1] = src[4];
+  dest[1][2] = src[5];
+}
+
+/*!
+ * @brief multiply m1 and m2 to dest
+ *
+ * @code
+ * glm_mat2x3_mul(mat2x3, mat3x2, mat2);
+ * @endcode
+ *
+ * @param[in]  m1   left matrix (mat2x3)
+ * @param[in]  m2   right matrix (mat3x2)
+ * @param[out] dest destination matrix (mat2)
+ */
+CGLM_INLINE
+void
+glm_mat2x3_mul(mat2x3 m1, mat3x2 m2, mat2 dest) {
+  float a00 = m1[0][0], a01 = m1[0][1], a02 = m1[0][2],
+        a10 = m1[1][0], a11 = m1[1][1], a12 = m1[1][2],
+
+        b00 = m2[0][0], b01 = m2[0][1],
+        b10 = m2[1][0], b11 = m2[1][1],
+        b20 = m2[2][0], b21 = m2[2][1];
+
+  dest[0][0] = a00 * b00 + a01 * b10 + a02 * b20;
+  dest[0][1] = a00 * b01 + a01 * b11 + a02 * b21;
+
+  dest[1][0] = a10 * b00 + a11 * b10 + a12 * b20;
+  dest[1][1] = a10 * b01 + a11 * b11 + a12 * b21;
+}
+
+/*!
+ * @brief multiply matrix with column vector and store in dest vector
+ *
+ * @param[in]  m    matrix (left)
+ * @param[in]  v    vector (right, column vector)
+ * @param[out] dest result vector
+ */
+CGLM_INLINE
+void
+glm_mat2x3_mulv(mat2x3 m, vec3 v, vec2 dest) {
+  float v0 = v[0], v1 = v[1], v2 = v[2];
+
+  dest[0] = m[0][0] * v0 + m[0][1] * v1 + m[0][2] * v2;
+  dest[1] = m[1][0] * v0 + m[1][1] * v1 + m[1][2] * v2;
+}
+
+/*!
+ * @brief transpose matrix and store in dest
+ *
+ * @param[in]  m     matrix
+ * @param[out] dest  result
+ */
+CGLM_INLINE
+void
+glm_mat2x3_transpose(mat2x3 m, mat3x2 dest) {
+  dest[0][0] = m[0][0];  dest[0][1] = m[1][0];
+  dest[1][0] = m[0][1];  dest[1][1] = m[1][1];
+  dest[2][0] = m[0][2];  dest[2][1] = m[1][2];
+}
+
+/*!
+ * @brief scale (multiply with scalar) matrix
+ *
+ * multiply matrix with scalar
+ *
+ * @param[in, out] m matrix
+ * @param[in]    s scalar
+ */
+CGLM_INLINE
+void
+glm_mat2x3_scale(mat2x3 m, float s) {
+  m[0][0] *= s;  m[0][1] *= s;  m[0][2] *= s;
+  m[1][0] *= s;  m[1][1] *= s;  m[1][2] *= s;
+}
+
+#endif

include/cglm/mat2x4.h

@@ -0,0 +1,154 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+/*
+ Macros:
+   GLM_MAT2X4_ZERO_INIT
+   GLM_MAT2X4_ZERO
+
+ Functions:
+   CGLM_INLINE void glm_mat2x4_copy(mat2x4 mat, mat2x4 dest);
+   CGLM_INLINE void glm_mat2x4_zero(mat2x4 mat);
+   CGLM_INLINE void glm_mat2x4_make(float * __restrict src, mat2x4 dest);
+   CGLM_INLINE void glm_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat2 dest);
+   CGLM_INLINE void glm_mat2x4_mulv(mat2x4 m, vec4 v, vec2 dest);
+   CGLM_INLINE void glm_mat2x4_transpose(mat2x4 m, mat4x2 dest);
+   CGLM_INLINE void glm_mat2x4_scale(mat2x4 m, float s);
+ */
+
+#ifndef cglm_mat2x4_h
+#define cglm_mat2x4_h
+
+#include "common.h"
+#include "vec4.h"
+
+#define GLM_MAT2X4_ZERO_INIT {{0.0f, 0.0f, 0.0f, 0.0f}, {0.0f, 0.0f, 0.0f, 0.0f}}
+
+/* for C only */
+#define GLM_MAT2X4_ZERO GLM_MAT2X4_ZERO_INIT
+
+/*!
+ * @brief copy all members of [mat] to [dest]
+ *
+ * @param[in]  mat  source
+ * @param[out] dest destination
+ */
+CGLM_INLINE
+void
+glm_mat2x4_copy(mat2x4 mat, mat2x4 dest) {
+  glm_vec4_ucopy(mat[0], dest[0]);
+  glm_vec4_ucopy(mat[1], dest[1]);
+}
+
+/*!
+ * @brief make given matrix zero.
+ *
+ * @param[in, out]  mat  matrix
+ */
+CGLM_INLINE
+void
+glm_mat2x4_zero(mat2x4 mat) {
+  CGLM_ALIGN_MAT mat2x4 t = GLM_MAT2X4_ZERO_INIT;
+  glm_mat2x4_copy(t, mat);
+}
+
+/*!
+ * @brief Create mat2x4 matrix from pointer
+ *
+ * @param[in]  src  pointer to an array of floats
+ * @param[out] dest matrix
+ */
+CGLM_INLINE
+void
+glm_mat2x4_make(float * __restrict src, mat2x4 dest) {
+  dest[0][0] = src[0];
+  dest[0][1] = src[1];
+  dest[0][2] = src[2];
+  dest[0][3] = src[3];
+
+  dest[1][0] = src[4];
+  dest[1][1] = src[5];
+  dest[1][2] = src[6];
+  dest[1][3] = src[7];
+}
+
+/*!
+ * @brief multiply m1 and m2 to dest
+ *
+ * @code
+ * glm_mat2x4_mul(mat2x4, mat4x2, mat2);
+ * @endcode
+ *
+ * @param[in]  m1   left matrix (mat2x4)
+ * @param[in]  m2   right matrix (mat4x2)
+ * @param[out] dest destination matrix (mat2)
+ */
+CGLM_INLINE
+void
+glm_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat2 dest) {
+  float a00 = m1[0][0], a01 = m1[0][1], a02 = m1[0][2], a03 = m1[0][3],
+        a10 = m1[1][0], a11 = m1[1][1], a12 = m1[1][2], a13 = m1[1][3],
+
+        b00 = m2[0][0], b01 = m2[0][1],
+        b10 = m2[1][0], b11 = m2[1][1],
+        b20 = m2[2][0], b21 = m2[2][1],
+        b30 = m2[3][0], b31 = m2[3][1];
+
+  dest[0][0] = a00 * b00 + a01 * b10 + a02 * b20 + a03 * b30;
+  dest[0][1] = a00 * b01 + a01 * b11 + a02 * b21 + a03 * b31;
+
+  dest[1][0] = a10 * b00 + a11 * b10 + a12 * b20 + a13 * b30;
+  dest[1][1] = a10 * b01 + a11 * b11 + a12 * b21 + a13 * b31;
+}
+
+/*!
+ * @brief multiply matrix with column vector and store in dest vector
+ *
+ * @param[in]  m    matrix (left)
+ * @param[in]  v    vector (right, column vector)
+ * @param[out] dest result vector
+ */
+CGLM_INLINE
+void
+glm_mat2x4_mulv(mat2x4 m, vec4 v, vec2 dest) {
+  float v0 = v[0], v1 = v[1], v2 = v[2], v3 = v[3];
+
+  dest[0] = m[0][0] * v0 + m[0][1] * v1 + m[0][2] * v2 + m[0][3] * v3;
+  dest[1] = m[1][0] * v0 + m[1][1] * v1 + m[1][2] * v2 + m[1][3] * v3;
+}
+
+/*!
+ * @brief transpose matrix and store in dest
+ *
+ * @param[in]  m     matrix
+ * @param[out] dest  result
+ */
+CGLM_INLINE
+void
+glm_mat2x4_transpose(mat2x4 m, mat4x2 dest) {
+  dest[0][0] = m[0][0];  dest[0][1] = m[1][0];
+  dest[1][0] = m[0][1];  dest[1][1] = m[1][1];
+  dest[2][0] = m[0][2];  dest[2][1] = m[1][2];
+  dest[3][0] = m[0][3];  dest[3][1] = m[1][3];
+}
+
+/*!
+ * @brief scale (multiply with scalar) matrix
+ *
+ * multiply matrix with scalar
+ *
+ * @param[in, out] m matrix
+ * @param[in]    s scalar
+ */
+CGLM_INLINE
+void
+glm_mat2x4_scale(mat2x4 m, float s) {
+  m[0][0] *= s;  m[0][1] *= s;  m[0][2] *= s;  m[0][3] *= s;
+  m[1][0] *= s;  m[1][1] *= s;  m[1][2] *= s;  m[1][3] *= s;
+}
+
+#endif

include/cglm/mat3.h

@@ -30,6 +30,7 @@
    CGLM_INLINE void  glm_mat3_swap_col(mat3 mat, int col1, int col2);
    CGLM_INLINE void  glm_mat3_swap_row(mat3 mat, int row1, int row2);
    CGLM_INLINE float glm_mat3_rmc(vec3 r, mat3 m, vec3 c);
+   CGLM_INLINE void  glm_mat3_make(float * restrict src, mat3 dest);
  */
 
 #ifndef cglm_mat3_h
@@ -42,6 +43,10 @@
 #  include "simd/sse2/mat3.h"
 #endif
 
+#ifdef CGLM_SIMD_WASM
+#  include "simd/wasm/mat3.h"
+#endif
+
 #define GLM_MAT3_IDENTITY_INIT  {{1.0f, 0.0f, 0.0f},                          \
                                  {0.0f, 1.0f, 0.0f},                          \
                                  {0.0f, 0.0f, 1.0f}}
@@ -148,7 +153,9 @@ glm_mat3_zero(mat3 mat) {
 CGLM_INLINE
 void
 glm_mat3_mul(mat3 m1, mat3 m2, mat3 dest) {
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_mat3_mul_wasm(m1, m2, dest);
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_mat3_mul_sse2(m1, m2, dest);
 #else
   float a00 = m1[0][0], a01 = m1[0][1], a02 = m1[0][2],
@@ -421,4 +428,26 @@ glm_mat3_rmc(vec3 r, mat3 m, vec3 c) {
   return glm_vec3_dot(r, tmp);
 }
 
+/*!
+ * @brief Create mat3 matrix from pointer
+ *
+ * @param[in]  src  pointer to an array of floats
+ * @param[out] dest matrix
+ */
+CGLM_INLINE
+void
+glm_mat3_make(float * __restrict src, mat3 dest) {
+  dest[0][0] = src[0];
+  dest[0][1] = src[1];
+  dest[0][2] = src[2];
+
+  dest[1][0] = src[3];
+  dest[1][1] = src[4];
+  dest[1][2] = src[5];
+
+  dest[2][0] = src[6];
+  dest[2][1] = src[7];
+  dest[2][2] = src[8];
+}
+
 #endif /* cglm_mat3_h */

include/cglm/mat3x2.h

@@ -0,0 +1,162 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+/*
+ Macros:
+   GLM_MAT3X2_ZERO_INIT
+   GLM_MAT3X2_ZERO
+
+ Functions:
+   CGLM_INLINE void glm_mat3x2_copy(mat3x2 mat, mat3x2 dest);
+   CGLM_INLINE void glm_mat3x2_zero(mat3x2 mat);
+   CGLM_INLINE void glm_mat3x2_make(float * __restrict src, mat3x2 dest);
+   CGLM_INLINE void glm_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat3 dest);
+   CGLM_INLINE void glm_mat3x2_mulv(mat3x2 m, vec2 v, vec3 dest);
+   CGLM_INLINE void glm_mat3x2_transpose(mat3x2 m, mat2x3 dest);
+   CGLM_INLINE void glm_mat3x2_scale(mat3x2 m, float s);
+ */
+
+#ifndef cglm_mat3x2_h
+#define cglm_mat3x2_h
+
+#include "common.h"
+
+#define GLM_MAT3X2_ZERO_INIT {{0.0f, 0.0f}, {0.0f, 0.0f}, {0.0f, 0.0f}}
+
+/* for C only */
+#define GLM_MAT3X2_ZERO GLM_MAT3X2_ZERO_INIT
+
+/*!
+ * @brief copy all members of [mat] to [dest]
+ *
+ * @param[in]  mat  source
+ * @param[out] dest destination
+ */
+CGLM_INLINE
+void
+glm_mat3x2_copy(mat3x2 mat, mat3x2 dest) {
+  dest[0][0] = mat[0][0];
+  dest[0][1] = mat[0][1];
+
+  dest[1][0] = mat[1][0];
+  dest[1][1] = mat[1][1];
+
+  dest[2][0] = mat[2][0];
+  dest[2][1] = mat[2][1];
+}
+
+/*!
+ * @brief make given matrix zero.
+ *
+ * @param[in, out]  mat  matrix
+ */
+CGLM_INLINE
+void
+glm_mat3x2_zero(mat3x2 mat) {
+  CGLM_ALIGN_MAT mat3x2 t = GLM_MAT3X2_ZERO_INIT;
+  glm_mat3x2_copy(t, mat);
+}
+
+/*!
+ * @brief Create mat3x2 matrix from pointer
+ *
+ * @param[in]  src  pointer to an array of floats
+ * @param[out] dest matrix
+ */
+CGLM_INLINE
+void
+glm_mat3x2_make(float * __restrict src, mat3x2 dest) {
+  dest[0][0] = src[0];
+  dest[0][1] = src[1];
+
+  dest[1][0] = src[2];
+  dest[1][1] = src[3];
+
+  dest[2][0] = src[4];
+  dest[2][1] = src[5];
+}
+
+/*!
+ * @brief multiply m1 and m2 to dest
+ *
+ * @code
+ * glm_mat3x2_mul(mat3x2, mat2x3, mat3);
+ * @endcode
+ *
+ * @param[in]  m1   left matrix (mat3x2)
+ * @param[in]  m2   right matrix (mat2x3)
+ * @param[out] dest destination matrix (mat3)
+ */
+CGLM_INLINE
+void
+glm_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat3 dest) {
+  float a00 = m1[0][0], a01 = m1[0][1],
+        a10 = m1[1][0], a11 = m1[1][1],
+        a20 = m1[2][0], a21 = m1[2][1],
+
+        b00 = m2[0][0], b01 = m2[0][1], b02 = m2[0][2],
+        b10 = m2[1][0], b11 = m2[1][1], b12 = m2[1][2];
+
+  dest[0][0] = a00 * b00 + a01 * b10;
+  dest[0][1] = a00 * b01 + a01 * b11;
+  dest[0][2] = a00 * b02 + a01 * b12;
+
+  dest[1][0] = a10 * b00 + a11 * b10;
+  dest[1][1] = a10 * b01 + a11 * b11;
+  dest[1][2] = a10 * b02 + a11 * b12;
+
+  dest[2][0] = a20 * b00 + a21 * b10;
+  dest[2][1] = a20 * b01 + a21 * b11;
+  dest[2][2] = a20 * b02 + a21 * b12;
+}
+
+/*!
+ * @brief multiply matrix with column vector and store in dest vector
+ *
+ * @param[in]  m    matrix (left)
+ * @param[in]  v    vector (right, column vector)
+ * @param[out] dest result vector
+ */
+CGLM_INLINE
+void
+glm_mat3x2_mulv(mat3x2 m, vec2 v, vec3 dest) {
+  float v0 = v[0], v1 = v[1];
+
+  dest[0] = m[0][0] * v0 + m[0][1] * v1;
+  dest[1] = m[1][0] * v0 + m[1][1] * v1;
+  dest[2] = m[2][0] * v0 + m[2][1] * v1;
+}
+
+/*!
+ * @brief transpose matrix and store in dest
+ *
+ * @param[in]  m     matrix
+ * @param[out] dest  result
+ */
+CGLM_INLINE
+void
+glm_mat3x2_transpose(mat3x2 m, mat2x3 dest) {
+  dest[0][0] = m[0][0];  dest[0][1] = m[1][0];  dest[0][2] = m[2][0];
+  dest[1][0] = m[0][1];  dest[1][1] = m[1][1];  dest[1][2] = m[2][1];
+}
+
+/*!
+ * @brief scale (multiply with scalar) matrix
+ *
+ * multiply matrix with scalar
+ *
+ * @param[in, out] m matrix
+ * @param[in]    s scalar
+ */
+CGLM_INLINE
+void
+glm_mat3x2_scale(mat3x2 m, float s) {
+  m[0][0] *= s;  m[0][1] *= s;  m[1][0] *= s;
+  m[1][1] *= s;  m[2][0] *= s;  m[2][1] *= s;
+}
+
+#endif

