Merge branch 'master' into optimize-inv

.github/workflows/ci.yml

@@ -13,8 +13,38 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        os: [macos-13, macos-14, ubuntu-22.04, ubuntu-24.04]
+        include:
-        simd: [none, sse, sse2, sse3, sse4, avx, avx2, neon]
+          # x86/x64 builds
+          - { os: macos-13, simd: none }
+          - { os: macos-13, simd: sse }
+          - { os: macos-13, simd: sse2 }
+          - { os: macos-13, simd: sse3 }
+          - { os: macos-13, simd: sse4 }
+          - { os: macos-13, simd: avx }
+          - { os: macos-13, simd: avx2 }
+          - { os: macos-14, simd: none }
+          - { os: macos-14, simd: sse }
+          - { os: macos-14, simd: sse2 }
+          - { os: macos-14, simd: sse3 }
+          - { os: macos-14, simd: sse4 }
+          - { os: macos-14, simd: avx }
+          - { os: macos-14, simd: avx2 }
+          - { os: ubuntu-22.04, simd: none }
+          - { os: ubuntu-22.04, simd: sse }
+          - { os: ubuntu-22.04, simd: sse2 }
+          - { os: ubuntu-22.04, simd: sse3 }
+          - { os: ubuntu-22.04, simd: sse4 }
+          - { os: ubuntu-22.04, simd: avx }
+          - { os: ubuntu-22.04, simd: avx2 }
+          - { os: ubuntu-24.04, simd: none }
+          - { os: ubuntu-24.04, simd: sse }
+          - { os: ubuntu-24.04, simd: sse2 }
+          - { os: ubuntu-24.04, simd: sse3 }
+          - { os: ubuntu-24.04, simd: sse4 }
+          - { os: ubuntu-24.04, simd: avx }
+          - { os: ubuntu-24.04, simd: avx2 }
+          # ARM64 builds
+          - { os: ubuntu-latest-arm64, simd: neon }
 
     steps:
     - uses: actions/checkout@v4
@@ -81,18 +111,58 @@ jobs:
       run: cmake --build build
 
   build_cmake_ubuntu:
-    name: CMake / ${{ matrix.target.os }} / ${{ matrix.target.cc }} / ${{ matrix.simd }}
+    name: CMake / ${{ matrix.target.os }} / ${{ matrix.target.cc }} / ${{ matrix.target.arch }} / ${{ matrix.target.simd }}
-    runs-on: ${{ matrix.target.os }}
+    runs-on: ${{ matrix.target.arch == 'arm64' && 'ubuntu-latest-arm64' || matrix.target.os }}
     strategy:
       fail-fast: false
       matrix:
         target:
-          - { os: ubuntu-20.04, cc: gcc-11 }
+          # GCC 11 builds
-          - { os: ubuntu-22.04, cc: gcc-12 }
+          - { os: ubuntu-20.04, cc: gcc-11, arch: x64, simd: none }
-          - { os: ubuntu-24.04, cc: gcc-13 }
+          - { os: ubuntu-20.04, cc: gcc-11, arch: x64, simd: sse }
-          - { os: ubuntu-20.04, cc: clang-12 }
+          - { os: ubuntu-20.04, cc: gcc-11, arch: x64, simd: sse2 }
-          - { os: ubuntu-22.04, cc: clang-15 }
+          - { os: ubuntu-20.04, cc: gcc-11, arch: x64, simd: sse3 }
-        simd: [none, sse, sse2, sse3, sse4, avx, avx2, neon]
+          - { os: ubuntu-20.04, cc: gcc-11, arch: x64, simd: sse4 }
+          - { os: ubuntu-20.04, cc: gcc-11, arch: x64, simd: avx }
+          - { os: ubuntu-20.04, cc: gcc-11, arch: x64, simd: avx2 }
+          # GCC 12 builds
+          - { os: ubuntu-22.04, cc: gcc-12, arch: x64, simd: none }
+          - { os: ubuntu-22.04, cc: gcc-12, arch: x64, simd: sse }
+          - { os: ubuntu-22.04, cc: gcc-12, arch: x64, simd: sse2 }
+          - { os: ubuntu-22.04, cc: gcc-12, arch: x64, simd: sse3 }
+          - { os: ubuntu-22.04, cc: gcc-12, arch: x64, simd: sse4 }
+          - { os: ubuntu-22.04, cc: gcc-12, arch: x64, simd: avx }
+          - { os: ubuntu-22.04, cc: gcc-12, arch: x64, simd: avx2 }
+          # GCC 13 builds
+          - { os: ubuntu-24.04, cc: gcc-13, arch: x64, simd: none }
+          - { os: ubuntu-24.04, cc: gcc-13, arch: x64, simd: sse }
+          - { os: ubuntu-24.04, cc: gcc-13, arch: x64, simd: sse2 }
+          - { os: ubuntu-24.04, cc: gcc-13, arch: x64, simd: sse3 }
+          - { os: ubuntu-24.04, cc: gcc-13, arch: x64, simd: sse4 }
+          - { os: ubuntu-24.04, cc: gcc-13, arch: x64, simd: avx }
+          - { os: ubuntu-24.04, cc: gcc-13, arch: x64, simd: avx2 }
+          # Clang 12 builds
+          - { os: ubuntu-20.04, cc: clang-12, arch: x64, simd: none }
+          - { os: ubuntu-20.04, cc: clang-12, arch: x64, simd: sse }
+          - { os: ubuntu-20.04, cc: clang-12, arch: x64, simd: sse2 }
+          - { os: ubuntu-20.04, cc: clang-12, arch: x64, simd: sse3 }
+          - { os: ubuntu-20.04, cc: clang-12, arch: x64, simd: sse4 }
+          - { os: ubuntu-20.04, cc: clang-12, arch: x64, simd: avx }
+          - { os: ubuntu-20.04, cc: clang-12, arch: x64, simd: avx2 }
+          # Clang 15 builds
+          - { os: ubuntu-22.04, cc: clang-15, arch: x64, simd: none }
+          - { os: ubuntu-22.04, cc: clang-15, arch: x64, simd: sse }
+          - { os: ubuntu-22.04, cc: clang-15, arch: x64, simd: sse2 }
+          - { os: ubuntu-22.04, cc: clang-15, arch: x64, simd: sse3 }
+          - { os: ubuntu-22.04, cc: clang-15, arch: x64, simd: sse4 }
+          - { os: ubuntu-22.04, cc: clang-15, arch: x64, simd: avx }
+          - { os: ubuntu-22.04, cc: clang-15, arch: x64, simd: avx2 }
+          # ARM64 builds
+          - { os: ubuntu-latest, cc: gcc-12, arch: arm64, simd: neon }
+          - { os: ubuntu-latest, cc: gcc-13, arch: arm64, simd: neon }
+          # ARMv7 builds
+          - { os: ubuntu-latest-arm64, cc: gcc-12, arch: armv7, simd: neon }
+          - { os: ubuntu-latest-arm64, cc: gcc-12, arch: armv7, simd: none }
 
