Align backslashes for great aesthetics (very important)

Redo test scripts
Test and fix _Generic
2025-12-28 15:44:33 +00:00 · 2025-02-24 17:45:35 -06:00 · 2025-02-23 20:33:06 -06:00 · 2025-02-23 20:33:06 -06:00 · 2024-05-28 09:46:45 -05:00 · 2024-05-28 09:46:22 -05:00
12 changed files with 550 additions and 283 deletions
--- a/HandmadeMath.h
+++ b/HandmadeMath.h
@@ -105,10 +105,15 @@
 # define ASSERT_COVERED(a)
 #endif

+#ifdef HANDMADE_MATH_NO_SSE
+# warning "HANDMADE_MATH_NO_SSE is deprecated, use HANDMADE_MATH_NO_SIMD instead"
+# define HANDMADE_MATH_NO_SIMD
+#endif
+
 /* let's figure out if SSE is really available (unless disabled anyway)
   (it isn't on non-x86/x86_64 platforms or even x86 without explicit SSE support)
   => only use "#ifdef HANDMADE_MATH__USE_SSE" to check for SSE support below this block! */
-#ifndef HANDMADE_MATH_NO_SSE
+#ifndef HANDMADE_MATH_NO_SIMD
 # ifdef _MSC_VER /* MSVC supports SSE in amd64 mode or _M_IX86_FP >= 1 (2 means SSE2) */
 #  if defined(_M_AMD64) || ( defined(_M_IX86_FP) && _M_IX86_FP >= 1 )
 #   define HANDMADE_MATH__USE_SSE 1
@@ -118,7 +123,10 @@
 #   define HANDMADE_MATH__USE_SSE 1
 #  endif /*  __SSE__ */
 # endif /* not _MSC_VER */
-#endif /* #ifndef HANDMADE_MATH_NO_SSE */
+# ifdef __ARM_NEON
+#  define HANDMADE_MATH__USE_NEON 1
+# endif /* NEON Supported */
+#endif /* #ifndef HANDMADE_MATH_NO_SIMD */

 #if (!defined(__cplusplus) && defined(__STDC_VERSION__) && __STDC_VERSION__ >= 201112L)
 # define HANDMADE_MATH__USE_C11_GENERICS 1
@@ -128,6 +136,10 @@
 # include <xmmintrin.h>
 #endif

+#ifdef HANDMADE_MATH__USE_NEON
+# include <arm_neon.h>
+#endif
+
 #ifdef _MSC_VER
 #pragma warning(disable:4201)
 #endif
@@ -135,9 +147,7 @@
 #if defined(__GNUC__) || defined(__clang__)
 # pragma GCC diagnostic push
 # pragma GCC diagnostic ignored "-Wfloat-equal"
-# if (defined(__GNUC__) && (__GNUC__ == 4 && __GNUC_MINOR__ < 8)) || defined(__clang__)
-#  pragma GCC diagnostic ignored "-Wmissing-braces"
-# endif
+# pragma GCC diagnostic ignored "-Wmissing-braces"
 # ifdef __clang__
 #  pragma GCC diagnostic ignored "-Wgnu-anonymous-struct"
 #  pragma GCC diagnostic ignored "-Wmissing-field-initializers"
@@ -354,6 +364,10 @@ typedef union HMM_Vec4
    __m128 SSE;
 #endif

+#ifdef HANDMADE_MATH__USE_NEON
+    float32x4_t NEON;
+#endif
+
 #ifdef __cplusplus
    inline float &operator[](int Index) { return Elements[Index]; }
    inline const float &operator[](int Index) const { return Elements[Index]; }
@@ -414,6 +428,9 @@ typedef union HMM_Quat
 #ifdef HANDMADE_MATH__USE_SSE
    __m128 SSE;
 #endif
+#ifdef HANDMADE_MATH__USE_NEON
+    float32x4_t NEON;
+#endif
 } HMM_Quat;

 typedef signed int HMM_Bool;
@@ -503,6 +520,10 @@ static inline float HMM_SqrtF(float Float)
    __m128 In = _mm_set_ss(Float);
    __m128 Out = _mm_sqrt_ss(In);
    Result = _mm_cvtss_f32(Out);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4_t In = vdupq_n_f32(Float);
+    float32x4_t Out = vsqrtq_f32(In);
+    Result = vgetq_lane_f32(Out, 0);
 #else
    Result = HMM_SQRTF(Float);
 #endif
@@ -593,6 +614,9 @@ static inline HMM_Vec4 HMM_V4(float X, float Y, float Z, float W)

 #ifdef HANDMADE_MATH__USE_SSE
    Result.SSE = _mm_setr_ps(X, Y, Z, W);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4_t v = {X, Y, Z, W};
+    Result.NEON = v;
 #else
    Result.X = X;
    Result.Y = Y;
@@ -612,6 +636,9 @@ static inline HMM_Vec4 HMM_V4V(HMM_Vec3 Vector, float W)

 #ifdef HANDMADE_MATH__USE_SSE
    Result.SSE = _mm_setr_ps(Vector.X, Vector.Y, Vector.Z, W);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4_t v = {Vector.X, Vector.Y, Vector.Z, W};
+    Result.NEON = v;
 #else
    Result.XYZ = Vector;
    Result.W = W;
@@ -659,6 +686,8 @@ static inline HMM_Vec4 HMM_AddV4(HMM_Vec4 Left, HMM_Vec4 Right)

 #ifdef HANDMADE_MATH__USE_SSE
    Result.SSE = _mm_add_ps(Left.SSE, Right.SSE);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    Result.NEON = vaddq_f32(Left.NEON, Right.NEON);
 #else
    Result.X = Left.X + Right.X;
    Result.Y = Left.Y + Right.Y;
@@ -703,6 +732,8 @@ static inline HMM_Vec4 HMM_SubV4(HMM_Vec4 Left, HMM_Vec4 Right)

 #ifdef HANDMADE_MATH__USE_SSE
    Result.SSE = _mm_sub_ps(Left.SSE, Right.SSE);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    Result.NEON = vsubq_f32(Left.NEON, Right.NEON);
 #else
    Result.X = Left.X - Right.X;
    Result.Y = Left.Y - Right.Y;
@@ -772,6 +803,8 @@ static inline HMM_Vec4 HMM_MulV4(HMM_Vec4 Left, HMM_Vec4 Right)

 #ifdef HANDMADE_MATH__USE_SSE
    Result.SSE = _mm_mul_ps(Left.SSE, Right.SSE);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    Result.NEON = vmulq_f32(Left.NEON, Right.NEON);
 #else
    Result.X = Left.X * Right.X;
    Result.Y = Left.Y * Right.Y;
@@ -792,6 +825,8 @@ static inline HMM_Vec4 HMM_MulV4F(HMM_Vec4 Left, float Right)
 #ifdef HANDMADE_MATH__USE_SSE
    __m128 Scalar = _mm_set1_ps(Right);
    Result.SSE = _mm_mul_ps(Left.SSE, Scalar);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    Result.NEON = vmulq_n_f32(Left.NEON, Right);
 #else
    Result.X = Left.X * Right;
    Result.Y = Left.Y * Right;
@@ -861,6 +896,8 @@ static inline HMM_Vec4 HMM_DivV4(HMM_Vec4 Left, HMM_Vec4 Right)

 #ifdef HANDMADE_MATH__USE_SSE
    Result.SSE = _mm_div_ps(Left.SSE, Right.SSE);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    Result.NEON = vdivq_f32(Left.NEON, Right.NEON);
 #else
    Result.X = Left.X / Right.X;
    Result.Y = Left.Y / Right.Y;
@@ -881,6 +918,9 @@ static inline HMM_Vec4 HMM_DivV4F(HMM_Vec4 Left, float Right)
 #ifdef HANDMADE_MATH__USE_SSE
    __m128 Scalar = _mm_set1_ps(Right);
    Result.SSE = _mm_div_ps(Left.SSE, Scalar);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4_t Scalar = vdupq_n_f32(Right);
+    Result.NEON = vdivq_f32(Left.NEON, Scalar);
 #else
    Result.X = Left.X / Right;
    Result.Y = Left.Y / Right;
@@ -943,6 +983,11 @@ static inline float HMM_DotV4(HMM_Vec4 Left, HMM_Vec4 Right)
    SSEResultTwo = _mm_shuffle_ps(SSEResultOne, SSEResultOne, _MM_SHUFFLE(0, 1, 2, 3));
    SSEResultOne = _mm_add_ps(SSEResultOne, SSEResultTwo);
    _mm_store_ss(&Result, SSEResultOne);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4_t NEONMultiplyResult = vmulq_f32(Left.NEON, Right.NEON);
+    float32x4_t NEONHalfAdd = vpaddq_f32(NEONMultiplyResult, NEONMultiplyResult);
+    float32x4_t NEONFullAdd = vpaddq_f32(NEONHalfAdd, NEONHalfAdd);
+    Result = vgetq_lane_f32(NEONFullAdd, 0);
 #else
    Result = ((Left.X * Right.X) + (Left.Z * Right.Z)) + ((Left.Y * Right.Y) + (Left.W * Right.W));
 #endif
@@ -1071,6 +1116,11 @@ static inline HMM_Vec4 HMM_LinearCombineV4M4(HMM_Vec4 Left, HMM_Mat4 Right)
    Result.SSE = _mm_add_ps(Result.SSE, _mm_mul_ps(_mm_shuffle_ps(Left.SSE, Left.SSE, 0x55), Right.Columns[1].SSE));
    Result.SSE = _mm_add_ps(Result.SSE, _mm_mul_ps(_mm_shuffle_ps(Left.SSE, Left.SSE, 0xaa), Right.Columns[2].SSE));
    Result.SSE = _mm_add_ps(Result.SSE, _mm_mul_ps(_mm_shuffle_ps(Left.SSE, Left.SSE, 0xff), Right.Columns[3].SSE));
+#elif defined(HANDMADE_MATH__USE_NEON)
+    Result.NEON = vmulq_laneq_f32(Right.Columns[0].NEON, Left.NEON, 0);
+    Result.NEON = vfmaq_laneq_f32(Result.NEON, Right.Columns[1].NEON, Left.NEON, 1);
+    Result.NEON = vfmaq_laneq_f32(Result.NEON, Right.Columns[2].NEON, Left.NEON, 2);
+    Result.NEON = vfmaq_laneq_f32(Result.NEON, Right.Columns[3].NEON, Left.NEON, 3);
 #else
    Result.X = Left.Elements[0] * Right.Columns[0].X;
    Result.Y = Left.Elements[0] * Right.Columns[0].Y;
@@ -1381,10 +1431,10 @@ static inline HMM_Mat3 HMM_MulM3F(HMM_Mat3 Matrix, float Scalar)
    return Result;
 }