include/cglm/mat3x4.h

@@ -0,0 +1,171 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+/*
+ Macros:
+   GLM_MAT3X4_ZERO_INIT
+   GLM_MAT3X4_ZERO
+
+ Functions:
+   CGLM_INLINE void glm_mat3x4_copy(mat3x4 mat, mat3x4 dest);
+   CGLM_INLINE void glm_mat3x4_zero(mat3x4 mat);
+   CGLM_INLINE void glm_mat3x4_make(float * __restrict src, mat3x4 dest);
+   CGLM_INLINE void glm_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat3 dest);
+   CGLM_INLINE void glm_mat3x4_mulv(mat3x4 m, vec4 v, vec3 dest);
+   CGLM_INLINE void glm_mat3x4_transpose(mat3x4 m, mat4x3 dest);
+   CGLM_INLINE void glm_mat3x4_scale(mat3x4 m, float s);
+ */
+
+#ifndef cglm_mat3x4_h
+#define cglm_mat3x4_h
+
+#include "common.h"
+
+#define GLM_MAT3X4_ZERO_INIT {{0.0f, 0.0f, 0.0f, 0.0f}, \
+                              {0.0f, 0.0f, 0.0f, 0.0f}, \
+                              {0.0f, 0.0f, 0.0f, 0.0f}}
+
+/* for C only */
+#define GLM_MAT3X4_ZERO GLM_MAT3X4_ZERO_INIT
+
+/*!
+ * @brief copy all members of [mat] to [dest]
+ *
+ * @param[in]  mat  source
+ * @param[out] dest destination
+ */
+CGLM_INLINE
+void
+glm_mat3x4_copy(mat3x4 mat, mat3x4 dest) {
+  glm_vec4_ucopy(mat[0], dest[0]);
+  glm_vec4_ucopy(mat[1], dest[1]);
+  glm_vec4_ucopy(mat[2], dest[2]);
+}
+
+/*!
+ * @brief make given matrix zero.
+ *
+ * @param[in, out]  mat  matrix
+ */
+CGLM_INLINE
+void
+glm_mat3x4_zero(mat3x4 mat) {
+  CGLM_ALIGN_MAT mat3x4 t = GLM_MAT3X4_ZERO_INIT;
+  glm_mat3x4_copy(t, mat);
+}
+
+/*!
+ * @brief Create mat3x4 matrix from pointer
+ *
+ * @param[in]  src  pointer to an array of floats
+ * @param[out] dest matrix
+ */
+CGLM_INLINE
+void
+glm_mat3x4_make(float * __restrict src, mat3x4 dest) {
+  dest[0][0] = src[0];
+  dest[0][1] = src[1];
+  dest[0][2] = src[2];
+  dest[0][3] = src[3];
+
+  dest[1][0] = src[4];
+  dest[1][1] = src[5];
+  dest[1][2] = src[6];
+  dest[1][3] = src[7];
+
+  dest[2][0] = src[8];
+  dest[2][1] = src[9];
+  dest[2][2] = src[10];
+  dest[2][3] = src[11];
+}
+
+/*!
+ * @brief multiply m1 and m2 to dest
+ *
+ * @code
+ * glm_mat3x4_mul(mat3x4, mat4x3, mat3);
+ * @endcode
+ *
+ * @param[in]  m1   left matrix (mat3x4)
+ * @param[in]  m2   right matrix (mat4x3)
+ * @param[out] dest destination matrix (mat3)
+ */
+CGLM_INLINE
+void
+glm_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat3 dest) {
+  float a00 = m1[0][0], a01 = m1[0][1], a02 = m1[0][2], a03 = m1[0][3],
+        a10 = m1[1][0], a11 = m1[1][1], a12 = m1[1][2], a13 = m1[1][3],
+        a20 = m1[2][0], a21 = m1[2][1], a22 = m1[2][2], a23 = m1[2][3],
+
+        b00 = m2[0][0], b01 = m2[0][1], b02 = m2[0][2],
+        b10 = m2[1][0], b11 = m2[1][1], b12 = m2[1][2],
+        b20 = m2[2][0], b21 = m2[2][1], b22 = m2[2][2],
+        b30 = m2[3][0], b31 = m2[3][1], b32 = m2[3][2];
+
+  dest[0][0] = a00 * b00 + a01 * b10 + a02 * b20 + a03 * b30;
+  dest[0][1] = a00 * b01 + a01 * b11 + a02 * b21 + a03 * b31;
+  dest[0][2] = a00 * b02 + a01 * b12 + a02 * b22 + a03 * b32;
+
+  dest[1][0] = a10 * b00 + a11 * b10 + a12 * b20 + a13 * b30;
+  dest[1][1] = a10 * b01 + a11 * b11 + a12 * b21 + a13 * b31;
+  dest[1][2] = a10 * b02 + a11 * b12 + a12 * b22 + a13 * b32;
+
+  dest[2][0] = a20 * b00 + a21 * b10 + a22 * b20 + a23 * b30;
+  dest[2][1] = a20 * b01 + a21 * b11 + a22 * b21 + a23 * b31;
+  dest[2][2] = a20 * b02 + a21 * b12 + a22 * b22 + a23 * b32;
+}
+
+/*!
+ * @brief multiply matrix with column vector and store in dest vector
+ *
+ * @param[in]  m    matrix (left)
+ * @param[in]  v    vector (right, column vector)
+ * @param[out] dest result vector
+ */
+CGLM_INLINE
+void
+glm_mat3x4_mulv(mat3x4 m, vec4 v, vec3 dest) {
+  float v0 = v[0], v1 = v[1], v2 = v[2], v3 = v[3];
+
+  dest[0] = m[0][0] * v0 + m[0][1] * v1 + m[0][2] * v2 + m[0][3] * v3;
+  dest[1] = m[1][0] * v0 + m[1][1] * v1 + m[1][2] * v2 + m[1][3] * v3;
+  dest[2] = m[2][0] * v0 + m[2][1] * v1 + m[2][2] * v2 + m[2][3] * v3;
+}
+
+/*!
+ * @brief transpose matrix and store in dest
+ *
+ * @param[in]  m     matrix
+ * @param[out] dest  result
+ */
+CGLM_INLINE
+void
+glm_mat3x4_transpose(mat3x4 m, mat4x3 dest) {
+  dest[0][0] = m[0][0];  dest[0][1] = m[1][0];  dest[0][2] = m[2][0];
+  dest[1][0] = m[0][1];  dest[1][1] = m[1][1];  dest[1][2] = m[2][1];
+  dest[2][0] = m[0][2];  dest[2][1] = m[1][2];  dest[2][2] = m[2][2];
+  dest[3][0] = m[0][3];  dest[3][1] = m[1][3];  dest[3][2] = m[2][3];
+}
+
+/*!
+ * @brief scale (multiply with scalar) matrix
+ *
+ * multiply matrix with scalar
+ *
+ * @param[in, out] m matrix
+ * @param[in]    s scalar
+ */
+CGLM_INLINE
+void
+glm_mat3x4_scale(mat3x4 m, float s) {
+  m[0][0] *= s;  m[1][0] *= s;   m[2][0] *= s;
+  m[0][1] *= s;  m[1][1] *= s;   m[2][1] *= s;
+  m[0][2] *= s;  m[1][2] *= s;   m[2][2] *= s;
+  m[0][3] *= s;  m[1][3] *= s;   m[2][3] *= s;
+}
+
+#endif

include/cglm/mat4.h

@@ -29,7 +29,7 @@
    CGLM_INLINE void  glm_mat4_mul(mat4 m1, mat4 m2, mat4 dest);
    CGLM_INLINE void  glm_mat4_mulN(mat4 *matrices[], int len, mat4 dest);
    CGLM_INLINE void  glm_mat4_mulv(mat4 m, vec4 v, vec4 dest);
-   CGLM_INLINE void  glm_mat4_mulv3(mat4 m, vec3 v, vec3 dest);
+   CGLM_INLINE void  glm_mat4_mulv3(mat4 m, vec3 v, float last, vec3 dest);
    CGLM_INLINE float glm_mat4_trace(mat4 m);
    CGLM_INLINE float glm_mat4_trace3(mat4 m);
    CGLM_INLINE void  glm_mat4_quat(mat4 m, versor dest) ;
@@ -43,6 +43,7 @@
    CGLM_INLINE void  glm_mat4_swap_col(mat4 mat, int col1, int col2);
    CGLM_INLINE void  glm_mat4_swap_row(mat4 mat, int row1, int row2);
    CGLM_INLINE float glm_mat4_rmc(vec4 r, mat4 m, vec4 c);
+   CGLM_INLINE void  glm_mat4_make(float * restrict src, mat4 dest);
  */
 
 #ifndef cglm_mat_h
@@ -64,6 +65,10 @@
 #  include "simd/neon/mat4.h"
 #endif
 
+#ifdef CGLM_SIMD_WASM
+#  include "simd/wasm/mat4.h"
+#endif
+
 #ifdef DEBUG
 # include <assert.h>
 #endif
@@ -121,7 +126,12 @@ glm_mat4_ucopy(mat4 mat, mat4 dest) {
 CGLM_INLINE
 void
 glm_mat4_copy(mat4 mat, mat4 dest) {
-#ifdef __AVX__
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glmm_store(dest[0], glmm_load(mat[0]));
+  glmm_store(dest[1], glmm_load(mat[1]));
+  glmm_store(dest[2], glmm_load(mat[2]));
+  glmm_store(dest[3], glmm_load(mat[3]));
+#elif defined(__AVX__)
   glmm_store256(dest[0], glmm_load256(mat[0]));
   glmm_store256(dest[2], glmm_load256(mat[2]));
 #elif defined( __SSE__ ) || defined( __SSE2__ )
@@ -187,7 +197,14 @@ glm_mat4_identity_array(mat4 * __restrict mat, size_t count) {
 CGLM_INLINE
 void
 glm_mat4_zero(mat4 mat) {
-#ifdef __AVX__
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glmm_128 x0;
+  x0 = wasm_f32x4_const_splat(0.f);
+  glmm_store(mat[0], x0);
+  glmm_store(mat[1], x0);
+  glmm_store(mat[2], x0);
+  glmm_store(mat[3], x0);
+#elif defined(__AVX__)
   __m256 y0;
   y0 = _mm256_setzero_ps();
   glmm_store256(mat[0], y0);
@@ -297,7 +314,9 @@ glm_mat4_ins3(mat3 mat, mat4 dest) {
 CGLM_INLINE
 void
 glm_mat4_mul(mat4 m1, mat4 m2, mat4 dest) {
-#ifdef __AVX__
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_mat4_mul_wasm(m1, m2, dest);
+#elif defined(__AVX__)
   glm_mat4_mul_avx(m1, m2, dest);
 #elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_mat4_mul_sse2(m1, m2, dest);
@@ -377,7 +396,9 @@ glm_mat4_mulN(mat4 * __restrict matrices[], uint32_t len, mat4 dest) {
 CGLM_INLINE
 void
 glm_mat4_mulv(mat4 m, vec4 v, vec4 dest) {
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_mat4_mulv_wasm(m, v, dest);
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_mat4_mulv_sse2(m, v, dest);
 #elif defined(CGLM_NEON_FP)
   glm_mat4_mulv_neon(m, v, dest);
@@ -497,7 +518,9 @@ glm_mat4_mulv3(mat4 m, vec3 v, float last, vec3 dest) {
 CGLM_INLINE
 void
 glm_mat4_transpose_to(mat4 m, mat4 dest) {
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_mat4_transp_wasm(m, dest);
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_mat4_transp_sse2(m, dest);
 #elif defined(CGLM_NEON_FP)
   glm_mat4_transp_neon(m, dest);
@@ -521,7 +544,9 @@ glm_mat4_transpose_to(mat4 m, mat4 dest) {
 CGLM_INLINE
 void
 glm_mat4_transpose(mat4 m) {
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_mat4_transp_wasm(m, m);
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_mat4_transp_sse2(m, m);
 #elif defined(CGLM_NEON_FP)
   glm_mat4_transp_neon(m, m);
@@ -560,7 +585,9 @@ glm_mat4_scale_p(mat4 m, float s) {
 CGLM_INLINE
 void
 glm_mat4_scale(mat4 m, float s) {
-#ifdef __AVX__
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_mat4_scale_wasm(m, s);
+#elif defined(__AVX__)
   glm_mat4_scale_avx(m, s);
 #elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_mat4_scale_sse2(m, s);
@@ -581,7 +608,9 @@ glm_mat4_scale(mat4 m, float s) {
 CGLM_INLINE
 float
 glm_mat4_det(mat4 mat) {
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  return glm_mat4_det_wasm(mat);
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   return glm_mat4_det_sse2(mat);
 #elif defined(CGLM_NEON_FP)
   return glm_mat4_det_neon(mat);
@@ -679,7 +708,9 @@ glm_mat4_inv(mat4 mat, mat4 dest) {
 CGLM_INLINE
 void
 glm_mat4_inv_fast(mat4 mat, mat4 dest) {
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_mat4_inv_fast_wasm(mat, dest);
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_mat4_inv_fast_sse2(mat, dest);
 #else
   glm_mat4_inv(mat, dest);
@@ -751,4 +782,24 @@ glm_mat4_rmc(vec4 r, mat4 m, vec4 c) {
   return glm_vec4_dot(r, tmp);
 }
 