     steps:
     - uses: actions/checkout@v4
@@ -130,7 +200,36 @@ jobs:
         fi
 
     - name: Configure CMake
-      run: cmake -B build -GNinja -DCMAKE_BUILD_TYPE=Release -DCMAKE_C_COMPILER=${{ matrix.target.cc }} -DCMAKE_C_FLAGS="$CFLAGS" -DCGLM_STATIC=ON -DCGLM_USE_TEST=ON
+      run: |
+        if [ "${{ matrix.target.arch }}" == "armv7" ]; then
+          # Build for ARMv7
+          neon_flags=""
+          if [ "${{ matrix.simd }}" == "neon" ]; then
+            neon_flags="-mfpu=neon -mfloat-abi=hard"
+          fi
+          cmake -B build -GNinja -DCMAKE_BUILD_TYPE=Release \
+            -DCMAKE_C_COMPILER=${{ matrix.target.cc }} \
+            -DCMAKE_C_FLAGS="$CFLAGS -m32 -march=armv7-a ${neon_flags}" \
+            -DCGLM_STATIC=ON -DCGLM_USE_TEST=ON
+        elif [ "${{ matrix.target.arch }}" == "arm64" ]; then
+          # Build for ARM64 (AArch64)
+          neon_flags=""
+          if [ "${{ matrix.simd }}" == "neon" ]; then
+            neon_flags="+simd" # Enable SIMD/NEON features on ARM64
+          else
+            neon_flags="+nosimd" # Explicitly disable SIMD/NEON
+          fi
+          cmake -B build -GNinja -DCMAKE_BUILD_TYPE=Release \
+            -DCMAKE_C_COMPILER=${{ matrix.target.cc }} \
+            -DCMAKE_C_FLAGS="$CFLAGS -march=armv8-a${neon_flags}" \
+            -DCGLM_STATIC=ON -DCGLM_USE_TEST=ON
+        else
+          # Normal build (x86/x64)
+          cmake -B build -GNinja -DCMAKE_BUILD_TYPE=Release \
+            -DCMAKE_C_COMPILER=${{ matrix.target.cc }} \
+            -DCMAKE_C_FLAGS="$CFLAGS" \
+            -DCGLM_STATIC=ON -DCGLM_USE_TEST=ON
+        fi
 