-COVERAGE(HMM_DivM3, 1)
+COVERAGE(HMM_DivM3F, 1)
 static inline HMM_Mat3 HMM_DivM3F(HMM_Mat3 Matrix, float Scalar)
 {
-    ASSERT_COVERED(HMM_DivM3);
+    ASSERT_COVERED(HMM_DivM3F);

    HMM_Mat3 Result;

@@ -1465,22 +1515,33 @@ static inline HMM_Mat4 HMM_TransposeM4(HMM_Mat4 Matrix)
 {
    ASSERT_COVERED(HMM_TransposeM4);

-    HMM_Mat4 Result = Matrix;
+    HMM_Mat4 Result;
 #ifdef HANDMADE_MATH__USE_SSE
+    Result = Matrix;
    _MM_TRANSPOSE4_PS(Result.Columns[0].SSE, Result.Columns[1].SSE, Result.Columns[2].SSE, Result.Columns[3].SSE);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4x4_t Transposed = vld4q_f32((float*)Matrix.Columns);
+    Result.Columns[0].NEON = Transposed.val[0];
+    Result.Columns[1].NEON = Transposed.val[1];
+    Result.Columns[2].NEON = Transposed.val[2];
+    Result.Columns[3].NEON = Transposed.val[3];
 #else
+    Result.Elements[0][0] = Matrix.Elements[0][0];
    Result.Elements[0][1] = Matrix.Elements[1][0];
    Result.Elements[0][2] = Matrix.Elements[2][0];
    Result.Elements[0][3] = Matrix.Elements[3][0];
    Result.Elements[1][0] = Matrix.Elements[0][1];
+    Result.Elements[1][1] = Matrix.Elements[1][1];
    Result.Elements[1][2] = Matrix.Elements[2][1];
    Result.Elements[1][3] = Matrix.Elements[3][1];
-    Result.Elements[2][1] = Matrix.Elements[1][2];
    Result.Elements[2][0] = Matrix.Elements[0][2];
+    Result.Elements[2][1] = Matrix.Elements[1][2];
+    Result.Elements[2][2] = Matrix.Elements[2][2];
    Result.Elements[2][3] = Matrix.Elements[3][2];
+    Result.Elements[3][0] = Matrix.Elements[0][3];
    Result.Elements[3][1] = Matrix.Elements[1][3];
    Result.Elements[3][2] = Matrix.Elements[2][3];
-    Result.Elements[3][0] = Matrix.Elements[0][3];
+    Result.Elements[3][3] = Matrix.Elements[3][3];
 #endif

    return Result;
@@ -1493,29 +1554,10 @@ static inline HMM_Mat4 HMM_AddM4(HMM_Mat4 Left, HMM_Mat4 Right)

    HMM_Mat4 Result;

-#ifdef HANDMADE_MATH__USE_SSE
-    Result.Columns[0].SSE = _mm_add_ps(Left.Columns[0].SSE, Right.Columns[0].SSE);
-    Result.Columns[1].SSE = _mm_add_ps(Left.Columns[1].SSE, Right.Columns[1].SSE);
-    Result.Columns[2].SSE = _mm_add_ps(Left.Columns[2].SSE, Right.Columns[2].SSE);
-    Result.Columns[3].SSE = _mm_add_ps(Left.Columns[3].SSE, Right.Columns[3].SSE);
-#else
-    Result.Elements[0][0] = Left.Elements[0][0] + Right.Elements[0][0];
-    Result.Elements[0][1] = Left.Elements[0][1] + Right.Elements[0][1];
-    Result.Elements[0][2] = Left.Elements[0][2] + Right.Elements[0][2];
-    Result.Elements[0][3] = Left.Elements[0][3] + Right.Elements[0][3];
-    Result.Elements[1][0] = Left.Elements[1][0] + Right.Elements[1][0];
-    Result.Elements[1][1] = Left.Elements[1][1] + Right.Elements[1][1];
-    Result.Elements[1][2] = Left.Elements[1][2] + Right.Elements[1][2];
-    Result.Elements[1][3] = Left.Elements[1][3] + Right.Elements[1][3];
-    Result.Elements[2][0] = Left.Elements[2][0] + Right.Elements[2][0];
-    Result.Elements[2][1] = Left.Elements[2][1] + Right.Elements[2][1];
-    Result.Elements[2][2] = Left.Elements[2][2] + Right.Elements[2][2];
-    Result.Elements[2][3] = Left.Elements[2][3] + Right.Elements[2][3];
-    Result.Elements[3][0] = Left.Elements[3][0] + Right.Elements[3][0];
-    Result.Elements[3][1] = Left.Elements[3][1] + Right.Elements[3][1];
-    Result.Elements[3][2] = Left.Elements[3][2] + Right.Elements[3][2];
-    Result.Elements[3][3] = Left.Elements[3][3] + Right.Elements[3][3];
-#endif
+    Result.Columns[0] = HMM_AddV4(Left.Columns[0], Right.Columns[0]);
+    Result.Columns[1] = HMM_AddV4(Left.Columns[1], Right.Columns[1]);
+    Result.Columns[2] = HMM_AddV4(Left.Columns[2], Right.Columns[2]);
+    Result.Columns[3] = HMM_AddV4(Left.Columns[3], Right.Columns[3]);

    return Result;
 }
@@ -1527,29 +1569,10 @@ static inline HMM_Mat4 HMM_SubM4(HMM_Mat4 Left, HMM_Mat4 Right)

    HMM_Mat4 Result;

-#ifdef HANDMADE_MATH__USE_SSE
-    Result.Columns[0].SSE = _mm_sub_ps(Left.Columns[0].SSE, Right.Columns[0].SSE);
-    Result.Columns[1].SSE = _mm_sub_ps(Left.Columns[1].SSE, Right.Columns[1].SSE);
-    Result.Columns[2].SSE = _mm_sub_ps(Left.Columns[2].SSE, Right.Columns[2].SSE);
-    Result.Columns[3].SSE = _mm_sub_ps(Left.Columns[3].SSE, Right.Columns[3].SSE);
-#else
-    Result.Elements[0][0] = Left.Elements[0][0] - Right.Elements[0][0];
-    Result.Elements[0][1] = Left.Elements[0][1] - Right.Elements[0][1];
-    Result.Elements[0][2] = Left.Elements[0][2] - Right.Elements[0][2];
-    Result.Elements[0][3] = Left.Elements[0][3] - Right.Elements[0][3];
-    Result.Elements[1][0] = Left.Elements[1][0] - Right.Elements[1][0];
-    Result.Elements[1][1] = Left.Elements[1][1] - Right.Elements[1][1];
-    Result.Elements[1][2] = Left.Elements[1][2] - Right.Elements[1][2];
-    Result.Elements[1][3] = Left.Elements[1][3] - Right.Elements[1][3];
-    Result.Elements[2][0] = Left.Elements[2][0] - Right.Elements[2][0];
-    Result.Elements[2][1] = Left.Elements[2][1] - Right.Elements[2][1];
-    Result.Elements[2][2] = Left.Elements[2][2] - Right.Elements[2][2];
-    Result.Elements[2][3] = Left.Elements[2][3] - Right.Elements[2][3];
-    Result.Elements[3][0] = Left.Elements[3][0] - Right.Elements[3][0];
-    Result.Elements[3][1] = Left.Elements[3][1] - Right.Elements[3][1];
-    Result.Elements[3][2] = Left.Elements[3][2] - Right.Elements[3][2];
-    Result.Elements[3][3] = Left.Elements[3][3] - Right.Elements[3][3];
-#endif
+    Result.Columns[0] = HMM_SubV4(Left.Columns[0], Right.Columns[0]);
+    Result.Columns[1] = HMM_SubV4(Left.Columns[1], Right.Columns[1]);
+    Result.Columns[2] = HMM_SubV4(Left.Columns[2], Right.Columns[2]);
+    Result.Columns[3] = HMM_SubV4(Left.Columns[3], Right.Columns[3]);

    return Result;
 }
@@ -1575,12 +1598,18 @@ static inline HMM_Mat4 HMM_MulM4F(HMM_Mat4 Matrix, float Scalar)