+/*!
+ * @brief Create mat4 matrix from pointer
+ *
+ * @param[in]  src  pointer to an array of floats
+ * @param[out] dest matrix
+ */
+CGLM_INLINE
+void
+glm_mat4_make(float * __restrict src, mat4 dest) {
+  dest[0][0] = src[0];   dest[1][0] = src[4];
+  dest[0][1] = src[1];   dest[1][1] = src[5];
+  dest[0][2] = src[2];   dest[1][2] = src[6];
+  dest[0][3] = src[3];   dest[1][3] = src[7];
+
+  dest[2][0] = src[8];   dest[3][0] = src[12];
+  dest[2][1] = src[9];   dest[3][1] = src[13];
+  dest[2][2] = src[10];  dest[3][2] = src[14];
+  dest[2][3] = src[11];  dest[3][3] = src[15];
+}
+
 #endif /* cglm_mat_h */

include/cglm/mat4x2.h

@@ -0,0 +1,184 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+/*
+ Macros:
+   GLM_MAT4X2_ZERO_INIT
+   GLM_MAT4X2_ZERO
+
+ Functions:
+   CGLM_INLINE void glm_mat4x2_copy(mat4x2 mat, mat4x2 dest);
+   CGLM_INLINE void glm_mat4x2_zero(mat4x2 mat);
+   CGLM_INLINE void glm_mat4x2_make(float * __restrict src, mat4x2 dest);
+   CGLM_INLINE void glm_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat4 dest);
+   CGLM_INLINE void glm_mat4x2_mulv(mat4x2 m, vec2 v, vec4 dest);
+   CGLM_INLINE void glm_mat4x2_transpose(mat4x2 m, mat2x4 dest);
+   CGLM_INLINE void glm_mat4x2_scale(mat4x2 m, float s);
+ */
+
+#ifndef cglm_mat4x2_h
+#define cglm_mat4x2_h
+
+#include "common.h"
+
+#define GLM_MAT4X2_ZERO_INIT {{0.0f, 0.0f}, {0.0f, 0.0f}, {0.0f, 0.0f}, {0.0f, 0.0f}}
+
+/* for C only */
+#define GLM_MAT4X2_ZERO GLM_MAT4X2_ZERO_INIT
+
+/*!
+ * @brief copy all members of [mat] to [dest]
+ *
+ * @param[in]  mat  source
+ * @param[out] dest destination
+ */
+CGLM_INLINE
+void
+glm_mat4x2_copy(mat4x2 mat, mat4x2 dest) {
+  dest[0][0] = mat[0][0];
+  dest[0][1] = mat[0][1];
+
+  dest[1][0] = mat[1][0];
+  dest[1][1] = mat[1][1];
+
+  dest[2][0] = mat[2][0];
+  dest[2][1] = mat[2][1];
+
+  dest[3][0] = mat[3][0];
+  dest[3][1] = mat[3][1];
+}
+
+/*!
+ * @brief make given matrix zero.
+ *
+ * @param[in, out]  mat  matrix
+ */
+CGLM_INLINE
+void
+glm_mat4x2_zero(mat4x2 mat) {
+  CGLM_ALIGN_MAT mat4x2 t = GLM_MAT4X2_ZERO_INIT;
+  glm_mat4x2_copy(t, mat);
+}
+
+/*!
+ * @brief Create mat4x2 matrix from pointer
+ *
+ * @param[in]  src  pointer to an array of floats
+ * @param[out] dest matrix
+ */
+CGLM_INLINE
+void
+glm_mat4x2_make(float * __restrict src, mat4x2 dest) {
+  dest[0][0] = src[0];
+  dest[0][1] = src[1];
+
+  dest[1][0] = src[2];
+  dest[1][1] = src[3];
+
+  dest[2][0] = src[4];
+  dest[2][1] = src[5];
+
+  dest[3][0] = src[6];
+  dest[3][1] = src[7];
+}
+
+/*!
+ * @brief multiply m1 and m2 to dest
+ *
+ * @code
+ * glm_mat4x2_mul(mat4x2, mat2x4, mat4);
+ * @endcode
+ *
+ * @param[in]  m1   left matrix (mat4x2)
+ * @param[in]  m2   right matrix (mat2x4)
+ * @param[out] dest destination matrix (mat4)
+ */
+CGLM_INLINE
+void
+glm_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat4 dest) {
+  float a00 = m1[0][0], a01 = m1[0][1],
+        a10 = m1[1][0], a11 = m1[1][1],
+        a20 = m1[2][0], a21 = m1[2][1],
+        a30 = m1[3][0], a31 = m1[3][1],
+
+        b00 = m2[0][0], b01 = m2[0][1], b02 = m2[0][2], b03 = m2[0][3],
+        b10 = m2[1][0], b11 = m2[1][1], b12 = m2[1][2], b13 = m2[1][3];
+
+  dest[0][0] = a00 * b00 + a01 * b10;
+  dest[0][1] = a00 * b01 + a01 * b11;
+  dest[0][2] = a00 * b02 + a01 * b12;
+  dest[0][3] = a00 * b03 + a01 * b13;
+
+  dest[1][0] = a10 * b00 + a11 * b10;
+  dest[1][1] = a10 * b01 + a11 * b11;
+  dest[1][2] = a10 * b02 + a11 * b12;
+  dest[1][3] = a10 * b03 + a11 * b13;
+
+  dest[2][0] = a20 * b00 + a21 * b10;
+  dest[2][1] = a20 * b01 + a21 * b11;
+  dest[2][2] = a20 * b02 + a21 * b12;
+  dest[2][3] = a20 * b03 + a21 * b13;
+
+  dest[3][0] = a30 * b00 + a31 * b10;
+  dest[3][1] = a30 * b01 + a31 * b11;
+  dest[3][2] = a30 * b02 + a31 * b12;
+  dest[3][3] = a30 * b03 + a31 * b13;
+}
+
+/*!
+ * @brief multiply matrix with column vector and store in dest vector
+ *
+ * @param[in]  m    matrix (left)
+ * @param[in]  v    vector (right, column vector)
+ * @param[out] dest result vector
+ */
+CGLM_INLINE
+void
+glm_mat4x2_mulv(mat4x2 m, vec2 v, vec4 dest) {
+  float v0 = v[0], v1 = v[1];
+
+  dest[0] = m[0][0] * v0 + m[0][1] * v1;
+  dest[1] = m[1][0] * v0 + m[1][1] * v1;
+  dest[2] = m[2][0] * v0 + m[2][1] * v1;
+  dest[3] = m[3][0] * v0 + m[3][1] * v1;
+}
+
+/*!
+ * @brief transpose matrix and store in dest
+ *
+ * @param[in]  m     matrix
+ * @param[out] dest  result
+ */
+CGLM_INLINE
+void
+glm_mat4x2_transpose(mat4x2 m, mat2x4 dest) {
+  dest[0][0] = m[0][0];
+  dest[0][1] = m[1][0];
+  dest[0][2] = m[2][0];
+  dest[0][3] = m[3][0];
+  dest[1][0] = m[0][1];
+  dest[1][1] = m[1][1];
+  dest[1][2] = m[2][1];
+  dest[1][3] = m[3][1];
+}
+
+/*!
+ * @brief scale (multiply with scalar) matrix
+ *
+ * multiply matrix with scalar
+ *
+ * @param[in, out] m matrix
+ * @param[in]    s scalar
+ */
+CGLM_INLINE
+void
+glm_mat4x2_scale(mat4x2 m, float s) {
+  m[0][0] *= s;  m[0][1] *= s;  m[1][0] *= s;  m[1][1] *= s;
+  m[2][0] *= s;  m[2][1] *= s;  m[3][0] *= s;  m[3][1] *= s;
+}
+
+#endif

include/cglm/mat4x3.h

@@ -0,0 +1,201 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+/*
+ Macros:
+   GLM_MAT4X3_ZERO_INIT
+   GLM_MAT4X3_ZERO
+
+ Functions:
+   CGLM_INLINE void glm_mat4x3_copy(mat4x3 mat, mat4x3 dest);
+   CGLM_INLINE void glm_mat4x3_zero(mat4x3 mat);
+   CGLM_INLINE void glm_mat4x3_make(float * __restrict src, mat4x3 dest);
+   CGLM_INLINE void glm_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat4 dest);
+   CGLM_INLINE void glm_mat4x3_mulv(mat4x3 m, vec3 v, vec4 dest);
+   CGLM_INLINE void glm_mat4x3_transpose(mat4x3 m, mat3x4 dest);
+   CGLM_INLINE void glm_mat4x3_scale(mat4x3 m, float s);
+ */
+
+#ifndef cglm_mat4x3_h
+#define cglm_mat4x3_h
+
+#include "common.h"
+
+#define GLM_MAT4X3_ZERO_INIT {{0.0f, 0.0f, 0.0f}, {0.0f, 0.0f, 0.0f}, \
+                              {0.0f, 0.0f, 0.0f}, {0.0f, 0.0f, 0.0f}}
+
+/* for C only */
+#define GLM_MAT4X3_ZERO GLM_MAT4X3_ZERO_INIT
+
+/*!
+ * @brief copy all members of [mat] to [dest]
+ *
+ * @param[in]  mat  source
+ * @param[out] dest destination
+ */
+CGLM_INLINE
+void
+glm_mat4x3_copy(mat4x3 mat, mat4x3 dest) {
+  dest[0][0] = mat[0][0];
+  dest[0][1] = mat[0][1];
+  dest[0][2] = mat[0][2];
+
+  dest[1][0] = mat[1][0];
+  dest[1][1] = mat[1][1];
+  dest[1][2] = mat[1][2];
+
+  dest[2][0] = mat[2][0];
+  dest[2][1] = mat[2][1];
+  dest[2][2] = mat[2][2];
+
+  dest[3][0] = mat[3][0];
+  dest[3][1] = mat[3][1];
+  dest[3][2] = mat[3][2];
+}
+
+/*!
+ * @brief make given matrix zero.
+ *
+ * @param[in, out]  mat  matrix
+ */
+CGLM_INLINE
+void
+glm_mat4x3_zero(mat4x3 mat) {
+  CGLM_ALIGN_MAT mat4x3 t = GLM_MAT4X3_ZERO_INIT;
+  glm_mat4x3_copy(t, mat);
+}
+
+/*!
+ * @brief Create mat4x3 matrix from pointer
+ *
+ * @param[in]  src  pointer to an array of floats
+ * @param[out] dest matrix
+ */
+CGLM_INLINE
+void
+glm_mat4x3_make(float * __restrict src, mat4x3 dest) {
+  dest[0][0] = src[0];
+  dest[0][1] = src[1];
+  dest[0][2] = src[2];
+
+  dest[1][0] = src[3];
+  dest[1][1] = src[4];
+  dest[1][2] = src[5];
+
+  dest[2][0] = src[6];
+  dest[2][1] = src[7];
+  dest[2][2] = src[8];
+
+  dest[3][0] = src[9];
+  dest[3][1] = src[10];
+  dest[3][2] = src[11];
+}
+
+/*!
+ * @brief multiply m1 and m2 to dest
+ *
+ * @code
+ * glm_mat4x3_mul(mat4x3, mat3x4, mat4);
+ * @endcode
+ *
+ * @param[in]  m1   left matrix (mat4x3)
+ * @param[in]  m2   right matrix (mat3x4)
+ * @param[out] dest destination matrix (mat4)
+ */
+CGLM_INLINE
+void
+glm_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat4 dest) {
+  float a00 = m1[0][0], a01 = m1[0][1], a02 = m1[0][2],
+        a10 = m1[1][0], a11 = m1[1][1], a12 = m1[1][2],
+        a20 = m1[2][0], a21 = m1[2][1], a22 = m1[2][2],
+        a30 = m1[3][0], a31 = m1[3][1], a32 = m1[3][2],
+
+        b00 = m2[0][0], b01 = m2[0][1], b02 = m2[0][2], b03 = m2[0][3],
+        b10 = m2[1][0], b11 = m2[1][1], b12 = m2[1][2], b13 = m2[1][3],
+        b20 = m2[2][0], b21 = m2[2][1], b22 = m2[2][2], b23 = m2[2][3];
+
+  dest[0][0] = a00 * b00 + a01 * b10 + a02 * b20;
+  dest[0][1] = a00 * b01 + a01 * b11 + a02 * b21;
+  dest[0][2] = a00 * b02 + a01 * b12 + a02 * b22;
+  dest[0][3] = a00 * b03 + a01 * b13 + a02 * b23;
+
+  dest[1][0] = a10 * b00 + a11 * b10 + a12 * b20;
+  dest[1][1] = a10 * b01 + a11 * b11 + a12 * b21;
+  dest[1][2] = a10 * b02 + a11 * b12 + a12 * b22;
+  dest[1][3] = a10 * b03 + a11 * b13 + a12 * b23;
+
+  dest[2][0] = a20 * b00 + a21 * b10 + a22 * b20;
+  dest[2][1] = a20 * b01 + a21 * b11 + a22 * b21;
+  dest[2][2] = a20 * b02 + a21 * b12 + a22 * b22;
+  dest[2][3] = a20 * b03 + a21 * b13 + a22 * b23;
+
+  dest[3][0] = a30 * b00 + a31 * b10 + a32 * b20;
+  dest[3][1] = a30 * b01 + a31 * b11 + a32 * b21;
+  dest[3][2] = a30 * b02 + a31 * b12 + a32 * b22;
+  dest[3][3] = a30 * b03 + a31 * b13 + a32 * b23;
+}
+
+/*!
+ * @brief multiply matrix with column vector and store in dest vector
+ *
+ * @param[in]  m    matrix (left)
+ * @param[in]  v    vector (right, column vector)
+ * @param[out] dest result vector
+ */
+CGLM_INLINE
+void
+glm_mat4x3_mulv(mat4x3 m, vec3 v, vec4 dest) {
+  float v0 = v[0], v1 = v[1], v2 = v[2];
+
+  dest[0] = m[0][0] * v0 + m[0][1] * v1 + m[0][2] * v2;
+  dest[1] = m[1][0] * v0 + m[1][1] * v1 + m[1][2] * v2;
+  dest[2] = m[2][0] * v0 + m[2][1] * v1 + m[2][2] * v2;
+  dest[3] = m[3][0] * v0 + m[3][1] * v1 + m[3][2] * v2;
+}
+
+/*!
+ * @brief transpose matrix and store in dest
+ *
+ * @param[in]  m     matrix
+ * @param[out] dest  result
+ */
+CGLM_INLINE
+void
+glm_mat4x3_transpose(mat4x3 m, mat3x4 dest) {
+  dest[0][0] = m[0][0];
+  dest[0][1] = m[1][0];
+  dest[0][2] = m[2][0];
+  dest[0][3] = m[3][0];
+
+  dest[1][0] = m[0][1];
+  dest[1][1] = m[1][1];
+  dest[1][2] = m[2][1];
+  dest[1][3] = m[3][1];
+
+  dest[2][0] = m[0][2];
+  dest[2][1] = m[1][2];
+  dest[2][2] = m[2][2];
+  dest[2][3] = m[3][2];
+}
+
+/*!
+ * @brief scale (multiply with scalar) matrix
+ *
+ * multiply matrix with scalar
+ *
+ * @param[in, out] m matrix
+ * @param[in]    s scalar
+ */
+CGLM_INLINE
+void
+glm_mat4x3_scale(mat4x3 m, float s) {
+  m[0][0] *= s;  m[0][1] *= s;  m[0][2] *= s;  m[1][0] *= s;
+  m[1][1] *= s;  m[1][2] *= s;  m[2][0] *= s;  m[2][1] *= s;
+  m[2][2] *= s;  m[3][0] *= s;  m[3][1] *= s;  m[3][2] *= s;
+}
+
+#endif

include/cglm/quat.h

@@ -49,6 +49,7 @@
                                   versor dest);
    CGLM_INLINE void glm_quat_rotatev(versor q, vec3 v, vec3 dest);
    CGLM_INLINE void glm_quat_rotate(mat4 m, versor q, mat4 dest);
+   CGLM_INLINE void glm_quat_make(float * restrict src, versor dest);
  */
 