     - name: Build
       run: cmake --build build
@@ -176,8 +275,31 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        os: [macos-13, macos-14, windows-2022]
+        include:
-        simd: [none, sse, sse2, sse3, sse4, avx, avx2, neon]
+          # x86/x64 builds
+          - { os: macos-13, simd: none }
+          - { os: macos-13, simd: sse }
+          - { os: macos-13, simd: sse2 }
+          - { os: macos-13, simd: sse3 }
+          - { os: macos-13, simd: sse4 }
+          - { os: macos-13, simd: avx }
+          - { os: macos-13, simd: avx2 }
+          - { os: macos-14, simd: none }
+          - { os: macos-14, simd: sse }
+          - { os: macos-14, simd: sse2 }
+          - { os: macos-14, simd: sse3 }
+          - { os: macos-14, simd: sse4 }
+          - { os: macos-14, simd: avx }
+          - { os: macos-14, simd: avx2 }
+          - { os: windows-2022, simd: none }
+          - { os: windows-2022, simd: sse }
+          - { os: windows-2022, simd: sse2 }
+          - { os: windows-2022, simd: sse3 }
+          - { os: windows-2022, simd: sse4 }
+          - { os: windows-2022, simd: avx }
+          - { os: windows-2022, simd: avx2 }
+          # ARM64 builds
+          - { os: macos-14-arm64, simd: neon }
 
     steps:
     - uses: actions/checkout@v4
@@ -258,12 +380,42 @@ jobs:
 
   build_meson:
     name: Meson / ${{ matrix.os }} / ${{ matrix.simd }}
-    runs-on: ${{ matrix.os }}
+    runs-on: ${{ contains(matrix.os, 'arm64') && 'ubuntu-latest-arm64' || matrix.os }}
     strategy:
       fail-fast: false
       matrix:
-        os: [macos-14, ubuntu-22.04, ubuntu-24.04, windows-2022]
+        include:
-        simd: [none, sse, sse2, sse3, sse4, avx, avx2, neon]
+          # x86/x64 builds
+          - { os: macos-14, simd: none }
+          - { os: macos-14, simd: sse }
+          - { os: macos-14, simd: sse2 }
+          - { os: macos-14, simd: sse3 }
+          - { os: macos-14, simd: sse4 }
+          - { os: macos-14, simd: avx }
+          - { os: macos-14, simd: avx2 }
+          - { os: ubuntu-22.04, simd: none }
+          - { os: ubuntu-22.04, simd: sse }
+          - { os: ubuntu-22.04, simd: sse2 }
+          - { os: ubuntu-22.04, simd: sse3 }
+          - { os: ubuntu-22.04, simd: sse4 }
+          - { os: ubuntu-22.04, simd: avx }
+          - { os: ubuntu-22.04, simd: avx2 }
+          - { os: ubuntu-24.04, simd: none }
+          - { os: ubuntu-24.04, simd: sse }
+          - { os: ubuntu-24.04, simd: sse2 }
+          - { os: ubuntu-24.04, simd: sse3 }
+          - { os: ubuntu-24.04, simd: sse4 }
+          - { os: ubuntu-24.04, simd: avx }
+          - { os: ubuntu-24.04, simd: avx2 }
+          - { os: windows-2022, simd: none }
+          - { os: windows-2022, simd: sse }
+          - { os: windows-2022, simd: sse2 }
+          - { os: windows-2022, simd: sse3 }
+          - { os: windows-2022, simd: sse4 }
+          - { os: windows-2022, simd: avx }
+          - { os: windows-2022, simd: avx2 }
+          # ARM64 builds
+          - { os: ubuntu-latest-arm64, simd: neon }
 
     steps:
     - uses: actions/checkout@v4
@@ -415,3 +567,79 @@ jobs:
 