    HMM_Mat4 Result;

+
 #ifdef HANDMADE_MATH__USE_SSE
    __m128 SSEScalar = _mm_set1_ps(Scalar);
    Result.Columns[0].SSE = _mm_mul_ps(Matrix.Columns[0].SSE, SSEScalar);
    Result.Columns[1].SSE = _mm_mul_ps(Matrix.Columns[1].SSE, SSEScalar);
    Result.Columns[2].SSE = _mm_mul_ps(Matrix.Columns[2].SSE, SSEScalar);
    Result.Columns[3].SSE = _mm_mul_ps(Matrix.Columns[3].SSE, SSEScalar);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    Result.Columns[0].NEON = vmulq_n_f32(Matrix.Columns[0].NEON, Scalar);
+    Result.Columns[1].NEON = vmulq_n_f32(Matrix.Columns[1].NEON, Scalar);
+    Result.Columns[2].NEON = vmulq_n_f32(Matrix.Columns[2].NEON, Scalar);
+    Result.Columns[3].NEON = vmulq_n_f32(Matrix.Columns[3].NEON, Scalar);
 #else
    Result.Elements[0][0] = Matrix.Elements[0][0] * Scalar;
    Result.Elements[0][1] = Matrix.Elements[0][1] * Scalar;
@@ -1623,6 +1652,12 @@ static inline HMM_Mat4 HMM_DivM4F(HMM_Mat4 Matrix, float Scalar)
    Result.Columns[1].SSE = _mm_div_ps(Matrix.Columns[1].SSE, SSEScalar);
    Result.Columns[2].SSE = _mm_div_ps(Matrix.Columns[2].SSE, SSEScalar);
    Result.Columns[3].SSE = _mm_div_ps(Matrix.Columns[3].SSE, SSEScalar);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4_t NEONScalar = vdupq_n_f32(Scalar);
+    Result.Columns[0].NEON = vdivq_f32(Matrix.Columns[0].NEON, NEONScalar);
+    Result.Columns[1].NEON = vdivq_f32(Matrix.Columns[1].NEON, NEONScalar);
+    Result.Columns[2].NEON = vdivq_f32(Matrix.Columns[2].NEON, NEONScalar);
+    Result.Columns[3].NEON = vdivq_f32(Matrix.Columns[3].NEON, NEONScalar);
 #else
    Result.Elements[0][0] = Matrix.Elements[0][0] / Scalar;
    Result.Elements[0][1] = Matrix.Elements[0][1] / Scalar;
@@ -2065,6 +2100,9 @@ static inline HMM_Quat HMM_Q(float X, float Y, float Z, float W)

 #ifdef HANDMADE_MATH__USE_SSE
    Result.SSE = _mm_setr_ps(X, Y, Z, W);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4_t v = { X, Y, Z, W };
+    Result.NEON = v;
 #else
    Result.X = X;
    Result.Y = Y;
@@ -2084,6 +2122,8 @@ static inline HMM_Quat HMM_QV4(HMM_Vec4 Vector)

 #ifdef HANDMADE_MATH__USE_SSE
    Result.SSE = Vector.SSE;
+#elif defined(HANDMADE_MATH__USE_NEON)
+    Result.NEON = Vector.NEON;
 #else
    Result.X = Vector.X;
    Result.Y = Vector.Y;
@@ -2103,6 +2143,8 @@ static inline HMM_Quat HMM_AddQ(HMM_Quat Left, HMM_Quat Right)

 #ifdef HANDMADE_MATH__USE_SSE
    Result.SSE = _mm_add_ps(Left.SSE, Right.SSE);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    Result.NEON = vaddq_f32(Left.NEON, Right.NEON);
 #else

    Result.X = Left.X + Right.X;
@@ -2123,6 +2165,8 @@ static inline HMM_Quat HMM_SubQ(HMM_Quat Left, HMM_Quat Right)

 #ifdef HANDMADE_MATH__USE_SSE
    Result.SSE = _mm_sub_ps(Left.SSE, Right.SSE);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    Result.NEON = vsubq_f32(Left.NEON, Right.NEON);
 #else
    Result.X = Left.X - Right.X;
    Result.Y = Left.Y - Right.Y;
@@ -2156,6 +2200,19 @@ static inline HMM_Quat HMM_MulQ(HMM_Quat Left, HMM_Quat Right)
    SSEResultOne = _mm_shuffle_ps(Left.SSE, Left.SSE, _MM_SHUFFLE(3, 3, 3, 3));
    SSEResultTwo = _mm_shuffle_ps(Right.SSE, Right.SSE, _MM_SHUFFLE(3, 2, 1, 0));
    Result.SSE = _mm_add_ps(SSEResultThree, _mm_mul_ps(SSEResultTwo, SSEResultOne));
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4_t Right1032 = vrev64q_f32(Right.NEON);
+    float32x4_t Right3210 = vcombine_f32(vget_high_f32(Right1032), vget_low_f32(Right1032));
+    float32x4_t Right2301 = vrev64q_f32(Right3210);
+
+    float32x4_t FirstSign = {1.0f, -1.0f, 1.0f, -1.0f};
+    Result.NEON = vmulq_f32(Right3210, vmulq_f32(vdupq_laneq_f32(Left.NEON, 0), FirstSign));
+    float32x4_t SecondSign = {1.0f, 1.0f, -1.0f, -1.0f};
+    Result.NEON = vfmaq_f32(Result.NEON, Right2301, vmulq_f32(vdupq_laneq_f32(Left.NEON, 1), SecondSign));
+    float32x4_t ThirdSign = {-1.0f, 1.0f, 1.0f, -1.0f};
+    Result.NEON = vfmaq_f32(Result.NEON, Right1032, vmulq_f32(vdupq_laneq_f32(Left.NEON, 2), ThirdSign));
+    Result.NEON = vfmaq_laneq_f32(Result.NEON, Right.NEON, Left.NEON, 3);
+
 #else
    Result.X =  Right.Elements[3] * +Left.Elements[0];
    Result.Y =  Right.Elements[2] * -Left.Elements[0];
@@ -2191,6 +2248,8 @@ static inline HMM_Quat HMM_MulQF(HMM_Quat Left, float Multiplicative)
 #ifdef HANDMADE_MATH__USE_SSE
    __m128 Scalar = _mm_set1_ps(Multiplicative);
    Result.SSE = _mm_mul_ps(Left.SSE, Scalar);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    Result.NEON = vmulq_n_f32(Left.NEON, Multiplicative);
 #else
    Result.X = Left.X * Multiplicative;
    Result.Y = Left.Y * Multiplicative;
@@ -2211,6 +2270,9 @@ static inline HMM_Quat HMM_DivQF(HMM_Quat Left, float Divnd)
 #ifdef HANDMADE_MATH__USE_SSE
    __m128 Scalar = _mm_set1_ps(Divnd);
    Result.SSE = _mm_div_ps(Left.SSE, Scalar);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4_t Scalar = vdupq_n_f32(Divnd);
+    Result.NEON = vdivq_f32(Left.NEON, Scalar);
 #else
    Result.X = Left.X / Divnd;
    Result.Y = Left.Y / Divnd;
@@ -2235,6 +2297,11 @@ static inline float HMM_DotQ(HMM_Quat Left, HMM_Quat Right)
    SSEResultTwo = _mm_shuffle_ps(SSEResultOne, SSEResultOne, _MM_SHUFFLE(0, 1, 2, 3));
    SSEResultOne = _mm_add_ps(SSEResultOne, SSEResultTwo);
    _mm_store_ss(&Result, SSEResultOne);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4_t NEONMultiplyResult = vmulq_f32(Left.NEON, Right.NEON);
+    float32x4_t NEONHalfAdd = vpaddq_f32(NEONMultiplyResult, NEONMultiplyResult);
+    float32x4_t NEONFullAdd = vpaddq_f32(NEONHalfAdd, NEONHalfAdd);
+    Result = vgetq_lane_f32(NEONFullAdd, 0);
 #else
    Result = ((Left.X * Right.X) + (Left.Z * Right.Z)) + ((Left.Y * Right.Y) + (Left.W * Right.W));
 #endif
@@ -2278,6 +2345,10 @@ static inline HMM_Quat _HMM_MixQ(HMM_Quat Left, float MixLeft, HMM_Quat Right, f
    __m128 SSEResultOne = _mm_mul_ps(Left.SSE, ScalarLeft);
    __m128 SSEResultTwo = _mm_mul_ps(Right.SSE, ScalarRight);
    Result.SSE = _mm_add_ps(SSEResultOne, SSEResultTwo);
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4_t ScaledLeft = vmulq_n_f32(Left.NEON, MixLeft);
+    float32x4_t ScaledRight = vmulq_n_f32(Right.NEON, MixRight);
+    Result.NEON = vaddq_f32(ScaledLeft, ScaledRight);
 #else
    Result.X = Left.X*MixLeft + Right.X*MixRight;
    Result.Y = Left.Y*MixLeft + Right.Y*MixRight;
@@ -2527,6 +2598,27 @@ static inline HMM_Quat HMM_QFromAxisAngle_LH(HMM_Vec3 Axis, float Angle)
    return HMM_QFromAxisAngle_RH(Axis, -Angle);
 }