 #ifndef cglm_quat_h
@@ -70,6 +71,10 @@
 #  include "simd/neon/quat.h"
 #endif
 
+#ifdef CGLM_SIMD_WASM
+#  include "simd/wasm/quat.h"
+#endif
+
 CGLM_INLINE void glm_quat_normalize(versor q);
 
 /*
@@ -238,7 +243,22 @@ glm_quat_norm(versor q) {
 CGLM_INLINE
 void
 glm_quat_normalize_to(versor q, versor dest) {
-#if defined( __SSE2__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glmm_128 xdot, x0;
+  float  dot;
+
+  x0   = glmm_load(q);
+  xdot = glmm_vdot(x0, x0);
+  /* dot  = _mm_cvtss_f32(xdot); */
+  dot  = wasm_f32x4_extract_lane(xdot, 0);
+
+  if (dot <= 0.0f) {
+    glm_quat_identity(dest);
+    return;
+  }
+
+  glmm_store(dest, wasm_f32x4_div(x0, wasm_f32x4_sqrt(xdot)));
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   __m128 xdot, x0;
   float  dot;
 
@@ -438,7 +458,9 @@ glm_quat_mul(versor p, versor q, versor dest) {
     + (a1 d2 + b1 c2 − c1 b2 + d1 a2)k
        a1 a2 − b1 b2 − c1 c2 − d1 d2
    */
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__wasm__) && defined(__wasm_simd128__)
+  glm_quat_mul_wasm(p, q, dest);
+#elif defined( __SSE__ ) || defined( __SSE2__ )
   glm_quat_mul_sse2(p, q, dest);
 #elif defined(CGLM_NEON_FP)
   glm_quat_mul_neon(p, q, dest);
@@ -864,4 +886,17 @@ glm_quat_rotate_atm(mat4 m, versor q, vec3 pivot) {
   glm_translate(m, pivotInv);
 }
 
+/*!
+ * @brief Create quaternion from pointer
+ *
+ * @param[in]  src  pointer to an array of floats
+ * @param[out] dest quaternion
+ */
+CGLM_INLINE
+void
+glm_quat_make(float * __restrict src, versor dest) {
+  dest[0] = src[0]; dest[1] = src[1];
+  dest[2] = src[2]; dest[3] = src[3];
+}
+
 #endif /* cglm_quat_h */

include/cglm/simd/arm.h

@@ -12,6 +12,8 @@
 
 #if defined(_M_ARM64) || defined(_M_HYBRID_X86_ARM64) || defined(_M_ARM64EC) || defined(__aarch64__)
 # define CGLM_ARM64 1
+#else
+# define CGLM_ARM64 0
 #endif
 
 #define glmm_load(p)      vld1q_f32(p)
@@ -38,11 +40,25 @@
 #define glmm_combine_lh(x, y) vcombine_f32(vget_low_f32(x),  vget_high_f32(y))
 #define glmm_combine_hh(x, y) vcombine_f32(vget_high_f32(x), vget_high_f32(y))
 
-static inline
+#if defined(_WIN32) && defined(_MSC_VER)
+/* #  define glmm_float32x4_init(x, y, z, w) { .n128_f32 = { x, y, z, w } } */
+CGLM_INLINE
 float32x4_t
-glmm_abs(float32x4_t v) {
+glmm_float32x4_init(float x, float y, float z, float w) {
-  return vabsq_f32(v);
+  CGLM_ALIGN(16) float v[4] = {x, y, z, w};
+  return vld1q_f32(v);
 }
+#else
+#  define glmm_float32x4_init(x, y, z, w) { x, y, z, w }
+#endif
+
+#define glmm_float32x4_SIGNMASK_PNPN glmm_float32x4_init( 0.f, -0.f,  0.f, -0.f)
+#define glmm_float32x4_SIGNMASK_NPNP glmm_float32x4_init(-0.f,  0.f, -0.f,  0.f)
+#define glmm_float32x4_SIGNMASK_NPPN glmm_float32x4_init(-0.f,  0.f,  0.f, -0.f)
+
+static inline float32x4_t glmm_abs(float32x4_t v)                { return vabsq_f32(v);    }
+static inline float32x4_t glmm_min(float32x4_t a, float32x4_t b) { return vminq_f32(a, b); }
+static inline float32x4_t glmm_max(float32x4_t a, float32x4_t b) { return vmaxq_f32(a, b); }
 
 static inline
 float32x4_t
@@ -156,11 +172,7 @@ glmm_fnmadd(float32x4_t a, float32x4_t b, float32x4_t c) {
 static inline
 float32x4_t
 glmm_fmsub(float32x4_t a, float32x4_t b, float32x4_t c) {
-#if CGLM_ARM64
+  return glmm_fmadd(a, b, vnegq_f32(c));
-  return vfmsq_f32(c, a, b);
-#else
-  return vmlsq_f32(c, a, b);
-#endif
 }
 
 static inline

include/cglm/simd/intrin.h

@@ -17,7 +17,7 @@
 #    ifndef __SSE__
 #      define __SSE__
 #    endif
-#  endif
+#endif
 /* do not use alignment for older visual studio versions */
 #  if _MSC_VER < 1913     /* Visual Studio 2017 version 15.6 */
 #    define CGLM_ALL_UNALIGNED
@@ -63,23 +63,56 @@
 #endif
 
 /* ARM Neon */
-#if defined(__ARM_NEON)
+#if defined(_WIN32) && defined(_MSC_VER)
-#  include <arm_neon.h>
+/* TODO: non-ARM stuff already inported, will this be better option */
-#  if defined(__ARM_NEON_FP)
+/* #  include <intrin.h> */
-#    define CGLM_NEON_FP 1
+
+#  if defined(_M_ARM64) || defined(_M_HYBRID_X86_ARM64) || defined(_M_ARM64EC)
+#    include <arm64intr.h>
+#    include <arm64_neon.h>
+#    ifndef CGLM_NEON_FP
+#      define CGLM_NEON_FP  1
+#    endif
+#    ifndef CGLM_SIMD_ARM
+#      define CGLM_SIMD_ARM
+#    endif
+#  elif defined(_M_ARM)
+#    include <armintr.h>
+#    include <arm_neon.h>
+#    ifndef CGLM_NEON_FP
+#      define CGLM_NEON_FP 1
+#    endif
+#    ifndef CGLM_SIMD_ARM
+#      define CGLM_SIMD_ARM
+#    endif
+#  endif
+
+#else /* non-windows */
+#  if defined(__ARM_NEON) || defined(__ARM_NEON__)
+#    include <arm_neon.h>
+#    if defined(__ARM_NEON_FP)
+#      define CGLM_NEON_FP 1
+#    endif
 #    ifndef CGLM_SIMD_ARM
 #      define CGLM_SIMD_ARM
 #    endif
 #  endif
 #endif
 
-#if defined(CGLM_SIMD_x86) || defined(CGLM_NEON_FP)
+/* WebAssembly */
+#if defined(__wasm__) && defined(__wasm_simd128__)
+#  ifndef CGLM_SIMD_WASM
+#    define CGLM_SIMD_WASM
+#  endif
+#endif
+
+#if defined(CGLM_SIMD_x86) || defined(CGLM_SIMD_ARM) || defined(CGLM_SIMD_WASM)
 #  ifndef CGLM_SIMD
 #    define CGLM_SIMD
 #  endif
 #endif
 
-#if defined(CGLM_SIMD_x86)
+#if defined(CGLM_SIMD_x86) && !defined(CGLM_SIMD_WASM)
 #  include "x86.h"
 #endif
 
@@ -87,4 +120,8 @@
 #  include "arm.h"
 #endif
 
+#if defined(CGLM_SIMD_WASM)
+#  include "wasm.h"
+#endif
+
 #endif /* cglm_intrin_h */

include/cglm/simd/neon/affine.h

@@ -7,7 +7,7 @@
 
 #ifndef cglm_affine_neon_h
 #define cglm_affine_neon_h
-#if defined(__ARM_NEON_FP)
+#if defined(CGLM_NEON_FP)
 
 #include "../../common.h"
 #include "../intrin.h"

include/cglm/simd/neon/mat2.h

@@ -7,7 +7,7 @@
 
 #ifndef cglm_mat2_neon_h
 #define cglm_mat2_neon_h
-#if defined(__ARM_NEON_FP)
+#if defined(CGLM_NEON_FP)
 
 #include "../../common.h"
 #include "../intrin.h"

include/cglm/simd/neon/mat4.h

@@ -7,7 +7,7 @@
 
 #ifndef cglm_mat4_neon_h
 #define cglm_mat4_neon_h
-#if defined(__ARM_NEON_FP)
+#if defined(CGLM_NEON_FP)
 
 #include "../../common.h"
 #include "../intrin.h"
@@ -108,7 +108,7 @@ glm_mat4_det_neon(mat4 mat) {
   float32x4_t   r0, r1, r2, r3, x0, x1, x2;
   float32x2_t   ij, op, mn, kl, nn, mm, jj, ii, gh, ef, t12, t34;
   float32x4x2_t a1;
-  float32x4_t   x3 = { 0.f, -0.f, 0.f, -0.f };
+  float32x4_t   x3 = glmm_float32x4_SIGNMASK_PNPN;
 
   /* 127 <- 0, [square] det(A) = det(At) */
   r0 = glmm_load(mat[0]);              /* d c b a */
@@ -181,7 +181,7 @@ glm_mat4_inv_neon(mat4 mat, mat4 dest) {
                 x0, x1, x2, x3, x4, x5, x6, x7, x8;
   float32x4x2_t a1;
   float32x2_t   lp, ko, hg, jn, im, fe, ae, bf, cg, dh;
-  float32x4_t   x9 = { -0.f, 0.f, -0.f, 0.f };
+  float32x4_t   x9 = glmm_float32x4_SIGNMASK_NPNP;
 
   x8 = vrev64q_f32(x9);
 

include/cglm/simd/neon/quat.h

@@ -7,7 +7,7 @@
 
 #ifndef cglm_quat_neon_h
 #define cglm_quat_neon_h
-#if defined(__ARM_NEON_FP)
+#if defined(CGLM_NEON_FP)
 
 #include "../../common.h"
 #include "../intrin.h"
@@ -23,7 +23,8 @@ glm_quat_mul_neon(versor p, versor q, versor dest) {
    */
 
   glmm_128 xp, xq, xqr, r, x, y, z, s2, s3;
-  glmm_128 s1 = {-0.f, 0.f, 0.f, -0.f};
+  glmm_128 s1 = glmm_float32x4_SIGNMASK_NPPN;
+
   float32x2_t   qh, ql;
   
   xp  = glmm_load(p); /* 3 2 1 0 */

include/cglm/simd/sse2/affine.h

@@ -98,7 +98,7 @@ glm_inv_tr_sse2(mat4 mat) {
   x2 = glmm_shuff1(r3, 0, 0, 0, 0);
   x3 = glmm_shuff1(r3, 1, 1, 1, 1);
   x4 = glmm_shuff1(r3, 2, 2, 2, 2);
-  x5 = _mm_set1_ps(-0.f);
+  x5 = glmm_float32x4_SIGNMASK_NEG;
 
   x0 = glmm_fmadd(r0, x2, glmm_fmadd(r1, x3, _mm_mul_ps(r2, x4)));
   x0 = _mm_xor_ps(x0, x5);

include/cglm/simd/sse2/mat4.h

@@ -153,7 +153,7 @@ glm_mat4_det_sse2(mat4 mat) {
                   _mm_shuffle_ps(x0, x1, _MM_SHUFFLE(2, 2, 3, 1)),
                   x2);
   
-  x2 = _mm_xor_ps(x2, _mm_set_ps(-0.f, 0.f, -0.f, 0.f));
+  x2 = _mm_xor_ps(x2, glmm_float32x4_SIGNMASK_NPNP);
   
   return glmm_hadd(_mm_mul_ps(x2, r0));
 }
@@ -166,7 +166,8 @@ glm_mat4_inv_fast_sse2(mat4 mat, mat4 dest) {
          t0, t1, t2, t3, t4, t5,
          x0, x1, x2, x3, x4, x5, x6, x7, x8, x9;
 
-  x8 = _mm_set_ps(-0.f, 0.f, -0.f, 0.f);
+  /* x8 = _mm_set_ps(-0.f, 0.f, -0.f, 0.f); */
+  x8 = glmm_float32x4_SIGNMASK_NPNP;
   x9 = glmm_shuff1(x8, 2, 1, 2, 1);
 
   /* 127 <- 0 */
@@ -302,7 +303,8 @@ glm_mat4_inv_sse2(mat4 mat, mat4 dest) {
          t0, t1, t2, t3, t4, t5,
          x0, x1, x2, x3, x4, x5, x6, x7, x8, x9;
 
-  x8 = _mm_set_ps(-0.f, 0.f, -0.f, 0.f);
+  /* x8 = _mm_set_ps(-0.f, 0.f, -0.f, 0.f); */
+  x8 = glmm_float32x4_SIGNMASK_NPNP;
   x9 = glmm_shuff1(x8, 2, 1, 2, 1);
 