       - name: Build
         run: swift build
+
+  build_cmake_arm:
+    name: CMake / ARM / ${{ matrix.os }} / ${{ matrix.arch }} / ${{ matrix.simd }}
+    runs-on: ${{ matrix.os }}
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          # Linux ARM builds
+          - os: ubuntu-latest-arm64
+            arch: arm64
+            simd: neon
+          - os: ubuntu-latest-arm64
+            arch: armv7
+            simd: neon
+          - os: ubuntu-latest-arm64
+            arch: armv7
+            simd: none
+          # Windows ARM builds
+          - os: windows-latest-arm64
+            arch: arm64
+            simd: neon
+          - os: windows-latest-arm64
+            arch: arm
+            simd: neon
+          - os: windows-latest-arm64
+            arch: arm
+            simd: none
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Configure CMake (Windows)
+      if: runner.os == 'Windows'
+      shell: pwsh
+      run: |
+        $flags = ""
+        if ("${{ matrix.arch }}" -eq "arm") {
+          $flags = "-m32 -march=armv7-a"
+          if ("${{ matrix.simd }}" -eq "neon") {
+            $flags += " -mfpu=neon"
+          }
+        }
+        elseif ("${{ matrix.simd }}" -eq "neon") {
+          $flags = "-march=armv8-a+simd"
+        }
+        
+        cmake -B build -G "Visual Studio 17 2022" -A ${{ matrix.arch == 'arm64' && 'ARM64' || 'ARM' }} `
+          -DCMAKE_BUILD_TYPE=Release `
+          -DCMAKE_C_FLAGS="$flags" `
+          -DCGLM_STATIC=ON -DCGLM_USE_TEST=ON
+
+    - name: Configure CMake (Unix)
+      if: runner.os != 'Windows'
+      shell: bash
+      run: |
+        flags=""
+        if [ "${{ matrix.arch }}" = "armv7" ]; then
+          flags="-m32 -march=armv7-a"
+          if [ "${{ matrix.simd }}" = "neon" ]; then
+            flags="$flags -mfpu=neon -mfloat-abi=hard"
+          fi
+        elif [ "${{ matrix.simd }}" = "neon" ]; then
+          flags="-march=armv8-a+simd"
+        fi
+        
+        cmake -B build -GNinja -DCMAKE_BUILD_TYPE=Release \
+          -DCMAKE_C_FLAGS="$flags" \
+          -DCGLM_STATIC=ON -DCGLM_USE_TEST=ON
+
+    - name: Build
+      run: cmake --build build
+
+    - name: Test
+      working-directory: build
+      run: ./tests

docs/source/opengl.rst

@@ -34,7 +34,7 @@ array of matrices:
    /* ... */
    glUniformMatrix4fv(location, count, GL_FALSE, matrix[0][0]);
 
-1. Cast matrix to pointer
+2. Cast matrix to pointer
 --------------------------
 
 .. code-block:: c

include/cglm/vec4.h

@@ -526,12 +526,8 @@ glm_vec4_div(vec4 a, vec4 b, vec4 dest) {
 CGLM_INLINE
 void
 glm_vec4_divs(vec4 v, float s, vec4 dest) {
-#if defined(__wasm__) && defined(__wasm_simd128__)
+#if defined(CGLM_SIMD)
-  glmm_store(dest, wasm_f32x4_div(glmm_load(v), wasm_f32x4_splat(s)));
+  glmm_store(dest, glmm_div(glmm_load(v), glmm_set1(s)));
-#elif defined( __SSE__ ) || defined( __SSE2__ )
-  glmm_store(dest, _mm_div_ps(glmm_load(v), glmm_set1(s)));
-#elif defined(CGLM_NEON_FP)
-  vst1q_f32(dest, vdivq_f32(vld1q_f32(v), vdupq_n_f32(s)));
 #else
   glm_vec4_scale(v, 1.0f / s, dest);
 #endif
@@ -926,7 +922,7 @@ glm_vec4_normalize_to(vec4 v, vec4 dest) {
     return;
   }
 
-  glmm_store(dest, wasm_f32x4_div(x0, wasm_f32x4_sqrt(xdot)));
+  glmm_store(dest, glmm_div(x0, wasm_f32x4_sqrt(xdot)));
 #elif defined( __SSE__ ) || defined( __SSE2__ )
   __m128 xdot, x0;
   float  dot;
@@ -940,7 +936,7 @@ glm_vec4_normalize_to(vec4 v, vec4 dest) {
     return;
   }
 
-  glmm_store(dest, _mm_div_ps(x0, _mm_sqrt_ps(xdot)));
+  glmm_store(dest, glmm_div(x0, _mm_sqrt_ps(xdot)));
 #else
   float norm;
 

win/cglm-test.vcxproj

@@ -53,6 +53,7 @@
   </ItemGroup>
   <ItemGroup>
     <ClInclude Include="..\test\include\common.h" />
+    <ClInclude Include="..\test\src\test_aabb2d.h" />
     <ClInclude Include="..\test\src\test_affine.h" />
     <ClInclude Include="..\test\src\test_affine2d.h" />
     <ClInclude Include="..\test\src\test_affine_mat.h" />
@@ -76,6 +77,7 @@
     <ClInclude Include="..\test\src\test_mat4.h" />
     <ClInclude Include="..\test\src\test_mat4x2.h" />
     <ClInclude Include="..\test\src\test_mat4x3.h" />
+    <ClInclude Include="..\test\src\test_noise.h" />
     <ClInclude Include="..\test\src\test_plane.h" />
     <ClInclude Include="..\test\src\test_project.h" />
     <ClInclude Include="..\test\src\test_quat.h" />