+COVERAGE(HMM_QFromNormPair, 1)
+static inline HMM_Quat HMM_QFromNormPair(HMM_Vec3 Left, HMM_Vec3 Right)
+{
+    ASSERT_COVERED(HMM_QFromNormPair);
+
+    HMM_Quat Result;
+
+    Result.XYZ = HMM_Cross(Left, Right);
+    Result.W = 1.0f + HMM_DotV3(Left, Right);
+
+    return HMM_NormQ(Result);
+}
+
+COVERAGE(HMM_QFromVecPair, 1)
+static inline HMM_Quat HMM_QFromVecPair(HMM_Vec3 Left, HMM_Vec3 Right)
+{
+    ASSERT_COVERED(HMM_QFromVecPair);
+
+    return HMM_QFromNormPair(HMM_NormV3(Left), HMM_NormV3(Right));
+}
+
 COVERAGE(HMM_RotateV2, 1)
 static inline HMM_Vec2 HMM_RotateV2(HMM_Vec2 V, float Angle)
 {
@@ -3695,6 +3787,9 @@ static inline HMM_Vec4 operator-(HMM_Vec4 In)
    HMM_Vec4 Result;
 #if HANDMADE_MATH__USE_SSE
    Result.SSE = _mm_xor_ps(In.SSE, _mm_set1_ps(-0.0f));
+#elif defined(HANDMADE_MATH__USE_NEON)
+    float32x4_t Zero = vdupq_n_f32(0.0f);
+    Result.NEON = vsubq_f32(Zero, In.NEON);
 #else
    Result.X = -In.X;
    Result.Y = -In.Y;
@@ -3708,124 +3803,131 @@ static inline HMM_Vec4 operator-(HMM_Vec4 In)
 #endif /* __cplusplus*/

 #ifdef HANDMADE_MATH__USE_C11_GENERICS
+
+void __hmm_invalid_generic();
+
 #define HMM_Add(A, B) _Generic((A), \
-        HMM_Vec2: HMM_AddV2, \
-        HMM_Vec3: HMM_AddV3, \
-        HMM_Vec4: HMM_AddV4, \
-        HMM_Mat2: HMM_AddM2, \
-        HMM_Mat3: HMM_AddM3, \
-        HMM_Mat4: HMM_AddM4, \
-        HMM_Quat: HMM_AddQ \
+    HMM_Vec2: HMM_AddV2, \
+    HMM_Vec3: HMM_AddV3, \
+    HMM_Vec4: HMM_AddV4, \
+    HMM_Mat2: HMM_AddM2, \
+    HMM_Mat3: HMM_AddM3, \
+    HMM_Mat4: HMM_AddM4, \
+    HMM_Quat: HMM_AddQ   \
 )(A, B)

 #define HMM_Sub(A, B) _Generic((A), \
-        HMM_Vec2: HMM_SubV2, \
-        HMM_Vec3: HMM_SubV3, \
-        HMM_Vec4: HMM_SubV4, \
-        HMM_Mat2: HMM_SubM2, \
-        HMM_Mat3: HMM_SubM3, \
-        HMM_Mat4: HMM_SubM4, \
-        HMM_Quat: HMM_SubQ \
+    HMM_Vec2: HMM_SubV2, \
+    HMM_Vec3: HMM_SubV3, \
+    HMM_Vec4: HMM_SubV4, \
+    HMM_Mat2: HMM_SubM2, \
+    HMM_Mat3: HMM_SubM3, \
+    HMM_Mat4: HMM_SubM4, \
+    HMM_Quat: HMM_SubQ   \
 )(A, B)

 #define HMM_Mul(A, B) _Generic((B), \
-     float: _Generic((A), \
+    float: _Generic((A), \
        HMM_Vec2: HMM_MulV2F, \
        HMM_Vec3: HMM_MulV3F, \
        HMM_Vec4: HMM_MulV4F, \
        HMM_Mat2: HMM_MulM2F, \
        HMM_Mat3: HMM_MulM3F, \
        HMM_Mat4: HMM_MulM4F, \
-        HMM_Quat: HMM_MulQF \
-     ), \
-     HMM_Mat2: HMM_MulM2, \
-     HMM_Mat3: HMM_MulM3, \
-     HMM_Mat4: HMM_MulM4, \
-     HMM_Quat: HMM_MulQ, \
-     default: _Generic((A), \
-        HMM_Vec2: HMM_MulV2, \
-        HMM_Vec3: HMM_MulV3, \
-        HMM_Vec4: HMM_MulV4, \
+        HMM_Quat: HMM_MulQF,  \
+        default: __hmm_invalid_generic \
+    ), \
+    HMM_Vec2: _Generic((A), \
+        HMM_Vec2: HMM_MulV2,   \
        HMM_Mat2: HMM_MulM2V2, \
+        default: __hmm_invalid_generic \
+    ), \
+    HMM_Vec3: _Generic((A), \
+        HMM_Vec3: HMM_MulV3,   \
        HMM_Mat3: HMM_MulM3V3, \
-        HMM_Mat4: HMM_MulM4V4 \
-    ) \
+        default: __hmm_invalid_generic \
+    ), \
+    HMM_Vec4: _Generic((A), \
+        HMM_Vec4: HMM_MulV4,   \
+        HMM_Mat4: HMM_MulM4V4, \
+        default: __hmm_invalid_generic \
+    ), \
+    HMM_Mat2: HMM_MulM2, \
+    HMM_Mat3: HMM_MulM3, \
+    HMM_Mat4: HMM_MulM4, \
+    HMM_Quat: HMM_MulQ   \
 )(A, B)

 #define HMM_Div(A, B) _Generic((B), \
-     float: _Generic((A), \
-        HMM_Mat2: HMM_DivM2F, \
-        HMM_Mat3: HMM_DivM3F, \
-        HMM_Mat4: HMM_DivM4F, \
+    float: _Generic((A), \
        HMM_Vec2: HMM_DivV2F, \
        HMM_Vec3: HMM_DivV3F, \
        HMM_Vec4: HMM_DivV4F, \
-        HMM_Quat: HMM_DivQF  \
-     ), \
-     HMM_Mat2: HMM_DivM2, \
-     HMM_Mat3: HMM_DivM3, \
-     HMM_Mat4: HMM_DivM4, \
-     HMM_Quat: HMM_DivQ, \
-     default: _Generic((A), \
-        HMM_Vec2: HMM_DivV2, \
-        HMM_Vec3: HMM_DivV3, \
-        HMM_Vec4: HMM_DivV4  \
-    ) \
+        HMM_Mat2: HMM_DivM2F, \
+        HMM_Mat3: HMM_DivM3F, \
+        HMM_Mat4: HMM_DivM4F, \
+        HMM_Quat: HMM_DivQF   \
+    ), \
+    HMM_Vec2: HMM_DivV2, \
+    HMM_Vec3: HMM_DivV3, \
+    HMM_Vec4: HMM_DivV4  \
 )(A, B)

 #define HMM_Len(A) _Generic((A), \
-        HMM_Vec2: HMM_LenV2, \
-        HMM_Vec3: HMM_LenV3, \
-        HMM_Vec4: HMM_LenV4  \
+    HMM_Vec2: HMM_LenV2, \
+    HMM_Vec3: HMM_LenV3, \
+    HMM_Vec4: HMM_LenV4  \
 )(A)

 #define HMM_LenSqr(A) _Generic((A), \
-        HMM_Vec2: HMM_LenSqrV2, \
-        HMM_Vec3: HMM_LenSqrV3, \
-        HMM_Vec4: HMM_LenSqrV4  \
+    HMM_Vec2: HMM_LenSqrV2, \
+    HMM_Vec3: HMM_LenSqrV3, \
+    HMM_Vec4: HMM_LenSqrV4  \
 )(A)

 #define HMM_Norm(A) _Generic((A), \
-        HMM_Vec2: HMM_NormV2, \
-        HMM_Vec3: HMM_NormV3, \
-        HMM_Vec4: HMM_NormV4  \
+    HMM_Vec2: HMM_NormV2, \
+    HMM_Vec3: HMM_NormV3, \
+    HMM_Vec4: HMM_NormV4, \
+    HMM_Quat: HMM_NormQ   \
 )(A)

 #define HMM_Dot(A, B) _Generic((A), \
-        HMM_Vec2: HMM_DotV2, \
-        HMM_Vec3: HMM_DotV3, \
-        HMM_Vec4: HMM_DotV4  \
+    HMM_Vec2: HMM_DotV2, \
+    HMM_Vec3: HMM_DotV3, \
+    HMM_Vec4: HMM_DotV4, \
+    HMM_Quat: HMM_DotQ   \
 )(A, B)

 #define HMM_Lerp(A, T, B) _Generic((A), \
-        float: HMM_Lerp, \
-        HMM_Vec2: HMM_LerpV2, \
-        HMM_Vec3: HMM_LerpV3, \
-        HMM_Vec4: HMM_LerpV4 \
+    float: HMM_Lerp, \
+    HMM_Vec2: HMM_LerpV2, \
+    HMM_Vec3: HMM_LerpV3, \
+    HMM_Vec4: HMM_LerpV4  \
 )(A, T, B)