   /* 127 <- 0 */

include/cglm/simd/sse2/quat.h

@@ -26,9 +26,9 @@ glm_quat_mul_sse2(versor p, versor q, versor dest) {
 
   xp = glmm_load(p); /* 3 2 1 0 */
   xq = glmm_load(q);
-  x1 = _mm_set_ps(-0.f, 0.f, -0.f, 0.f); /* TODO: _mm_set1_ss() + shuff ? */
+  x1 = glmm_float32x4_SIGNMASK_NPNP; /* TODO: _mm_set1_ss() + shuff ? */
   r  = _mm_mul_ps(glmm_splat_w(xp), xq);
-  
+
   x2 = _mm_unpackhi_ps(x1, x1);
   x3 = glmm_shuff1(x1, 3, 2, 0, 1);
   x  = glmm_splat_x(xp);

include/cglm/simd/wasm.h

@@ -0,0 +1,196 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglm_simd_wasm_h
+#define cglm_simd_wasm_h
+#include "intrin.h"
+#ifdef CGLM_SIMD_WASM
+#include <wasm_simd128.h>
+
+#define glmm_load(p)      wasm_v128_load(p)
+#define glmm_store(p, a)  wasm_v128_store(p, (a))
+
+#define glmm_set1(x) wasm_f32x4_splat(x)
+#define glmm_128     v128_t
+
+#define glmm_shuff1(xmm, z, y, x, w) wasm_i32x4_shuffle(xmm, xmm, w, x, y, z)
+
+#define glmm_splat(x, lane) glmm_shuff1(x, lane, lane, lane, lane)
+
+#define glmm_splat_x(x) glmm_splat(x, 0)
+#define glmm_splat_y(x) glmm_splat(x, 1)
+#define glmm_splat_z(x) glmm_splat(x, 2)
+#define glmm_splat_w(x) glmm_splat(x, 3)
+
+#define GLMM_NEGZEROf 0x80000000 /*  0x80000000 ---> -0.0f  */
+
+/* _mm_set_ps(X, Y, Z, W); */
+#define GLMM__SIGNMASKf(X, Y, Z, W) wasm_i32x4_const(X, Y, Z, W)
+
+#define glmm_float32x4_SIGNMASK_PNPN GLMM__SIGNMASKf(0, GLMM_NEGZEROf, 0, GLMM_NEGZEROf)
+#define glmm_float32x4_SIGNMASK_NPNP GLMM__SIGNMASKf(GLMM_NEGZEROf, 0, GLMM_NEGZEROf, 0)
+#define glmm_float32x4_SIGNMASK_NPPN GLMM__SIGNMASKf(GLMM_NEGZEROf, 0, 0, GLMM_NEGZEROf)
+#define glmm_float32x4_SIGNMASK_NEG  wasm_i32x4_const_splat(GLMM_NEGZEROf)
+
+static inline glmm_128 glmm_abs(glmm_128 x)             { return wasm_f32x4_abs(x);     }
+static inline glmm_128 glmm_min(glmm_128 a, glmm_128 b) { return wasm_f32x4_pmin(b, a); }
+static inline glmm_128 glmm_max(glmm_128 a, glmm_128 b) { return wasm_f32x4_pmax(b, a); }
+
+static inline
+glmm_128
+glmm_vhadd(glmm_128 v) {
+  glmm_128 x0;
+  x0 = wasm_f32x4_add(v,  glmm_shuff1(v, 0, 1, 2, 3));
+  x0 = wasm_f32x4_add(x0, glmm_shuff1(x0, 1, 0, 0, 1));
+  return x0;
+}
+
+static inline
+glmm_128
+glmm_vhadds(glmm_128 v) {
+  glmm_128 shuf, sums;
+  shuf = glmm_shuff1(v, 2, 3, 0, 1);
+  sums = wasm_f32x4_add(v, shuf);
+  /* shuf = _mm_movehl_ps(shuf, sums); */
+  shuf = wasm_i32x4_shuffle(shuf, sums, 6, 7, 2, 3);
+  sums = wasm_i32x4_shuffle(sums, wasm_f32x4_add(sums, shuf), 4, 1, 2, 3);
+  return sums;
+}
+
+static inline
+float
+glmm_hadd(glmm_128 v) {
+  return wasm_f32x4_extract_lane(glmm_vhadds(v), 0);
+}
+
+static inline
+glmm_128
+glmm_vhmin(glmm_128 v) {
+  glmm_128 x0, x1, x2;
+  x0 = glmm_shuff1(v, 2, 3, 2, 3);     /* [2, 3, 2, 3] */
+  x1 = wasm_f32x4_pmin(x0, v);         /* [0|2, 1|3, 2|2, 3|3] */
+  x2 = glmm_splat(x1, 1);              /* [1|3, 1|3, 1|3, 1|3] */
+  return wasm_f32x4_pmin(x1, x2);
+}
+
+static inline
+float
+glmm_hmin(glmm_128 v) {
+  return wasm_f32x4_extract_lane(glmm_vhmin(v), 0);
+}
+
+static inline
+glmm_128
+glmm_vhmax(glmm_128 v) {
+  glmm_128 x0, x1, x2;
+  x0 = glmm_shuff1(v, 2, 3, 2, 3);     /* [2, 3, 2, 3] */
+  x1 = wasm_f32x4_pmax(x0, v);         /* [0|2, 1|3, 2|2, 3|3] */
+  x2 = glmm_splat(x1, 1);              /* [1|3, 1|3, 1|3, 1|3] */
+  /* _mm_max_ss */
+  return wasm_i32x4_shuffle(x1, wasm_f32x4_pmax(x1, x2), 4, 1, 2, 3);
+}
+
+static inline
+float
+glmm_hmax(glmm_128 v) {
+  return wasm_f32x4_extract_lane(glmm_vhmax(v), 0);
+}
+
+static inline
+glmm_128
+glmm_vdots(glmm_128 a, glmm_128 b) {
+  return glmm_vhadds(wasm_f32x4_mul(a, b));
+}
+
+static inline
+glmm_128
+glmm_vdot(glmm_128 a, glmm_128 b) {
+  glmm_128 x0;
+  x0 = wasm_f32x4_mul(a, b);
+  x0 = wasm_f32x4_add(x0, glmm_shuff1(x0, 1, 0, 3, 2));
+  return wasm_f32x4_add(x0, glmm_shuff1(x0, 0, 1, 0, 1));
+}
+
+static inline
+float
+glmm_dot(glmm_128 a, glmm_128 b) {
+  return wasm_f32x4_extract_lane(glmm_vdots(a, b), 0);
+}
+
+static inline
+float
+glmm_norm(glmm_128 a) {
+  glmm_128 x0;
+  x0 = glmm_vhadds(wasm_f32x4_mul(a, a));
+  return wasm_f32x4_extract_lane(
+          wasm_i32x4_shuffle(x0, wasm_f32x4_sqrt(x0),4, 1, 2, 3), 0);
+}
+
+static inline
+float
+glmm_norm2(glmm_128 a) {
+  return wasm_f32x4_extract_lane(glmm_vhadds(wasm_f32x4_mul(a, a)), 0);
+}
+
+static inline
+float
+glmm_norm_one(glmm_128 a) {
+  return wasm_f32x4_extract_lane(glmm_vhadds(glmm_abs(a)), 0);
+}
+
+static inline
+float
+glmm_norm_inf(glmm_128 a) {
+  return wasm_f32x4_extract_lane(glmm_vhmax(glmm_abs(a)), 0);
+}
+
+static inline
+glmm_128
+glmm_load3(float v[3]) {
+  glmm_128 xy = wasm_v128_load64_zero(v);
+  return wasm_f32x4_replace_lane(xy, 2, v[2]);
+}
+
+static inline
+void
+glmm_store3(float v[3], glmm_128 vx) {
+  wasm_v128_store64_lane(v, vx, 0);
+  wasm_v128_store32_lane(&v[2], vx, 2);
+}
+
+static inline
+glmm_128
+glmm_div(glmm_128 a, glmm_128 b) {
+  return wasm_f32x4_div(a, b);
+}
+
+static inline
+glmm_128
+glmm_fmadd(glmm_128 a, glmm_128 b, glmm_128 c) {
+  return wasm_f32x4_add(c, wasm_f32x4_mul(a, b));
+}
+
+static inline
+glmm_128
+glmm_fnmadd(glmm_128 a, glmm_128 b, glmm_128 c) {
+  return wasm_f32x4_sub(c, wasm_f32x4_mul(a, b));
+}
+
+static inline
+glmm_128
+glmm_fmsub(glmm_128 a, glmm_128 b, glmm_128 c) {
+  return wasm_f32x4_sub(wasm_f32x4_mul(a, b), c);
+}
+
+static inline
+glmm_128
+glmm_fnmsub(glmm_128 a, glmm_128 b, glmm_128 c) {
+  return wasm_f32x4_neg(wasm_f32x4_add(wasm_f32x4_mul(a, b), c));
+}
+
+#endif
+#endif /* cglm_simd_wasm_h */

include/cglm/simd/wasm/affine.h

@@ -0,0 +1,127 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglm_affine_mat_wasm_h
+#define cglm_affine_mat_wasm_h
+#if defined(__wasm__) && defined(__wasm_simd128__)
+
+#include "../../common.h"
+#include "../intrin.h"
+
+CGLM_INLINE
+void
+glm_mul_wasm(mat4 m1, mat4 m2, mat4 dest) {
+  /* D = R * L (Column-Major) */
+  glmm_128 l, r0, r1, r2, r3, v0, v1, v2, v3;
+
+  l  = glmm_load(m1[0]);
+  r0 = glmm_load(m2[0]);
+  r1 = glmm_load(m2[1]);
+  r2 = glmm_load(m2[2]);
+  r3 = glmm_load(m2[3]);
+
+  v0 = wasm_f32x4_mul(glmm_splat_x(r0), l);
+  v1 = wasm_f32x4_mul(glmm_splat_x(r1), l);
+  v2 = wasm_f32x4_mul(glmm_splat_x(r2), l);
+  v3 = wasm_f32x4_mul(glmm_splat_x(r3), l);
+
+  l  = glmm_load(m1[1]);
+  v0 = glmm_fmadd(glmm_splat_y(r0), l, v0);
+  v1 = glmm_fmadd(glmm_splat_y(r1), l, v1);
+  v2 = glmm_fmadd(glmm_splat_y(r2), l, v2);
+  v3 = glmm_fmadd(glmm_splat_y(r3), l, v3);
+
+  l  = glmm_load(m1[2]);
+  v0 = glmm_fmadd(glmm_splat_z(r0), l, v0);
+  v1 = glmm_fmadd(glmm_splat_z(r1), l, v1);
+  v2 = glmm_fmadd(glmm_splat_z(r2), l, v2);
+  v3 = glmm_fmadd(glmm_splat_z(r3), l, v3);
+
+  l  = glmm_load(m1[3]);
+  v3 = glmm_fmadd(glmm_splat_w(r3), l, v3);
+
+  glmm_store(dest[0], v0);
+  glmm_store(dest[1], v1);
+  glmm_store(dest[2], v2);
+  glmm_store(dest[3], v3);
+}
+
+CGLM_INLINE
+void
+glm_mul_rot_wasm(mat4 m1, mat4 m2, mat4 dest) {
+  /* D = R * L (Column-Major) */
+
+  glmm_128 l, r0, r1, r2, v0, v1, v2;
+
+  l  = glmm_load(m1[0]);
+  r0 = glmm_load(m2[0]);
+  r1 = glmm_load(m2[1]);
+  r2 = glmm_load(m2[2]);
+
+  v0 = wasm_f32x4_mul(glmm_splat_x(r0), l);
+  v1 = wasm_f32x4_mul(glmm_splat_x(r1), l);
+  v2 = wasm_f32x4_mul(glmm_splat_x(r2), l);
+
+  l  = glmm_load(m1[1]);
+  v0 = glmm_fmadd(glmm_splat_y(r0), l, v0);
+  v1 = glmm_fmadd(glmm_splat_y(r1), l, v1);
+  v2 = glmm_fmadd(glmm_splat_y(r2), l, v2);
+
+  l  = glmm_load(m1[2]);
+  v0 = glmm_fmadd(glmm_splat_z(r0), l, v0);
+  v1 = glmm_fmadd(glmm_splat_z(r1), l, v1);
+  v2 = glmm_fmadd(glmm_splat_z(r2), l, v2);
+
+  glmm_store(dest[0], v0);
+  glmm_store(dest[1], v1);
+  glmm_store(dest[2], v2);
+  glmm_store(dest[3], glmm_load(m1[3]));
+}
+
+CGLM_INLINE
+void
+glm_inv_tr_wasm(mat4 mat) {
+  glmm_128 r0, r1, r2, r3, x0, x1, x2, x3, x4, x5;
+
+  r0 = glmm_load(mat[0]);
+  r1 = glmm_load(mat[1]);
+  r2 = glmm_load(mat[2]);
+  r3 = glmm_load(mat[3]);
+  x1 = wasm_f32x4_const(0.0f, 0.0f, 0.0f, 1.0f);
+
+  /* _MM_TRANSPOSE4_PS(r0, r1, r2, x1); */
+  x2 = wasm_i32x4_shuffle(r0, r1, 0, 4, 1, 5);
+  x3 = wasm_i32x4_shuffle(r0, r1, 2, 6, 3, 7);
+  x4 = wasm_i32x4_shuffle(r2, x1, 0, 4, 1, 5);
+  x5 = wasm_i32x4_shuffle(r2, x1, 2, 6, 3, 7);
+  /* r0 = _mm_movelh_ps(x2, x4); */
+  r0 = wasm_i32x4_shuffle(x2, x4, 0, 1, 4, 5);
+  /* r1 = _mm_movehl_ps(x4, x2); */
+  r1 = wasm_i32x4_shuffle(x4, x2, 6, 7, 2, 3);
+  /* r2 = _mm_movelh_ps(x3, x5); */
+  r2 = wasm_i32x4_shuffle(x3, x5, 0, 1, 4, 5);
+  /* x1 = _mm_movehl_ps(x5, x3); */
+  x1 = wasm_i32x4_shuffle(x5, x3, 6, 7, 2, 3);
+
+  x2 = glmm_shuff1(r3, 0, 0, 0, 0);
+  x3 = glmm_shuff1(r3, 1, 1, 1, 1);
+  x4 = glmm_shuff1(r3, 2, 2, 2, 2);
+
+  x0 = glmm_fmadd(r0, x2,
+                  glmm_fmadd(r1, x3, wasm_f32x4_mul(r2, x4)));
+  x0 = wasm_f32x4_neg(x0);
+
+  x0 = wasm_f32x4_add(x0, x1);
+
+  glmm_store(mat[0], r0);
+  glmm_store(mat[1], r1);
+  glmm_store(mat[2], r2);
+  glmm_store(mat[3], x0);
+}
+
+#endif
+#endif /* cglm_affine_mat_wasm_h */

include/cglm/simd/wasm/mat2.h

@@ -0,0 +1,50 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglm_mat2_wasm_h
+#define cglm_mat2_wasm_h
+#if defined(__wasm__) && defined(__wasm_simd128__)
+
+#include "../../common.h"
+#include "../intrin.h"
+
+CGLM_INLINE
+void
+glm_mat2_mul_wasm(mat2 m1, mat2 m2, mat2 dest) {
+  glmm_128 x0, x1, x2, x3, x4;
+
+  x1 = glmm_load(m1[0]); /* d c b a */
+  x2 = glmm_load(m2[0]); /* h g f e */
+
+  x3 = glmm_shuff1(x2, 2, 2, 0, 0);
+  x4 = glmm_shuff1(x2, 3, 3, 1, 1);
+  /* x0 = _mm_movelh_ps(x1, x1); */
+  x0 = wasm_i32x4_shuffle(x1, x1, 0, 1, 4, 5);
+  /* x2 = _mm_movehl_ps(x1, x1); */
+  x2 = wasm_i32x4_shuffle(x1, x1, 6, 7, 2, 3);
+
+  /*
+   dest[0][0] = a * e + c * f;
+   dest[0][1] = b * e + d * f;
+   dest[1][0] = a * g + c * h;
+   dest[1][1] = b * g + d * h;
+   */
+  x0 = glmm_fmadd(x0, x3, wasm_f32x4_mul(x2, x4));
+
+  glmm_store(dest[0], x0);
+}
+
+CGLM_INLINE
+void
+glm_mat2_transp_wasm(mat2 m, mat2 dest) {
+  /* d c b a */
+  /* d b c a */
+  glmm_store(dest[0], glmm_shuff1(glmm_load(m[0]), 3, 1, 2, 0));
+}
+
+#endif
+#endif /* cglm_mat2_wasm_h */