win/cglm-test.vcxproj.filters

@@ -130,5 +130,11 @@
     <ClInclude Include="..\test\src\test_euler_to_quat_rh.h">
       <Filter>src</Filter>
     </ClInclude>
+    <ClInclude Include="..\test\src\test_aabb2d.h">
+      <Filter>src</Filter>
+    </ClInclude>
+    <ClInclude Include="..\test\src\test_noise.h">
+      <Filter>src</Filter>
+    </ClInclude>
   </ItemGroup>
 </Project>

win/cglm.vcxproj

@@ -80,6 +80,7 @@
     <ClCompile Include="..\src\mat4.c" />
     <ClCompile Include="..\src\mat4x2.c" />
     <ClCompile Include="..\src\mat4x3.c" />
+    <ClCompile Include="..\src\noise.c" />
     <ClCompile Include="..\src\plane.c" />
     <ClCompile Include="..\src\project.c" />
     <ClCompile Include="..\src\quat.c" />
@@ -137,6 +138,7 @@
     <ClInclude Include="..\include\cglm\call\mat4.h" />
     <ClInclude Include="..\include\cglm\call\mat4x2.h" />
     <ClInclude Include="..\include\cglm\call\mat4x3.h" />
+    <ClInclude Include="..\include\cglm\call\noise.h" />
     <ClInclude Include="..\include\cglm\call\plane.h" />
     <ClInclude Include="..\include\cglm\call\project.h" />
     <ClInclude Include="..\include\cglm\call\quat.h" />
@@ -183,6 +185,7 @@
     <ClInclude Include="..\include\cglm\mat4.h" />
     <ClInclude Include="..\include\cglm\mat4x2.h" />
     <ClInclude Include="..\include\cglm\mat4x3.h" />
+    <ClInclude Include="..\include\cglm\noise.h" />
     <ClInclude Include="..\include\cglm\plane.h" />
     <ClInclude Include="..\include\cglm\project.h" />
     <ClInclude Include="..\include\cglm\quat.h" />
@@ -250,6 +253,7 @@
     <ClInclude Include="..\include\cglm\struct\mat4.h" />
     <ClInclude Include="..\include\cglm\struct\mat4x2.h" />
     <ClInclude Include="..\include\cglm\struct\mat4x3.h" />
+    <ClInclude Include="..\include\cglm\struct\noise.h" />
     <ClInclude Include="..\include\cglm\struct\plane.h" />
     <ClInclude Include="..\include\cglm\struct\project.h" />
     <ClInclude Include="..\include\cglm\struct\quat.h" />
@@ -278,7 +282,7 @@
     <ProjectGuid>{CA8BCAF9-CD25-4133-8F62-3D1449B5D2FC}</ProjectGuid>
     <Keyword>Win32Proj</Keyword>
     <RootNamespace>cglm</RootNamespace>
-    <WindowsTargetPlatformVersion>10.0.17763.0</WindowsTargetPlatformVersion>
+    <WindowsTargetPlatformVersion>10.0</WindowsTargetPlatformVersion>
   </PropertyGroup>
   <Import Project="$(VCTargetsPath)\Microsoft.Cpp.Default.props" />
   <PropertyGroup Condition="'$(Configuration)|$(Platform)'=='Debug|Win32'" Label="Configuration">

win/cglm.vcxproj.filters

@@ -184,6 +184,9 @@
     <ClCompile Include="..\src\aabb2d.c">
       <Filter>src</Filter>
     </ClCompile>
+    <ClCompile Include="..\src\noise.c">
+      <Filter>src</Filter>
+    </ClCompile>
   </ItemGroup>
   <ItemGroup>
     <ClInclude Include="..\src\config.h">
@@ -732,5 +735,14 @@
     <ClInclude Include="..\include\cglm\struct\ray.h">
       <Filter>include\cglm\struct</Filter>
     </ClInclude>
+    <ClInclude Include="..\include\cglm\noise.h">
+      <Filter>include\cglm</Filter>
+    </ClInclude>
+    <ClInclude Include="..\include\cglm\call\noise.h">
+      <Filter>include\cglm\call</Filter>
+    </ClInclude>
+    <ClInclude Include="..\include\cglm\struct\noise.h">
+      <Filter>include\cglm\struct</Filter>
+    </ClInclude>
   </ItemGroup>
 </Project>