 #define HMM_Eq(A, B) _Generic((A), \
-        HMM_Vec2: HMM_EqV2, \
-        HMM_Vec3: HMM_EqV3, \
-        HMM_Vec4: HMM_EqV4  \
+    HMM_Vec2: HMM_EqV2, \
+    HMM_Vec3: HMM_EqV3, \
+    HMM_Vec4: HMM_EqV4  \
 )(A, B)

 #define HMM_Transpose(M) _Generic((M), \
-        HMM_Mat2: HMM_TransposeM2, \
-        HMM_Mat3: HMM_TransposeM3, \
-        HMM_Mat4: HMM_TransposeM4  \
+    HMM_Mat2: HMM_TransposeM2, \
+    HMM_Mat3: HMM_TransposeM3, \
+    HMM_Mat4: HMM_TransposeM4  \
 )(M)

 #define HMM_Determinant(M) _Generic((M), \
-        HMM_Mat2: HMM_DeterminantM2, \
-        HMM_Mat3: HMM_DeterminantM3, \
-        HMM_Mat4: HMM_DeterminantM4  \
+    HMM_Mat2: HMM_DeterminantM2, \
+    HMM_Mat3: HMM_DeterminantM3, \
+    HMM_Mat4: HMM_DeterminantM4  \
 )(M)

 #define HMM_InvGeneral(M) _Generic((M), \
-        HMM_Mat2: HMM_InvGeneralM2, \
-        HMM_Mat3: HMM_InvGeneralM3, \
-        HMM_Mat4: HMM_InvGeneralM4  \
+    HMM_Mat2: HMM_InvGeneralM2, \
+    HMM_Mat3: HMM_InvGeneralM3, \
+    HMM_Mat4: HMM_InvGeneralM4  \
 )(M)

 #endif
@@ -3835,6 +3937,3 @@ static inline HMM_Vec4 operator-(HMM_Vec4 In)
 #endif

 #endif /* HANDMADE_MATH_H */
-
-
-
--- a/test/Makefile
+++ b/test/Makefile
@@ -2,77 +2,123 @@ BUILD_DIR=./build

 CXXFLAGS+=-g -Wall -Wextra -pthread -Wno-missing-braces -Wno-missing-field-initializers -Wfloat-equal

-all: c c_no_sse cpp cpp_no_sse build_c_without_coverage build_cpp_without_coverage
-
-build_all: build_c build_c_no_sse build_cpp build_cpp_no_sse
+.PHONY: all all_c all_cpp
+all: all_c all_cpp
+all_c: c99 c99_no_simd c11 c17
+all_cpp: cpp98 cpp98_no_simd cpp03 cpp11 cpp14 cpp17 cpp20

+.PHONY: clean
 clean:
 	rm -rf $(BUILD_DIR)

-c: build_c
-	$(BUILD_DIR)/hmm_test_c
-
-build_c: HandmadeMath.c test_impl
-	@echo "\nCompiling in C mode"
+.PHONY: c99
+c99:
+	@echo "\nCompiling as C99"
 	mkdir -p $(BUILD_DIR)
 	cd $(BUILD_DIR)\
 		&& $(CC) $(CPPFLAGS) $(CXXFLAGS) -std=c99 \
-			-c ../HandmadeMath.c ../hmm_test.c \
-			-lm \
-		&& $(CC) -ohmm_test_c HandmadeMath.o hmm_test.o -lm
+			../HandmadeMath.c ../hmm_test.c \
+			-lm -o hmm_test_c99 \
+		&& ./hmm_test_c99

-c_no_sse: build_c_no_sse
-	$(BUILD_DIR)/hmm_test_c_no_sse
-
-build_c_no_sse: HandmadeMath.c test_impl
-	@echo "\nCompiling in C mode (no SSE)"
+.PHONY: c99_no_simd
+c99_no_simd:
+	@echo "\nCompiling as C99 (no SIMD)"
 	mkdir -p $(BUILD_DIR)
 	cd $(BUILD_DIR) \
 		&& $(CC) $(CPPFLAGS) $(CXXFLAGS) -std=c99 \
-			-DHANDMADE_MATH_NO_SSE \
-			-c ../HandmadeMath.c ../hmm_test.c \
-			-lm \
-		&& $(CC) -ohmm_test_c_no_sse HandmadeMath.o hmm_test.o -lm
+			-DHANDMADE_MATH_NO_SIMD \
+			../HandmadeMath.c ../hmm_test.c \
+			-lm -o hmm_test_c99_no_simd \
+		&& ./hmm_test_c99_no_simd

-cpp: build_cpp
-	$(BUILD_DIR)/hmm_test_cpp
-
-build_cpp: HandmadeMath.cpp test_impl
-	@echo "\nCompiling in C++ mode"
-	mkdir -p $(BUILD_DIR)
-	cd $(BUILD_DIR) \
-		&& $(CXX) $(CPPFLAGS) $(CXXFLAGS) -ohmm_test_cpp \
-			-DHANDMADE_MATH_CPP_MODE \
-			../HandmadeMath.cpp ../hmm_test.cpp
-
-cpp_no_sse: build_cpp_no_sse
-	$(BUILD_DIR)/hmm_test_cpp_no_sse
-
-build_cpp_no_sse: HandmadeMath.cpp test_impl
-	@echo "\nCompiling in C++ mode (no SSE)"
-	mkdir -p $(BUILD_DIR)
-	cd $(BUILD_DIR) \
-		&& $(CXX) $(CPPFLAGS) $(CXXFLAGS) -ohmm_test_cpp_no_sse \
-			-DHANDMADE_MATH_CPP_MODE -DHANDMADE_MATH_NO_SSE \
-			../HandmadeMath.cpp ../hmm_test.cpp
-
-test_impl: hmm_test.cpp hmm_test.c
-
-build_c_without_coverage: HandmadeMath.c test_impl
-	@echo "\nCompiling in C mode"
+.PHONY: c11
+c11:
+	@echo "\nCompiling as C11"
 	mkdir -p $(BUILD_DIR)
 	cd $(BUILD_DIR)\
-		&& $(CC) $(CPPFLAGS) $(CXXFLAGS) -std=c99 \
-			-DWITHOUT_COVERAGE \
-			-c ../HandmadeMath.c ../hmm_test.c \
-			-lm \
-		&& $(CC) -ohmm_test_c HandmadeMath.o hmm_test.o -lm
+		&& $(CC) $(CPPFLAGS) $(CXXFLAGS) -std=c11 \
+			../HandmadeMath.c ../hmm_test.c \
+			-lm -o hmm_test_c11 \
+		&& ./hmm_test_c11

-build_cpp_without_coverage: HandmadeMath.cpp test_impl
-	@echo "\nCompiling in C++ mode (no SSE)"
+.PHONY: c17
+c17:
+	@echo "\nCompiling as C17"
+	mkdir -p $(BUILD_DIR)
+	cd $(BUILD_DIR)\
+		&& $(CC) $(CPPFLAGS) $(CXXFLAGS) -std=c17 \
+			../HandmadeMath.c ../hmm_test.c \
+			-lm -o hmm_test_c17 \
+		&& ./hmm_test_c17
+
+.PHONY: cpp98
+cpp98:
+	@echo "\nCompiling as C++98"
 	mkdir -p $(BUILD_DIR)
 	cd $(BUILD_DIR) \
-		&& $(CXX) $(CPPFLAGS) $(CXXFLAGS) -ohmm_test_cpp_no_sse \
-			-DHANDMADE_MATH_CPP_MODE -DWITHOUT_COVERAGE \
-			../HandmadeMath.cpp ../hmm_test.cpp
+		&& $(CXX) $(CPPFLAGS) $(CXXFLAGS) -std=c++98 \
+			../HandmadeMath.cpp ../hmm_test.cpp \
+			-lm -o hmm_test_cpp98 \
+		&& ./hmm_test_cpp98