include/cglm/simd/wasm/mat3.h

@@ -0,0 +1,85 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglm_mat3_wasm_h
+#define cglm_mat3_wasm_h
+#if defined(__wasm__) && defined(__wasm_simd128__)
+
+#include "../../common.h"
+#include "../intrin.h"
+
+CGLM_INLINE
+void
+glm_mat3_mul_wasm(mat3 m1, mat3 m2, mat3 dest) {
+  glmm_128 l0, l1, l2, r0, r1, r2, x0, x1, x2, x3, x4, x5, x6, x7, x8, x9;
+  
+  l0 = wasm_v128_load(m1[0]);
+  l1 = wasm_v128_load(&m1[1][1]);
+
+  r0 = wasm_v128_load(m2[0]);
+  r1 = wasm_v128_load(&m2[1][1]);
+
+  x8 = glmm_shuff1(l0, 0, 2, 1, 0);                     /* a00 a02 a01 a00 */
+  x1 = glmm_shuff1(r0, 3, 0, 0, 0);                     /* b10 b00 b00 b00 */
+  x2 = wasm_i32x4_shuffle(l0, l1, 3, 3, 4, 5);          /* a12 a11 a10 a10 */
+  x3 = wasm_i32x4_shuffle(r0, r1, 1, 3, 4, 6);          /* b20 b11 b10 b01 */
+  x0 = wasm_f32x4_mul(x8, x1);
+
+  x6 = glmm_shuff1(l0, 1, 0, 2, 1);                     /* a01 a00 a02 a01 */
+  x7 = glmm_shuff1(x3, 3, 3, 1, 1);                     /* b20 b20 b10 b10 */
+  l2 = wasm_v128_load32_zero(&m1[2][2]);
+  r2 = wasm_v128_load32_zero(&m2[2][2]);
+  x1 = wasm_f32x4_mul(x6, x7);
+  l2 = glmm_shuff1(l2, 0, 0, 1, 0);                     /* a22 a22 0.f a22 */
+  r2 = glmm_shuff1(r2, 0, 0, 1, 0);                     /* b22 b22 0.f b22 */
+
+  x4 = glmm_shuff1(x2, 0, 3, 2, 0);                     /* a10 a12 a11 a10 */
+  x5 = glmm_shuff1(x2, 2, 0, 3, 2);                     /* a11 a10 a12 a11 */
+  x6 = glmm_shuff1(x3, 2, 0, 0, 0);                     /* b11 b01 b01 b01 */
+  x2 = glmm_shuff1(r1, 3, 3, 0, 0);                     /* b21 b21 b11 b11 */
+
+  /* x8 = _mm_unpackhi_ps(x8, x4); */
+  /* x9 = _mm_unpackhi_ps(x7, x2); */
+  x8 = wasm_i32x4_shuffle(x8, x4, 2, 6, 3, 7);          /* a10 a00 a12 a02 */
+  x9 = wasm_i32x4_shuffle(x7, x2, 2, 6, 3, 7);          /* b21 b20 b21 b20 */
+
+  x0 = glmm_fmadd(x4, x6, x0);
+  x1 = glmm_fmadd(x5, x2, x1);
+
+  /* x2 = _mm_movehl_ps(l2, l1); */
+  x2 = wasm_i32x4_shuffle(l2, l1, 6, 7, 2, 3);          /* a22 a22 a21 a20 */
+  x3 = glmm_shuff1(x2, 0, 2, 1, 0);                     /* a20 a22 a21 a20 */
+  x2 = glmm_shuff1(x2, 1, 0, 2, 1);                     /* a21 a20 a22 a21 */
+  x4 = wasm_i32x4_shuffle(r0, r1, 2, 2, 5, 5);          /* b12 b12 b02 b02 */
+  
+  x5 = glmm_shuff1(x4, 3, 0, 0, 0);                     /* b12 b02 b02 b02 */
+  /* x4 = _mm_movehl_ps(r2, x4); */
+  x4 = wasm_i32x4_shuffle(r2, x4, 6, 7, 2, 3);          /* b22 b22 b12 b12 */
+  x0 = glmm_fmadd(x3, x5, x0);
+  x1 = glmm_fmadd(x2, x4, x1);
+
+  /*
+   Dot Product : dest[2][2] =  a02 * b20 +
+                               a12 * b21 +
+                               a22 * b22 +
+                               0   * 00                                    */
+  /* x2 = _mm_movelh_ps(x8, l2); */
+  /* x3 = _mm_movelh_ps(x9, r2); */
+  x2 = wasm_i32x4_shuffle(x8, l2, 0, 1, 4, 5);           /* 0.f a22 a12 a02 */
+  x3 = wasm_i32x4_shuffle(x9, r2, 0, 1, 4, 5);           /* 0.f b22 b21 b20 */
+  x2 = glmm_vdots(x2, x3);
+
+  /* _mm_storeu_ps(&dest[0][0], x0); */
+  wasm_v128_store(&dest[0][0], x0);
+  /* _mm_storeu_ps(&dest[1][1], x1); */
+  wasm_v128_store(&dest[1][1], x1);
+  /* _mm_store_ss (&dest[2][2], x2); */
+  wasm_v128_store32_lane(&dest[2][2], x2, 0);
+}
+
+#endif
+#endif /* cglm_mat3_wasm_h */