+.PHONY: cpp98_no_simd
+cpp98_no_simd:
+	@echo "\nCompiling as C++98 (no SIMD)"
+	mkdir -p $(BUILD_DIR)
+	cd $(BUILD_DIR) \
+		&& $(CXX) $(CPPFLAGS) $(CXXFLAGS) -std=c++98 \
+			-DHANDMADE_MATH_NO_SIMD \
+			../HandmadeMath.cpp ../hmm_test.cpp \
+			-lm -o hmm_test_cpp98 \
+		&& ./hmm_test_cpp98
+
+.PHONY: cpp03
+cpp03:
+	@echo "\nCompiling as C++03"
+	mkdir -p $(BUILD_DIR)
+	cd $(BUILD_DIR) \
+		&& $(CXX) $(CPPFLAGS) $(CXXFLAGS) -std=c++03 \
+			../HandmadeMath.cpp ../hmm_test.cpp \
+			-lm -o hmm_test_cpp03 \
+		&& ./hmm_test_cpp03
+
+.PHONY: cpp11
+cpp11:
+	@echo "\nCompiling as C++11"
+	mkdir -p $(BUILD_DIR)
+	cd $(BUILD_DIR) \
+		&& $(CXX) $(CPPFLAGS) $(CXXFLAGS) -std=c++11 \
+			../HandmadeMath.cpp ../hmm_test.cpp \
+			-lm -o hmm_test_cpp11 \
+		&& ./hmm_test_cpp11
+
+.PHONY: cpp14
+cpp14:
+	@echo "\nCompiling as C++14"
+	mkdir -p $(BUILD_DIR)
+	cd $(BUILD_DIR) \
+		&& $(CXX) $(CPPFLAGS) $(CXXFLAGS) -std=c++14 \
+			../HandmadeMath.cpp ../hmm_test.cpp \
+			-lm -o hmm_test_cpp14 \
+		&& ./hmm_test_cpp14
+
+.PHONY: cpp17
+cpp17:
+	@echo "\nCompiling as C++17"
+	mkdir -p $(BUILD_DIR)
+	cd $(BUILD_DIR) \
+		&& $(CXX) $(CPPFLAGS) $(CXXFLAGS) -std=c++17 \
+			../HandmadeMath.cpp ../hmm_test.cpp \
+			-lm -o hmm_test_cpp17 \
+		&& ./hmm_test_cpp17
+
+.PHONY: cpp20
+cpp20:
+	@echo "\nCompiling as C++20"
+	mkdir -p $(BUILD_DIR)
+	cd $(BUILD_DIR) \
+		&& $(CXX) $(CPPFLAGS) $(CXXFLAGS) -std=c++20 \
+			../HandmadeMath.cpp ../hmm_test.cpp \
+			-lm -o hmm_test_cpp20 \
+		&& ./hmm_test_cpp20
--- a/test/categories/Addition.h
+++ b/test/categories/Addition.h
@@ -10,12 +10,14 @@ TEST(Addition, Vec2)
        EXPECT_FLOAT_EQ(result.X, 4.0f);
        EXPECT_FLOAT_EQ(result.Y, 6.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec2 result = HMM_Add(v2_1, v2_2);
        EXPECT_FLOAT_EQ(result.X, 4.0f);
        EXPECT_FLOAT_EQ(result.Y, 6.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec2 result = v2_1 + v2_2;
        EXPECT_FLOAT_EQ(result.X, 4.0f);
@@ -39,13 +41,15 @@ TEST(Addition, Vec3)
        EXPECT_FLOAT_EQ(result.Y, 7.0f);
        EXPECT_FLOAT_EQ(result.Z, 9.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec3 result = HMM_Add(v3_1, v3_2);
        EXPECT_FLOAT_EQ(result.X, 5.0f);
        EXPECT_FLOAT_EQ(result.Y, 7.0f);
        EXPECT_FLOAT_EQ(result.Z, 9.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec3 result = v3_1 + v3_2;
        EXPECT_FLOAT_EQ(result.X, 5.0f);
@@ -72,7 +76,7 @@ TEST(Addition, Vec4)
        EXPECT_FLOAT_EQ(result.Z, 10.0f);
        EXPECT_FLOAT_EQ(result.W, 12.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec4 result = HMM_Add(v4_1, v4_2);
        EXPECT_FLOAT_EQ(result.X, 6.0f);
@@ -80,6 +84,8 @@ TEST(Addition, Vec4)
        EXPECT_FLOAT_EQ(result.Z, 10.0f);
        EXPECT_FLOAT_EQ(result.W, 12.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec4 result = v4_1 + v4_2;
        EXPECT_FLOAT_EQ(result.X, 6.0f);
@@ -125,14 +131,16 @@ TEST(Addition, Mat2)
        EXPECT_FLOAT_EQ(result.Elements[1][1], 12.0f); 
    }

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat2 result = HMM_Add(a, b);
-        EXPECT_FLOAT_EQ(result.Elements[0][0], 6.0f); 
-        EXPECT_FLOAT_EQ(result.Elements[0][1], 8.0f); 
-        EXPECT_FLOAT_EQ(result.Elements[1][0], 10.0f); 
-        EXPECT_FLOAT_EQ(result.Elements[1][1], 12.0f); 
+        EXPECT_FLOAT_EQ(result.Elements[0][0], 6.0f);
+        EXPECT_FLOAT_EQ(result.Elements[0][1], 8.0f);
+        EXPECT_FLOAT_EQ(result.Elements[1][0], 10.0f);
+        EXPECT_FLOAT_EQ(result.Elements[1][1], 12.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Mat2 result = a + b;
        EXPECT_FLOAT_EQ(result.Elements[0][0], 6.0f); 
@@ -182,19 +190,21 @@ TEST(Addition, Mat3)
        EXPECT_FLOAT_EQ(result.Elements[2][2], 27.0f);   
    }

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat3 result = HMM_Add(a, b);
-        EXPECT_FLOAT_EQ(result.Elements[0][0], 11.0f); 
-        EXPECT_FLOAT_EQ(result.Elements[0][1], 13.0f); 
-        EXPECT_FLOAT_EQ(result.Elements[0][2], 15.0f); 
-        EXPECT_FLOAT_EQ(result.Elements[1][0], 17.0f); 
+        EXPECT_FLOAT_EQ(result.Elements[0][0], 11.0f);
+        EXPECT_FLOAT_EQ(result.Elements[0][1], 13.0f);
+        EXPECT_FLOAT_EQ(result.Elements[0][2], 15.0f);
+        EXPECT_FLOAT_EQ(result.Elements[1][0], 17.0f);
        EXPECT_FLOAT_EQ(result.Elements[1][1], 19.0f);
-        EXPECT_FLOAT_EQ(result.Elements[1][2], 21.0f); 
-        EXPECT_FLOAT_EQ(result.Elements[2][0], 23.0f); 
+        EXPECT_FLOAT_EQ(result.Elements[1][2], 21.0f);
+        EXPECT_FLOAT_EQ(result.Elements[2][0], 23.0f);
        EXPECT_FLOAT_EQ(result.Elements[2][1], 25.0f);
-        EXPECT_FLOAT_EQ(result.Elements[2][2], 27.0f);   
+        EXPECT_FLOAT_EQ(result.Elements[2][2], 27.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Mat3 result = a + b;
        EXPECT_FLOAT_EQ(result.Elements[0][0], 11.0f); 
@@ -257,7 +267,7 @@ TEST(Addition, Mat4)
            }
        }
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat4 result = HMM_Add(m4_1, m4_2);
        float Expected = 18.0f;
@@ -270,6 +280,8 @@ TEST(Addition, Mat4)
            }
        }
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Mat4 result = m4_1 + m4_2;
        float Expected = 18.0f;
@@ -308,7 +320,7 @@ TEST(Addition, Quaternion)
        EXPECT_FLOAT_EQ(result.Z, 10.0f);
        EXPECT_FLOAT_EQ(result.W, 12.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Quat result = HMM_Add(q1, q2);
        EXPECT_FLOAT_EQ(result.X, 6.0f);
@@ -316,6 +328,8 @@ TEST(Addition, Quaternion)
        EXPECT_FLOAT_EQ(result.Z, 10.0f);
        EXPECT_FLOAT_EQ(result.W, 12.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Quat result = q1 + q2;
        EXPECT_FLOAT_EQ(result.X, 6.0f);
--- a/test/categories/Division.h
+++ b/test/categories/Division.h
@@ -10,12 +10,14 @@ TEST(Division, Vec2Vec2)
        EXPECT_FLOAT_EQ(result.X, 0.5f);
        EXPECT_FLOAT_EQ(result.Y, 0.75f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec2 result = HMM_Div(v2_1, v2_2);
        EXPECT_FLOAT_EQ(result.X, 0.5f);
        EXPECT_FLOAT_EQ(result.Y, 0.75f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec2 result = v2_1 / v2_2;
        EXPECT_FLOAT_EQ(result.X, 0.5f);
@@ -38,12 +40,14 @@ TEST(Division, Vec2Scalar)
        EXPECT_FLOAT_EQ(result.X, 0.5f);
        EXPECT_FLOAT_EQ(result.Y, 1.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec2 result = HMM_Div(v2, s);
        EXPECT_FLOAT_EQ(result.X, 0.5f);
        EXPECT_FLOAT_EQ(result.Y, 1.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec2 result = v2 / s;
        EXPECT_FLOAT_EQ(result.X, 0.5f);
@@ -67,13 +71,15 @@ TEST(Division, Vec3Vec3)
        EXPECT_FLOAT_EQ(result.Y, 0.75f);
        EXPECT_FLOAT_EQ(result.Z, 10.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec3 result = HMM_Div(v3_1, v3_2);
        EXPECT_FLOAT_EQ(result.X, 0.5f);
        EXPECT_FLOAT_EQ(result.Y, 0.75f);
        EXPECT_FLOAT_EQ(result.Z, 10.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec3 result = v3_1 / v3_2;
        EXPECT_FLOAT_EQ(result.X, 0.5f);
@@ -99,13 +105,15 @@ TEST(Division, Vec3Scalar)
        EXPECT_FLOAT_EQ(result.Y, 1.0f);
        EXPECT_FLOAT_EQ(result.Z, 1.5f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec3 result = HMM_Div(v3, s);
        EXPECT_FLOAT_EQ(result.X, 0.5f);
        EXPECT_FLOAT_EQ(result.Y, 1.0f);
        EXPECT_FLOAT_EQ(result.Z, 1.5f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec3 result = v3 / s;
        EXPECT_FLOAT_EQ(result.X, 0.5f);
@@ -132,7 +140,7 @@ TEST(Division, Vec4Vec4)
        EXPECT_FLOAT_EQ(result.Z, 10.0f);
        EXPECT_FLOAT_EQ(result.W, 0.25f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec4 result = HMM_Div(v4_1, v4_2);
        EXPECT_FLOAT_EQ(result.X, 0.5f);
@@ -140,6 +148,8 @@ TEST(Division, Vec4Vec4)
        EXPECT_FLOAT_EQ(result.Z, 10.0f);
        EXPECT_FLOAT_EQ(result.W, 0.25f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec4 result = v4_1 / v4_2;
        EXPECT_FLOAT_EQ(result.X, 0.5f);
@@ -168,7 +178,7 @@ TEST(Division, Vec4Scalar)
        EXPECT_FLOAT_EQ(result.Z, 1.5f);
        EXPECT_FLOAT_EQ(result.W, 2.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec4 result = HMM_Div(v4, s);
        EXPECT_FLOAT_EQ(result.X, 0.5f);
@@ -176,6 +186,8 @@ TEST(Division, Vec4Scalar)
        EXPECT_FLOAT_EQ(result.Z, 1.5f);
        EXPECT_FLOAT_EQ(result.W, 2.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec4 result = v4 / s;
        EXPECT_FLOAT_EQ(result.X, 0.5f);
@@ -212,7 +224,7 @@ TEST(Division, Mat2Scalar)
        EXPECT_FLOAT_EQ(result.Elements[1][1], 8.0f);
    }

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat2 result = HMM_Div(m, s);
        EXPECT_FLOAT_EQ(result.Elements[0][0], 2.0f);
@@ -220,7 +232,9 @@ TEST(Division, Mat2Scalar)
        EXPECT_FLOAT_EQ(result.Elements[1][0], 6.0f);
        EXPECT_FLOAT_EQ(result.Elements[1][1], 8.0f);
    }
+#endif

+#ifdef __cplusplus
    {
        HMM_Mat2 result = m / s;
        EXPECT_FLOAT_EQ(result.Elements[0][0], 2.0f);
@@ -257,7 +271,7 @@ TEST(Division, Mat3Scalar)
        EXPECT_FLOAT_EQ(result.Elements[2][2], 18.0f);
    }

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat3 result = HMM_Div(m, s);
        EXPECT_FLOAT_EQ(result.Elements[0][0], 2.0f);
@@ -270,7 +284,9 @@ TEST(Division, Mat3Scalar)
        EXPECT_FLOAT_EQ(result.Elements[2][1], 16.0f);
        EXPECT_FLOAT_EQ(result.Elements[2][2], 18.0f);
    }
+#endif

+#ifdef __cplusplus
    {
        HMM_Mat3 result = m / s;
        EXPECT_FLOAT_EQ(result.Elements[0][0], 2.0f);
@@ -322,7 +338,7 @@ TEST(Division, Mat4Scalar)
        EXPECT_FLOAT_EQ(result.Elements[3][2], 7.5f);
        EXPECT_FLOAT_EQ(result.Elements[3][3], 8.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat4 result = HMM_Div(m4, s);
        EXPECT_FLOAT_EQ(result.Elements[0][0], 0.5f);
@@ -342,6 +358,8 @@ TEST(Division, Mat4Scalar)
        EXPECT_FLOAT_EQ(result.Elements[3][2], 7.5f);
        EXPECT_FLOAT_EQ(result.Elements[3][3], 8.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Mat4 result = m4 / s;
        EXPECT_FLOAT_EQ(result.Elements[0][0], 0.5f);
@@ -394,7 +412,7 @@ TEST(Division, QuaternionScalar)
        EXPECT_FLOAT_EQ(result.Z, 1.5f);
        EXPECT_FLOAT_EQ(result.W, 2.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Quat result = HMM_Div(q, f);
        EXPECT_FLOAT_EQ(result.X, 0.5f);
@@ -402,6 +420,8 @@ TEST(Division, QuaternionScalar)
        EXPECT_FLOAT_EQ(result.Z, 1.5f);
        EXPECT_FLOAT_EQ(result.W, 2.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Quat result = q / f;
        EXPECT_FLOAT_EQ(result.X, 0.5f);
--- a/test/categories/Equality.h
+++ b/test/categories/Equality.h
@@ -9,10 +9,12 @@ TEST(Equality, Vec2)
    EXPECT_TRUE(HMM_EqV2(a, b));
    EXPECT_FALSE(HMM_EqV2(a, c));

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    EXPECT_TRUE(HMM_Eq(a, b));
    EXPECT_FALSE(HMM_Eq(a, c));
+#endif

+#ifdef __cplusplus
    EXPECT_TRUE(a == b);
    EXPECT_FALSE(a == c);

@@ -30,10 +32,12 @@ TEST(Equality, Vec3)
    EXPECT_TRUE(HMM_EqV3(a, b));
    EXPECT_FALSE(HMM_EqV3(a, c));

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    EXPECT_TRUE(HMM_Eq(a, b));
    EXPECT_FALSE(HMM_Eq(a, c));
+#endif

+#ifdef __cplusplus
    EXPECT_TRUE(a == b);
    EXPECT_FALSE(a == c);

@@ -51,10 +55,12 @@ TEST(Equality, Vec4)
    EXPECT_TRUE(HMM_EqV4(a, b));
    EXPECT_FALSE(HMM_EqV4(a, c));

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    EXPECT_TRUE(HMM_Eq(a, b));
    EXPECT_FALSE(HMM_Eq(a, c));
+#endif

+#ifdef __cplusplus
    EXPECT_TRUE(a == b);
    EXPECT_FALSE(a == c);

--- a/test/categories/MatrixOps.h
+++ b/test/categories/MatrixOps.h
@@ -19,7 +19,7 @@ TEST(InvMatrix, Transpose)
            EXPECT_FLOAT_EQ(result.Elements[1][0], Expect.Elements[1][0]);
            EXPECT_FLOAT_EQ(result.Elements[1][1], Expect.Elements[1][1]);
        }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
        {
            HMM_Mat2 result = HMM_Transpose(Matrix);
            EXPECT_FLOAT_EQ(result.Elements[0][0], Expect.Elements[0][0]);
@@ -54,7 +54,7 @@ TEST(InvMatrix, Transpose)
            EXPECT_FLOAT_EQ(result.Elements[2][1], Expect.Elements[2][1]);
            EXPECT_FLOAT_EQ(result.Elements[2][2], Expect.Elements[2][2]);
        }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
        {
            HMM_Mat3 result = HMM_Transpose(Matrix);
            EXPECT_FLOAT_EQ(result.Elements[0][0], Expect.Elements[0][0]);
@@ -94,7 +94,7 @@ TEST(InvMatrix, Transpose)
            EXPECT_FLOAT_EQ(result.Elements[2][1], Expect.Elements[2][1]);
            EXPECT_FLOAT_EQ(result.Elements[2][2], Expect.Elements[2][2]);
        }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
        {
            HMM_Mat4 result = HMM_Transpose(Matrix);
            EXPECT_FLOAT_EQ(result.Elements[0][0], Expect.Elements[0][0]);
--- a/test/categories/Multiplication.h
+++ b/test/categories/Multiplication.h
@@ -10,12 +10,14 @@ TEST(Multiplication, Vec2Vec2)
        EXPECT_FLOAT_EQ(result.X, 3.0f);
        EXPECT_FLOAT_EQ(result.Y, 8.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec2 result = HMM_Mul(v2_1, v2_2);
        EXPECT_FLOAT_EQ(result.X, 3.0f);
        EXPECT_FLOAT_EQ(result.Y, 8.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec2 result = v2_1 * v2_2;
        EXPECT_FLOAT_EQ(result.X, 3.0f);
@@ -38,12 +40,14 @@ TEST(Multiplication, Vec2Scalar)
        EXPECT_FLOAT_EQ(result.X, 3.0f);
        EXPECT_FLOAT_EQ(result.Y, 6.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec2 result = HMM_Mul(v2, s);
        EXPECT_FLOAT_EQ(result.X, 3.0f);
        EXPECT_FLOAT_EQ(result.Y, 6.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec2 result = v2 * s;
        EXPECT_FLOAT_EQ(result.X, 3.0f);
@@ -72,13 +76,15 @@ TEST(Multiplication, Vec3Vec3)
        EXPECT_FLOAT_EQ(result.Y, 10.0f);
        EXPECT_FLOAT_EQ(result.Z, 18.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec3 result = HMM_Mul(v3_1, v3_2);
        EXPECT_FLOAT_EQ(result.X, 4.0f);
        EXPECT_FLOAT_EQ(result.Y, 10.0f);
        EXPECT_FLOAT_EQ(result.Z, 18.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec3 result = v3_1 * v3_2;
        EXPECT_FLOAT_EQ(result.X, 4.0f);
@@ -104,13 +110,15 @@ TEST(Multiplication, Vec3Scalar)
        EXPECT_FLOAT_EQ(result.Y, 6.0f);
        EXPECT_FLOAT_EQ(result.Z, 9.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec3 result = HMM_Mul(v3, s);
        EXPECT_FLOAT_EQ(result.X, 3.0f);
        EXPECT_FLOAT_EQ(result.Y, 6.0f);
        EXPECT_FLOAT_EQ(result.Z, 9.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec3 result = v3 * s;
        EXPECT_FLOAT_EQ(result.X, 3.0f);
@@ -143,7 +151,7 @@ TEST(Multiplication, Vec4Vec4)
        EXPECT_FLOAT_EQ(result.Z, 21.0f);
        EXPECT_FLOAT_EQ(result.W, 32.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec4 result = HMM_Mul(v4_1, v4_2);
        EXPECT_FLOAT_EQ(result.X, 5.0f);
@@ -151,6 +159,8 @@ TEST(Multiplication, Vec4Vec4)
        EXPECT_FLOAT_EQ(result.Z, 21.0f);
        EXPECT_FLOAT_EQ(result.W, 32.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec4 result = v4_1 * v4_2;
        EXPECT_FLOAT_EQ(result.X, 5.0f);
@@ -179,7 +189,7 @@ TEST(Multiplication, Vec4Scalar)
        EXPECT_FLOAT_EQ(result.Z, 9.0f);
        EXPECT_FLOAT_EQ(result.W, 12.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec4 result = HMM_Mul(v4, s);
        EXPECT_FLOAT_EQ(result.X, 3.0f);
@@ -187,6 +197,8 @@ TEST(Multiplication, Vec4Scalar)
        EXPECT_FLOAT_EQ(result.Z, 9.0f);
        EXPECT_FLOAT_EQ(result.W, 12.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec4 result = v4 * s;
        EXPECT_FLOAT_EQ(result.X, 3.0f);
@@ -234,15 +246,17 @@ TEST(Multiplication, Mat2Mat2) {
        EXPECT_FLOAT_EQ(result.Elements[1][1], 46.0f);        
    }