include/cglm/simd/wasm/mat4.h

@@ -0,0 +1,454 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglm_mat_wasm_h
+#define cglm_mat_wasm_h
+#if defined(__wasm__) && defined(__wasm_simd128__)
+
+#include "../../common.h"
+#include "../intrin.h"
+
+#define glm_mat4_inv_precise_wasm(mat, dest) glm_mat4_inv_wasm(mat, dest)
+
+CGLM_INLINE
+void
+glm_mat4_scale_wasm(mat4 m, float s) {
+  glmm_128 x0;
+  x0 = wasm_f32x4_splat(s);
+
+  glmm_store(m[0], wasm_f32x4_mul(glmm_load(m[0]), x0));
+  glmm_store(m[1], wasm_f32x4_mul(glmm_load(m[1]), x0));
+  glmm_store(m[2], wasm_f32x4_mul(glmm_load(m[2]), x0));
+  glmm_store(m[3], wasm_f32x4_mul(glmm_load(m[3]), x0));
+}
+
+CGLM_INLINE
+void
+glm_mat4_transp_wasm(mat4 m, mat4 dest) {
+  glmm_128 r0, r1, r2, r3, tmp0, tmp1, tmp2, tmp3;
+
+  r0 = glmm_load(m[0]);
+  r1 = glmm_load(m[1]);
+  r2 = glmm_load(m[2]);
+  r3 = glmm_load(m[3]);
+
+  /* _MM_TRANSPOSE4_PS(r0, r1, r2, r3); */
+  tmp0 = wasm_i32x4_shuffle(r0, r1, 0, 4, 1, 5);
+  tmp1 = wasm_i32x4_shuffle(r0, r1, 2, 6, 3, 7);
+  tmp2 = wasm_i32x4_shuffle(r2, r3, 0, 4, 1, 5);
+  tmp3 = wasm_i32x4_shuffle(r2, r3, 2, 6, 3, 7);
+  /* r0 = _mm_movelh_ps(tmp0, tmp2); */
+  r0 = wasm_i32x4_shuffle(tmp0, tmp2, 0, 1, 4, 5);
+  /* r1 = _mm_movehl_ps(tmp2, tmp0); */
+  r1 = wasm_i32x4_shuffle(tmp2, tmp0, 6, 7, 2, 3);
+  /* r2 = _mm_movelh_ps(tmp1, tmp3); */
+  r2 = wasm_i32x4_shuffle(tmp1, tmp3, 0, 1, 4, 5);
+  /* r3 = _mm_movehl_ps(tmp3, tmp1); */
+  r3 = wasm_i32x4_shuffle(tmp3, tmp1, 6, 7, 2, 3);
+
+  glmm_store(dest[0], r0);
+  glmm_store(dest[1], r1);
+  glmm_store(dest[2], r2);
+  glmm_store(dest[3], r3);
+}
+
+CGLM_INLINE
+void
+glm_mat4_mul_wasm(mat4 m1, mat4 m2, mat4 dest) {
+  /* D = R * L (Column-Major) */
+
+  glmm_128 l, r0, r1, r2, r3, v0, v1, v2, v3;
+
+  l  = glmm_load(m1[0]);
+  r0 = glmm_load(m2[0]);
+  r1 = glmm_load(m2[1]);
+  r2 = glmm_load(m2[2]);
+  r3 = glmm_load(m2[3]);
+
+  v0 = wasm_f32x4_mul(glmm_splat_x(r0), l);
+  v1 = wasm_f32x4_mul(glmm_splat_x(r1), l);
+  v2 = wasm_f32x4_mul(glmm_splat_x(r2), l);
+  v3 = wasm_f32x4_mul(glmm_splat_x(r3), l);
+
+  l  = glmm_load(m1[1]);
+  v0 = glmm_fmadd(glmm_splat_y(r0), l, v0);
+  v1 = glmm_fmadd(glmm_splat_y(r1), l, v1);
+  v2 = glmm_fmadd(glmm_splat_y(r2), l, v2);
+  v3 = glmm_fmadd(glmm_splat_y(r3), l, v3);
+
+  l  = glmm_load(m1[2]);
+  v0 = glmm_fmadd(glmm_splat_z(r0), l, v0);
+  v1 = glmm_fmadd(glmm_splat_z(r1), l, v1);
+  v2 = glmm_fmadd(glmm_splat_z(r2), l, v2);
+  v3 = glmm_fmadd(glmm_splat_z(r3), l, v3);
+
+  l  = glmm_load(m1[3]);
+  v0 = glmm_fmadd(glmm_splat_w(r0), l, v0);
+  v1 = glmm_fmadd(glmm_splat_w(r1), l, v1);
+  v2 = glmm_fmadd(glmm_splat_w(r2), l, v2);
+  v3 = glmm_fmadd(glmm_splat_w(r3), l, v3);
+
+  glmm_store(dest[0], v0);
+  glmm_store(dest[1], v1);
+  glmm_store(dest[2], v2);
+  glmm_store(dest[3], v3);
+}
+
+CGLM_INLINE
+void
+glm_mat4_mulv_wasm(mat4 m, vec4 v, vec4 dest) {
+  glmm_128 x0, x1, m0, m1, m2, m3, v0, v1, v2, v3;
+
+  m0 = glmm_load(m[0]);
+  m1 = glmm_load(m[1]);
+  m2 = glmm_load(m[2]);
+  m3 = glmm_load(m[3]);
+
+  x0 = glmm_load(v);
+  v0 = glmm_splat_x(x0);
+  v1 = glmm_splat_y(x0);
+  v2 = glmm_splat_z(x0);
+  v3 = glmm_splat_w(x0);
+
+  x1 = wasm_f32x4_mul(m3, v3);
+  x1 = glmm_fmadd(m2, v2, x1);
+  x1 = glmm_fmadd(m1, v1, x1);
+  x1 = glmm_fmadd(m0, v0, x1);
+
+  glmm_store(dest, x1);
+}
+
+CGLM_INLINE
+float
+glm_mat4_det_wasm(mat4 mat) {
+  glmm_128 r0, r1, r2, r3, x0, x1, x2;
+
+  /* 127 <- 0, [square] det(A) = det(At) */
+  r0 = glmm_load(mat[0]); /* d c b a */
+  r1 = glmm_load(mat[1]); /* h g f e */
+  r2 = glmm_load(mat[2]); /* l k j i */
+  r3 = glmm_load(mat[3]); /* p o n m */
+
+  /*
+   t[1] = j * p - n * l;
+   t[2] = j * o - n * k;
+   t[3] = i * p - m * l;
+   t[4] = i * o - m * k;
+   */
+  x0 = glmm_fnmadd(glmm_shuff1(r3, 0, 0, 1, 1), glmm_shuff1(r2, 2, 3, 2, 3),
+                   wasm_f32x4_mul(glmm_shuff1(r2, 0, 0, 1, 1),
+                              glmm_shuff1(r3, 2, 3, 2, 3)));
+  /*
+   t[0] = k * p - o * l;
+   t[0] = k * p - o * l;
+   t[5] = i * n - m * j;
+   t[5] = i * n - m * j;
+   */
+  x1 = glmm_fnmadd(glmm_shuff1(r3, 0, 0, 2, 2), glmm_shuff1(r2, 1, 1, 3, 3),
+                   wasm_f32x4_mul(glmm_shuff1(r2, 0, 0, 2, 2),
+                              glmm_shuff1(r3, 1, 1, 3, 3)));
+
+  /*
+     a * (f * t[0] - g * t[1] + h * t[2])
+   - b * (e * t[0] - g * t[3] + h * t[4])
+   + c * (e * t[1] - f * t[3] + h * t[5])
+   - d * (e * t[2] - f * t[4] + g * t[5])
+   */
+  x2 = glmm_fnmadd(glmm_shuff1(r1, 1, 1, 2, 2), glmm_shuff1(x0, 3, 2, 2, 0),
+                   wasm_f32x4_mul(glmm_shuff1(r1, 0, 0, 0, 1),
+                              wasm_i32x4_shuffle(x1, x0, 0, 0, 4, 5)));
+  x2 = glmm_fmadd(glmm_shuff1(r1, 2, 3, 3, 3),
+                  wasm_i32x4_shuffle(x0, x1, 1, 3, 6, 6),
+                  x2);
+  /* x2 = wasm_v128_xor(x2, wasm_f32x4_const(0.f, -0.f, 0.f, -0.f)); */
+  x2 = wasm_v128_xor(x2, glmm_float32x4_SIGNMASK_PNPN);
+  
+  return glmm_hadd(wasm_f32x4_mul(x2, r0));
+}
+
+CGLM_INLINE
+void
+glm_mat4_inv_fast_wasm(mat4 mat, mat4 dest) {
+  glmm_128 r0, r1, r2, r3,
+         v0, v1, v2, v3,
+         t0, t1, t2, t3, t4, t5,
+         x0, x1, x2, x3, x4, x5, x6, x7, x8, x9;
+
+  /* x8 = wasm_f32x4_const(0.f, -0.f, 0.f, -0.f); */
+  x8 = glmm_float32x4_SIGNMASK_PNPN;
+  x9 = glmm_shuff1(x8, 2, 1, 2, 1);
+
+  /* 127 <- 0 */
+  r0 = glmm_load(mat[0]); /* d c b a */
+  r1 = glmm_load(mat[1]); /* h g f e */
+  r2 = glmm_load(mat[2]); /* l k j i */
+  r3 = glmm_load(mat[3]); /* p o n m */
+  /* x0 = _mm_movehl_ps(r3, r2); */
+  x0 = wasm_i32x4_shuffle(r3, r2, 6, 7, 2, 3);           /* p o l k */
+  /* x3 = _mm_movelh_ps(r2, r3); */
+  x3 = wasm_i32x4_shuffle(r2, r3, 0, 1, 4, 5);           /* n m j i */
+  x1 = glmm_shuff1(x0, 1, 3, 3 ,3);                      /* l p p p */
+  x2 = glmm_shuff1(x0, 0, 2, 2, 2);                      /* k o o o */
+  x4 = glmm_shuff1(x3, 1, 3, 3, 3);                      /* j n n n */
+  x7 = glmm_shuff1(x3, 0, 2, 2, 2);                      /* i m m m */
+
+  x6 = wasm_i32x4_shuffle(r2, r1, 0, 0, 4, 4);           /* e e i i */
+  x5 = wasm_i32x4_shuffle(r2, r1, 1, 1, 5, 5);           /* f f j j */
+  x3 = wasm_i32x4_shuffle(r2, r1, 2, 2, 6, 6);           /* g g k k */
+  x0 = wasm_i32x4_shuffle(r2, r1, 3, 3, 7, 7);           /* h h l l */
+  
+  t0 = wasm_f32x4_mul(x3, x1);
+  t1 = wasm_f32x4_mul(x5, x1);
+  t2 = wasm_f32x4_mul(x5, x2);
+  t3 = wasm_f32x4_mul(x6, x1);
+  t4 = wasm_f32x4_mul(x6, x2);
+  t5 = wasm_f32x4_mul(x6, x4);
+  
+  /* t1[0] = k * p - o * l;
+     t1[0] = k * p - o * l;
+     t2[0] = g * p - o * h;
+     t3[0] = g * l - k * h; */
+  t0 = glmm_fnmadd(x2, x0, t0);
+  
+  /* t1[1] = j * p - n * l;
+     t1[1] = j * p - n * l;
+     t2[1] = f * p - n * h;
+     t3[1] = f * l - j * h; */
+   t1 = glmm_fnmadd(x4, x0, t1);
+  
+  /* t1[2] = j * o - n * k
+     t1[2] = j * o - n * k;
+     t2[2] = f * o - n * g;
+     t3[2] = f * k - j * g; */
+  t2 = glmm_fnmadd(x4, x3, t2);
+  
+  /* t1[3] = i * p - m * l;
+     t1[3] = i * p - m * l;
+     t2[3] = e * p - m * h;
+     t3[3] = e * l - i * h; */
+  t3 = glmm_fnmadd(x7, x0, t3);
+  
+  /* t1[4] = i * o - m * k;
+     t1[4] = i * o - m * k;
+     t2[4] = e * o - m * g;
+     t3[4] = e * k - i * g; */
+  t4 = glmm_fnmadd(x7, x3, t4);
+  
+  /* t1[5] = i * n - m * j;
+     t1[5] = i * n - m * j;
+     t2[5] = e * n - m * f;
+     t3[5] = e * j - i * f; */
+  t5 = glmm_fnmadd(x7, x5, t5);
+  /* x4 = _mm_movelh_ps(r0, r1); */
+  x4 = wasm_i32x4_shuffle(r0, r1, 0, 1, 4, 5);           /* f e b a */
+  /* x5 = _mm_movehl_ps(r1, r0); */
+  x5 = wasm_i32x4_shuffle(r1, r0, 6, 7, 2, 3);           /* h g d c */
+  
+  x0 = glmm_shuff1(x4, 0, 0, 0, 2);                      /* a a a e */
+  x1 = glmm_shuff1(x4, 1, 1, 1, 3);                      /* b b b f */
+  x2 = glmm_shuff1(x5, 0, 0, 0, 2);                      /* c c c g */
+  x3 = glmm_shuff1(x5, 1, 1, 1, 3);                      /* d d d h */
+  
+  v2 = wasm_f32x4_mul(x0, t1);
+  v1 = wasm_f32x4_mul(x0, t0);
+  v3 = wasm_f32x4_mul(x0, t2);
+  v0 = wasm_f32x4_mul(x1, t0);
+  
+  v2 = glmm_fnmadd(x1, t3, v2);
+  v3 = glmm_fnmadd(x1, t4, v3);
+  v0 = glmm_fnmadd(x2, t1, v0);
+  v1 = glmm_fnmadd(x2, t3, v1);
+  
+  v3 = glmm_fmadd(x2, t5, v3);
+  v0 = glmm_fmadd(x3, t2, v0);
+  v2 = glmm_fmadd(x3, t5, v2);
+  v1 = glmm_fmadd(x3, t4, v1);
+
+  /*
+   dest[0][0] =  f * t1[0] - g * t1[1] + h * t1[2];
+   dest[0][1] =-(b * t1[0] - c * t1[1] + d * t1[2]);
+   dest[0][2] =  b * t2[0] - c * t2[1] + d * t2[2];
+   dest[0][3] =-(b * t3[0] - c * t3[1] + d * t3[2]); */
+  v0 = wasm_v128_xor(v0, x8);
+  
+  /*
+   dest[2][0] =  e * t1[1] - f * t1[3] + h * t1[5];
+   dest[2][1] =-(a * t1[1] - b * t1[3] + d * t1[5]);
+   dest[2][2] =  a * t2[1] - b * t2[3] + d * t2[5];
+   dest[2][3] =-(a * t3[1] - b * t3[3] + d * t3[5]);*/
+  v2 = wasm_v128_xor(v2, x8);
+
+  /*
+   dest[1][0] =-(e * t1[0] - g * t1[3] + h * t1[4]);
+   dest[1][1] =  a * t1[0] - c * t1[3] + d * t1[4];
+   dest[1][2] =-(a * t2[0] - c * t2[3] + d * t2[4]);
+   dest[1][3] =  a * t3[0] - c * t3[3] + d * t3[4]; */
+  v1 = wasm_v128_xor(v1, x9);
+
+  /*
+   dest[3][0] =-(e * t1[2] - f * t1[4] + g * t1[5]);
+   dest[3][1] =  a * t1[2] - b * t1[4] + c * t1[5];
+   dest[3][2] =-(a * t2[2] - b * t2[4] + c * t2[5]);
+   dest[3][3] =  a * t3[2] - b * t3[4] + c * t3[5]; */
+  v3 = wasm_v128_xor(v3, x9);
+
+  /* determinant */
+  x0 = wasm_i32x4_shuffle(v0, v1, 0, 0, 4, 4);
+  x1 = wasm_i32x4_shuffle(v2, v3, 0, 0, 4, 4);
+  x0 = wasm_i32x4_shuffle(x0, x1, 0, 2, 4, 6);
+
+  /* x0 = _mm_rcp_ps(glmm_vhadd(wasm_f32x4_mul(x0, r0))); */
+  x0 = wasm_f32x4_div(wasm_f32x4_const_splat(1.0f),
+                      glmm_vhadd(wasm_f32x4_mul(x0, r0)));
+
+  glmm_store(dest[0], wasm_f32x4_mul(v0, x0));
+  glmm_store(dest[1], wasm_f32x4_mul(v1, x0));
+  glmm_store(dest[2], wasm_f32x4_mul(v2, x0));
+  glmm_store(dest[3], wasm_f32x4_mul(v3, x0));
+}
+
+CGLM_INLINE
+void
+glm_mat4_inv_wasm(mat4 mat, mat4 dest) {
+  glmm_128 r0, r1, r2, r3,
+         v0, v1, v2, v3,
+         t0, t1, t2, t3, t4, t5,
+         x0, x1, x2, x3, x4, x5, x6, x7, x8, x9;
+
+  /* x8 = wasm_f32x4_const(0.f, -0.f, 0.f, -0.f); */
+  x8 = glmm_float32x4_SIGNMASK_PNPN;
+  x9 = glmm_shuff1(x8, 2, 1, 2, 1);
+
+  /* 127 <- 0 */
+  r0 = glmm_load(mat[0]); /* d c b a */
+  r1 = glmm_load(mat[1]); /* h g f e */
+  r2 = glmm_load(mat[2]); /* l k j i */
+  r3 = glmm_load(mat[3]); /* p o n m */
+  /* x0 = _mm_movehl_ps(r3, r2); */
+  x0 = wasm_i32x4_shuffle(r3, r2, 6, 7, 2, 3);           /* p o l k */
+  /* x3 = _mm_movelh_ps(r2, r3); */
+  x3 = wasm_i32x4_shuffle(r2, r3, 0, 1, 4, 5);           /* n m j i */
+  x1 = glmm_shuff1(x0, 1, 3, 3 ,3);                      /* l p p p */
+  x2 = glmm_shuff1(x0, 0, 2, 2, 2);                      /* k o o o */
+  x4 = glmm_shuff1(x3, 1, 3, 3, 3);                      /* j n n n */
+  x7 = glmm_shuff1(x3, 0, 2, 2, 2);                      /* i m m m */
+
+  x6 = wasm_i32x4_shuffle(r2, r1, 0, 0, 4, 4);           /* e e i i */
+  x5 = wasm_i32x4_shuffle(r2, r1, 1, 1, 5, 5);           /* f f j j */
+  x3 = wasm_i32x4_shuffle(r2, r1, 2, 2, 6, 6);           /* g g k k */
+  x0 = wasm_i32x4_shuffle(r2, r1, 3, 3, 7, 7);           /* h h l l */
+  
+  t0 = wasm_f32x4_mul(x3, x1);
+  t1 = wasm_f32x4_mul(x5, x1);
+  t2 = wasm_f32x4_mul(x5, x2);
+  t3 = wasm_f32x4_mul(x6, x1);
+  t4 = wasm_f32x4_mul(x6, x2);
+  t5 = wasm_f32x4_mul(x6, x4);
+  
+  /* t1[0] = k * p - o * l;
+     t1[0] = k * p - o * l;
+     t2[0] = g * p - o * h;
+     t3[0] = g * l - k * h; */
+  t0 = glmm_fnmadd(x2, x0, t0);
+  
+  /* t1[1] = j * p - n * l;
+     t1[1] = j * p - n * l;
+     t2[1] = f * p - n * h;
+     t3[1] = f * l - j * h; */
+   t1 = glmm_fnmadd(x4, x0, t1);
+  
+  /* t1[2] = j * o - n * k
+     t1[2] = j * o - n * k;
+     t2[2] = f * o - n * g;
+     t3[2] = f * k - j * g; */
+  t2 = glmm_fnmadd(x4, x3, t2);
+  
+  /* t1[3] = i * p - m * l;
+     t1[3] = i * p - m * l;
+     t2[3] = e * p - m * h;
+     t3[3] = e * l - i * h; */
+  t3 = glmm_fnmadd(x7, x0, t3);
+  
+  /* t1[4] = i * o - m * k;
+     t1[4] = i * o - m * k;
+     t2[4] = e * o - m * g;
+     t3[4] = e * k - i * g; */
+  t4 = glmm_fnmadd(x7, x3, t4);
+  
+  /* t1[5] = i * n - m * j;
+     t1[5] = i * n - m * j;
+     t2[5] = e * n - m * f;
+     t3[5] = e * j - i * f; */
+  t5 = glmm_fnmadd(x7, x5, t5);
+  /* x4 = _mm_movelh_ps(r0, r1); */
+  x4 = wasm_i32x4_shuffle(r0, r1, 0, 1, 4, 5);           /* f e b a */
+  /* x5 = _mm_movehl_ps(r1, r0); */
+  x5 = wasm_i32x4_shuffle(r1, r0, 6, 7, 2, 3);           /* h g d c */
+  
+  x0 = glmm_shuff1(x4, 0, 0, 0, 2);                      /* a a a e */
+  x1 = glmm_shuff1(x4, 1, 1, 1, 3);                      /* b b b f */
+  x2 = glmm_shuff1(x5, 0, 0, 0, 2);                      /* c c c g */
+  x3 = glmm_shuff1(x5, 1, 1, 1, 3);                      /* d d d h */
+  
+  v2 = wasm_f32x4_mul(x0, t1);
+  v1 = wasm_f32x4_mul(x0, t0);
+  v3 = wasm_f32x4_mul(x0, t2);
+  v0 = wasm_f32x4_mul(x1, t0);
+  
+  v2 = glmm_fnmadd(x1, t3, v2);
+  v3 = glmm_fnmadd(x1, t4, v3);
+  v0 = glmm_fnmadd(x2, t1, v0);
+  v1 = glmm_fnmadd(x2, t3, v1);
+  
+  v3 = glmm_fmadd(x2, t5, v3);
+  v0 = glmm_fmadd(x3, t2, v0);
+  v2 = glmm_fmadd(x3, t5, v2);
+  v1 = glmm_fmadd(x3, t4, v1);
+
+  /*
+   dest[0][0] =  f * t1[0] - g * t1[1] + h * t1[2];
+   dest[0][1] =-(b * t1[0] - c * t1[1] + d * t1[2]);
+   dest[0][2] =  b * t2[0] - c * t2[1] + d * t2[2];
+   dest[0][3] =-(b * t3[0] - c * t3[1] + d * t3[2]); */
+  v0 = wasm_v128_xor(v0, x8);
+  
+  /*
+   dest[2][0] =  e * t1[1] - f * t1[3] + h * t1[5];
+   dest[2][1] =-(a * t1[1] - b * t1[3] + d * t1[5]);
+   dest[2][2] =  a * t2[1] - b * t2[3] + d * t2[5];
+   dest[2][3] =-(a * t3[1] - b * t3[3] + d * t3[5]);*/
+  v2 = wasm_v128_xor(v2, x8);
+
+  /*
+   dest[1][0] =-(e * t1[0] - g * t1[3] + h * t1[4]);
+   dest[1][1] =  a * t1[0] - c * t1[3] + d * t1[4];
+   dest[1][2] =-(a * t2[0] - c * t2[3] + d * t2[4]);
+   dest[1][3] =  a * t3[0] - c * t3[3] + d * t3[4]; */
+  v1 = wasm_v128_xor(v1, x9);
+
+  /*
+   dest[3][0] =-(e * t1[2] - f * t1[4] + g * t1[5]);
+   dest[3][1] =  a * t1[2] - b * t1[4] + c * t1[5];
+   dest[3][2] =-(a * t2[2] - b * t2[4] + c * t2[5]);
+   dest[3][3] =  a * t3[2] - b * t3[4] + c * t3[5]; */
+  v3 = wasm_v128_xor(v3, x9);
+
+  /* determinant */
+  x0 = wasm_i32x4_shuffle(v0, v1, 0, 0, 4, 4);
+  x1 = wasm_i32x4_shuffle(v2, v3, 0, 0, 4, 4);
+  x0 = wasm_i32x4_shuffle(x0, x1, 0, 2, 4, 6);
+
+  x0 = wasm_f32x4_div(wasm_f32x4_splat(1.0f), glmm_vhadd(wasm_f32x4_mul(x0, r0)));
+
+  glmm_store(dest[0], wasm_f32x4_mul(v0, x0));
+  glmm_store(dest[1], wasm_f32x4_mul(v1, x0));
+  glmm_store(dest[2], wasm_f32x4_mul(v2, x0));
+  glmm_store(dest[3], wasm_f32x4_mul(v3, x0));
+}
+
+#endif
+#endif /* cglm_mat_wasm_h */

include/cglm/simd/wasm/quat.h

@@ -0,0 +1,55 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglm_quat_wasm_h
+#define cglm_quat_wasm_h
+#if defined(__wasm__) && defined(__wasm_simd128__)
+
+#include "../../common.h"
+#include "../intrin.h"
+
+CGLM_INLINE
+void
+glm_quat_mul_wasm(versor p, versor q, versor dest) {
+  /*
+   + (a1 b2 + b1 a2 + c1 d2 − d1 c2)i
+   + (a1 c2 − b1 d2 + c1 a2 + d1 b2)j
+   + (a1 d2 + b1 c2 − c1 b2 + d1 a2)k
+     a1 a2 − b1 b2 − c1 c2 − d1 d2
+   */
+
+  glmm_128 xp, xq, x1, x2, x3, r, x, y, z;
+
+  xp = glmm_load(p); /* 3 2 1 0 */
+  xq = glmm_load(q);
+  /* x1 = wasm_f32x4_const(0.f, -0.f, 0.f, -0.f); */
+  x1 = glmm_float32x4_SIGNMASK_PNPN; /* TODO: _mm_set1_ss() + shuff ? */
+  r  = wasm_f32x4_mul(glmm_splat_w(xp), xq);
+  /* x2 = _mm_unpackhi_ps(x1, x1); */
+  x2 = wasm_i32x4_shuffle(x1, x1, 2, 6, 3, 7);
+  x3 = glmm_shuff1(x1, 3, 2, 0, 1);
+  x  = glmm_splat_x(xp);
+  y  = glmm_splat_y(xp);
+  z  = glmm_splat_z(xp);
+
+  x  = wasm_v128_xor(x, x1);
+  y  = wasm_v128_xor(y, x2);
+  z  = wasm_v128_xor(z, x3);
+  
+  x1 = glmm_shuff1(xq, 0, 1, 2, 3);
+  x2 = glmm_shuff1(xq, 1, 0, 3, 2);
+  x3 = glmm_shuff1(xq, 2, 3, 0, 1);
+  
+  r  = glmm_fmadd(x, x1, r);
+  r  = glmm_fmadd(y, x2, r);
+  r  = glmm_fmadd(z, x3, r);
+
+  glmm_store(dest, r);
+}
+
+#endif
+#endif /* cglm_quat_wasm_h */

include/cglm/simd/x86.h

@@ -54,12 +54,29 @@
 #  endif
 #endif
 
+/* Note that `0x80000000` corresponds to `INT_MIN` for a 32-bit int. */
+#define GLMM_NEGZEROf ((int)0x80000000) /*  0x80000000 ---> -0.0f  */
+
+#define GLMM__SIGNMASKf(X, Y, Z, W)                                           \
+   _mm_castsi128_ps(_mm_set_epi32(X, Y, Z, W))
+  /* _mm_set_ps(X, Y, Z, W); */
+
+#define glmm_float32x4_SIGNMASK_PNPN GLMM__SIGNMASKf(0, GLMM_NEGZEROf, 0, GLMM_NEGZEROf)
+#define glmm_float32x4_SIGNMASK_NPNP GLMM__SIGNMASKf(GLMM_NEGZEROf, 0, GLMM_NEGZEROf, 0)
+#define glmm_float32x4_SIGNMASK_NPPN GLMM__SIGNMASKf(GLMM_NEGZEROf, 0, 0, GLMM_NEGZEROf)
+
+#define glmm_float32x4_SIGNMASK_NEG _mm_castsi128_ps(_mm_set1_epi32(GLMM_NEGZEROf)) /* _mm_set1_ps(-0.0f) */
+#define glmm_float32x8_SIGNMASK_NEG _mm256_castsi256_ps(_mm256_set1_epi32(GLMM_NEGZEROf))
+
 static inline
 __m128
 glmm_abs(__m128 x) {
-  return _mm_andnot_ps(_mm_set1_ps(-0.0f), x);
+  return _mm_andnot_ps(glmm_float32x4_SIGNMASK_NEG, x);
 }
 
+static inline __m128 glmm_min(__m128 a, __m128 b) { return _mm_min_ps(a, b); }
+static inline __m128 glmm_max(__m128 a, __m128 b) { return _mm_max_ps(a, b); }
+
 static inline
 __m128
 glmm_vhadd(__m128 v) {
@@ -256,7 +273,8 @@ glmm_fnmsub(__m128 a, __m128 b, __m128 c) {
 #ifdef __FMA__
   return _mm_fnmsub_ps(a, b, c);
 #else
-  return _mm_xor_ps(_mm_add_ps(_mm_mul_ps(a, b), c), _mm_set1_ps(-0.0f));
+  return _mm_xor_ps(_mm_add_ps(_mm_mul_ps(a, b), c),
+                    glmm_float32x4_SIGNMASK_NEG);
 #endif
 }
 
@@ -298,7 +316,7 @@ glmm256_fnmsub(__m256 a, __m256 b, __m256 c) {
   return _mm256_fmsub_ps(a, b, c);
 #else
   return _mm256_xor_ps(_mm256_sub_ps(_mm256_mul_ps(a, b), c),
-                       _mm256_set1_ps(-0.0f));
+                       glmm_float32x8_SIGNMASK_NEG);
 #endif
 }
 #endif

include/cglm/struct.h

@@ -17,8 +17,14 @@ extern "C" {
 #include "struct/vec3.h"
 #include "struct/vec4.h"
 #include "struct/mat2.h"
+#include "struct/mat2x3.h"
+#include "struct/mat2x4.h"
 #include "struct/mat3.h"
+#include "struct/mat3x2.h"
+#include "struct/mat3x4.h"
 #include "struct/mat4.h"
+#include "struct/mat4x2.h"
+#include "struct/mat4x3.h"
 #include "struct/affine.h"
 #include "struct/frustum.h"
 #include "struct/plane.h"

include/cglm/struct/aabb2d.h

@@ -0,0 +1,235 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+#ifndef cglms_aabb2ds_h
+#define cglms_aabb2ds_h
+
+#include "../common.h"
+#include "../types-struct.h"
+#include "../aabb2d.h"
+#include "vec2.h"
+#include "vec4.h"
+#include "mat4.h"
+
+/* api definition */
+#define glms_aabb2d_(NAME) CGLM_STRUCTAPI(aabb, NAME)
+
+/*!
+ * @brief apply transform to Axis-Aligned Bounding Box
+ *
+ * @param[in]  aabb  bounding box
+ * @param[in]  m    transform matrix
+ * @param[out] dest transformed bounding box
+ */
+CGLM_INLINE
+void
+glms_aabb2d_(transform)(vec2s aabb[2], mat3s m, vec2s dest[2]) {
+  vec2 rawAabb[2];
+  vec2 rawDest[2];
+
+  glms_vec2_(unpack)(rawAabb, aabb, 2);
+  glm_aabb2d_transform(rawAabb, m.raw, rawDest);
+  glms_vec2_(pack)(dest, rawDest, 2);
+}
+
+/*!
+ * @brief merges two AABB bounding box and creates new one
+ *
+ * two box must be in same space, if one of box is in different space then
+ * you should consider to convert it's space by glm_box_space
+ *
+ * @param[in]  aabb1 bounding box 1
+ * @param[in]  aabb2 bounding box 2
+ * @param[out] dest merged bounding box
+ */
+CGLM_INLINE
+void
+glms_aabb2d_(merge)(vec2s aabb1[2], vec2s aabb2[2], vec2s dest[2]) {
+  vec2 rawAabb1[2];
+  vec2 rawAabb2[2];
+  vec2 rawDest[2];
+
+  glms_vec2_(unpack)(rawAabb1, aabb1, 2);
+  glms_vec2_(unpack)(rawAabb2, aabb2, 2);
+  glm_aabb2d_merge(rawAabb1, rawAabb2, rawDest);
+  glms_vec2_(pack)(dest, rawDest, 2);
+}
+
+/*!
+ * @brief crops a bounding box with another one.
+ *
+ * this could be useful for gettng a bbox which fits with view frustum and
+ * object bounding boxes. In this case you crop view frustum box with objects
+ * box
+ *
+ * @param[in]  aabb     bounding box 1
+ * @param[in]  cropAabb crop box
+ * @param[out] dest    cropped bounding box
+ */
+CGLM_INLINE
+void
+glms_aabb2d_(crop)(vec2s aabb[2], vec2s cropAabb[2], vec2s dest[2]) {
+  vec2 rawAabb[2];
+  vec2 rawCropAabb[2];
+  vec2 rawDest[2];
+
+  glms_vec2_(unpack)(rawAabb, aabb, 2);
+  glms_vec2_(unpack)(rawCropAabb, cropAabb, 2);
+  glm_aabb2d_crop(rawAabb, rawCropAabb, rawDest);
+  glms_vec2_(pack)(dest, rawDest, 2);
+}
+
+/*!
+ * @brief crops a bounding box with another one.
+ *
+ * this could be useful for gettng a bbox which fits with view frustum and
+ * object bounding boxes. In this case you crop view frustum box with objects
+ * box
+ *
+ * @param[in]  aabb      bounding box
+ * @param[in]  cropAabb  crop box
+ * @param[in]  clampAabb miniumum box
+ * @param[out] dest     cropped bounding box
+ */
+CGLM_INLINE
+void
+glms_aabb2d_(crop_until)(vec2s aabb[2],
+                       vec2s cropAabb[2],
+                       vec2s clampAabb[2],
+                       vec2s dest[2]) {
+  glms_aabb2d_(crop)(aabb, cropAabb, dest);
+  glms_aabb2d_(merge)(clampAabb, dest, dest);
+}
+
+/*!
+ * @brief invalidate AABB min and max values
+ *
+ * @param[in, out]  aabb bounding box
+ */
+CGLM_INLINE
+void
+glms_aabb2d_(invalidate)(vec2s box[2]) {
+  box[0] = glms_vec2_(fill)(FLT_MAX);
+  box[1] = glms_vec2_(fill)(-FLT_MAX);
+}
+
+/*!
+ * @brief check if AABB is valid or not
+ *
+ * @param[in]  aabb bounding box
+ */
+CGLM_INLINE
+bool
+glms_aabb2d_(isvalid)(vec2s aabb[2]) {
+  vec2 rawAabb[2];
+  glms_vec2_(unpack)(rawAabb, aabb, 2);
+  return glm_aabb2d_isvalid(rawAabb);
+}
+
+/*!
+ * @brief distance between of min and max
+ *
+ * @param[in]  aabb bounding box
+ */
+CGLM_INLINE
+float
+glms_aabb2d_(size)(vec2s aabb[2]) {
+  return glm_vec2_distance(aabb[0].raw, aabb[1].raw);
+}
+
+/*!
+ * @brief radius of sphere which surrounds AABB
+ *
+ * @param[in]  aabb bounding box
+ */
+CGLM_INLINE
+float
+glms_aabb2d_(radius)(vec2s aabb[2]) {
+  return glms_aabb2d_(size)(aabb) * 0.5f;
+}
+
+/*!
+ * @brief computes center point of AABB
+ *
+ * @param[in]   aabb  bounding box
+ * @returns center of bounding box
+ */
+CGLM_INLINE
+vec2s
+glms_aabb2d_(center)(vec2s aabb[2]) {
+  return glms_vec2_(center)(aabb[0], aabb[1]);
+}
+
+/*!
+ * @brief check if two AABB intersects
+ *
+ * @param[in]   aabb   bounding box
+ * @param[in]   other  other bounding box
+ */
+CGLM_INLINE
+bool
+glms_aabb2d_(aabb)(vec2s aabb[2], vec2s other[2]) {
+  vec2 rawAabb[2];
+  vec2 rawOther[2];
+
+  glms_vec2_(unpack)(rawAabb, aabb, 2);
+  glms_vec2_(unpack)(rawOther, other, 2);
+  return glm_aabb2d_aabb(rawAabb, rawOther);
+}
+
+/*!
+ * @brief check if AABB intersects with a circle
+ *
+ * https://github.com/erich666/GraphicsGems/blob/master/gems/BoxSphere.c
+ * Solid Box - Solid Sphere test.
+ *
+ * @param[in]   aabb   solid bounding box
+ * @param[in]   s      solid sphere
+ */
+CGLM_INLINE
+bool
+glms_aabb2d_(circle)(vec2s aabb[2], vec3s c) {
+  vec2 rawAabb[2];
+
+  glms_vec2_(unpack)(rawAabb, aabb, 2);
+  return glm_aabb2d_circle(rawAabb, c.raw);
+}
+
+/*!
+ * @brief check if point is inside of AABB
+ *
+ * @param[in]   aabb   bounding box
+ * @param[in]   point  point
+ */
+CGLM_INLINE
+bool
+glms_aabb2d_(point)(vec2s aabb[2], vec2s point) {
+  vec2 rawAabb[2];
+
+  glms_vec2_(unpack)(rawAabb, aabb, 2);
+  return glm_aabb2d_point(rawAabb, point.raw);
+}
+
+/*!
+ * @brief check if AABB contains other AABB
+ *
+ * @param[in]   box    bounding box
+ * @param[in]   other  other bounding box
+ */
+CGLM_INLINE
+bool
+glms_aabb2d_(contains)(vec2s aabb[2], vec2s other[2]) {
+  vec2 rawAabb[2];
+  vec2 rawOther[2];
+
+  glms_vec2_(unpack)(rawAabb, aabb, 2);
+  glms_vec2_(unpack)(rawOther, other, 2);
+  return glm_aabb2d_contains(rawAabb, rawOther);
+}
+
+#endif /* cglms_aabb2ds_h */
+

include/cglm/struct/affine-mat.h

@@ -0,0 +1,90 @@
+/*
+ * Copyright (c), Recep Aslantas.
+ *
+ * MIT License (MIT), http://opensource.org/licenses/MIT
+ * Full license can be found in the LICENSE file
+ */
+
+/*
+ Functions:
+   CGLM_INLINE mat4s glms_mul(mat4 m1, mat4 m2);
+   CGLM_INLINE mat4s glms_mul_rot(mat4 m1, mat4 m2);
+   CGLM_INLINE mat4s glms_inv_tr();
+ */
+
+#ifndef cglms_affine_mat_h
+#define cglms_affine_mat_h
+
+#include "../common.h"
+#include "../types-struct.h"
+#include "../affine-mat.h"
+#include "vec3.h"
+#include "vec4.h"
+#include "mat4.h"
+
+/*!
+ * @brief this is similar to glms_mat4_mul but specialized to affine transform
+ *
+ * Matrix format should be:
+ *   R  R  R  X
+ *   R  R  R  Y
+ *   R  R  R  Z
+ *   0  0  0  W
+ *
+ * this reduces some multiplications. It should be faster than mat4_mul.
+ * if you are not sure about matrix format then DON'T use this! use mat4_mul
+ *
+ * @param[in]   m1    affine matrix 1
+ * @param[in]   m2    affine matrix 2
+ * @returns         destination matrix
+ */
+CGLM_INLINE
+mat4s
+glms_mul(mat4s m1, mat4s m2){
+  mat4s r;
+  glm_mul(m1.raw, m2.raw, r.raw);
+  return r;
+}
+
+/*!
+ * @brief this is similar to glm_mat4_mul but specialized to affine transform
+ *
+ * Right Matrix format should be:
+ *   R  R  R  0
+ *   R  R  R  0
+ *   R  R  R  0
+ *   0  0  0  1
+ *
+ * this reduces some multiplications. It should be faster than mat4_mul.
+ * if you are not sure about matrix format then DON'T use this! use mat4_mul
+ *
+ * @param[in]   m1    affine matrix 1
+ * @param[in]   m2    affine matrix 2
+ * @returns         destination matrix
+ */
+CGLM_INLINE
+mat4s
+glms_mul_rot(mat4s m1, mat4s m2){
+  mat4s r;
+  glm_mul_rot(m1.raw, m2.raw, r.raw);
+  return r;
+}
+
+/*!
+ * @brief inverse orthonormal rotation + translation matrix (ridig-body)
+ *
+ * @code
+ * X = | R  T |   X' = | R' -R'T |
+ *     | 0  1 |        | 0     1 |
+ * @endcode
+ *
+ * @param[in]  m  matrix
+ * @returns      destination matrix
+ */
+CGLM_INLINE
+mat4s
+glms_inv_tr(mat4s m){
+  glm_inv_tr(m.raw);
+  return m;
+}
+#endif /* cglms_affine_mat_h */