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat2 result = HMM_Mul(a,b);
        EXPECT_FLOAT_EQ(result.Elements[0][0], 23.0f);
        EXPECT_FLOAT_EQ(result.Elements[0][1], 34.0f);
        EXPECT_FLOAT_EQ(result.Elements[1][0], 31.0f);
-        EXPECT_FLOAT_EQ(result.Elements[1][1], 46.0f);        
+        EXPECT_FLOAT_EQ(result.Elements[1][1], 46.0f);
    }
+#endif

+#ifdef __cplusplus
    {
        HMM_Mat2 result = a * b;
        EXPECT_FLOAT_EQ(result.Elements[0][0], 23.0f);
@@ -271,7 +285,7 @@ TEST(Multiplication, Mat2Scalar) {
        EXPECT_FLOAT_EQ(result.Elements[1][0], 30.0f);
        EXPECT_FLOAT_EQ(result.Elements[1][1], 40.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat2 result = HMM_Mul(m, s);
        EXPECT_FLOAT_EQ(result.Elements[0][0], 10.0f);
@@ -279,6 +293,8 @@ TEST(Multiplication, Mat2Scalar) {
        EXPECT_FLOAT_EQ(result.Elements[1][0], 30.0f);
        EXPECT_FLOAT_EQ(result.Elements[1][1], 40.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Mat2 result = m * s;
        EXPECT_FLOAT_EQ(result.Elements[0][0], 10.0f);
@@ -322,13 +338,15 @@ TEST(Multiplication, Mat2Vec2) {
        EXPECT_FLOAT_EQ(result.Elements[1], 34.0f);
    }

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec2 result = HMM_Mul(m, v);
        EXPECT_FLOAT_EQ(result.Elements[0], 23.0f);
        EXPECT_FLOAT_EQ(result.Elements[1], 34.0f);
    }
-    
+#endif
+
+#ifdef __cplusplus
    {
        HMM_Vec2 result = m * v;
        EXPECT_FLOAT_EQ(result.Elements[0], 23.0f);
@@ -368,7 +386,7 @@ TEST(Multiplication, Mat3Mat3)
        EXPECT_FLOAT_EQ(result.Elements[2][2], 312.0f);
    }

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat3 result = HMM_Mul(a,b);
        EXPECT_FLOAT_EQ(result.Elements[0][0], 138.0f);
@@ -381,7 +399,9 @@ TEST(Multiplication, Mat3Mat3)
        EXPECT_FLOAT_EQ(result.Elements[2][1], 261.0f);
        EXPECT_FLOAT_EQ(result.Elements[2][2], 312.0f);
    }
+#endif

+#ifdef __cplusplus
    {
        HMM_Mat3 result = a * b;
        EXPECT_FLOAT_EQ(result.Elements[0][0], 138.0f);
@@ -420,7 +440,7 @@ TEST(Multiplication, Mat3Scalar) {
        EXPECT_FLOAT_EQ(result.Elements[2][1], 80.0f);
        EXPECT_FLOAT_EQ(result.Elements[2][2], 90.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat3 result = HMM_Mul(m, s);
        EXPECT_FLOAT_EQ(result.Elements[0][0], 10.0f);
@@ -433,6 +453,8 @@ TEST(Multiplication, Mat3Scalar) {
        EXPECT_FLOAT_EQ(result.Elements[2][1], 80.0f);
        EXPECT_FLOAT_EQ(result.Elements[2][2], 90.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Mat3 result = m * s;
        EXPECT_FLOAT_EQ(result.Elements[0][0], 10.0f);
@@ -491,14 +513,15 @@ TEST(Multiplication, Mat3Vec3) {
        EXPECT_FLOAT_EQ(result.Elements[1], 171.0f);
        EXPECT_FLOAT_EQ(result.Elements[2], 204.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec3 result = HMM_Mul(m, v);
        EXPECT_FLOAT_EQ(result.Elements[0], 138.0f);
        EXPECT_FLOAT_EQ(result.Elements[1], 171.0f);
        EXPECT_FLOAT_EQ(result.Elements[2], 204.0f);
    }
-    
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec3 result = m * v;
        EXPECT_FLOAT_EQ(result.Elements[0], 138.0f);
@@ -552,7 +575,7 @@ TEST(Multiplication, Mat4Mat4)
        EXPECT_FLOAT_EQ(result.Elements[3][2], 1118.0f);
        EXPECT_FLOAT_EQ(result.Elements[3][3], 1240.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat4 result = HMM_Mul(m4_1, m4_2);
        EXPECT_FLOAT_EQ(result.Elements[0][0], 538.0f);
@@ -572,6 +595,8 @@ TEST(Multiplication, Mat4Mat4)
        EXPECT_FLOAT_EQ(result.Elements[3][2], 1118.0f);
        EXPECT_FLOAT_EQ(result.Elements[3][3], 1240.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Mat4 result = m4_1 * m4_2;
        EXPECT_FLOAT_EQ(result.Elements[0][0], 538.0f);
@@ -634,7 +659,7 @@ TEST(Multiplication, Mat4Scalar)
        EXPECT_FLOAT_EQ(result.Elements[3][2], 45.0f);
        EXPECT_FLOAT_EQ(result.Elements[3][3], 48.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat4 result = HMM_Mul(m4, s);
        EXPECT_FLOAT_EQ(result.Elements[0][0], 3.0f);
@@ -654,6 +679,8 @@ TEST(Multiplication, Mat4Scalar)
        EXPECT_FLOAT_EQ(result.Elements[3][2], 45.0f);
        EXPECT_FLOAT_EQ(result.Elements[3][3], 48.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Mat4 result = m4 * s;
        EXPECT_FLOAT_EQ(result.Elements[0][0], 3.0f);
@@ -737,7 +764,7 @@ TEST(Multiplication, Mat4Vec4)
        EXPECT_FLOAT_EQ(result.Z, 110.0f);
        EXPECT_FLOAT_EQ(result.W, 120.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec4 result = HMM_Mul(m4, v4);
        EXPECT_FLOAT_EQ(result.X, 90.0f);
@@ -745,6 +772,8 @@ TEST(Multiplication, Mat4Vec4)
        EXPECT_FLOAT_EQ(result.Z, 110.0f);
        EXPECT_FLOAT_EQ(result.W, 120.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec4 result = m4 * v4;
        EXPECT_FLOAT_EQ(result.X, 90.0f);
@@ -769,7 +798,7 @@ TEST(Multiplication, QuaternionQuaternion)
        EXPECT_FLOAT_EQ(result.Z, 48.0f);
        EXPECT_FLOAT_EQ(result.W, -6.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Quat result = HMM_Mul(q1, q2);
        EXPECT_FLOAT_EQ(result.X, 24.0f);
@@ -777,6 +806,8 @@ TEST(Multiplication, QuaternionQuaternion)
        EXPECT_FLOAT_EQ(result.Z, 48.0f);
        EXPECT_FLOAT_EQ(result.W, -6.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Quat result = q1 * q2;
        EXPECT_FLOAT_EQ(result.X, 24.0f);
@@ -803,7 +834,7 @@ TEST(Multiplication, QuaternionScalar)
        EXPECT_FLOAT_EQ(result.Z, 6.0f);
        EXPECT_FLOAT_EQ(result.W, 8.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Quat result = HMM_Mul(q, f);
        EXPECT_FLOAT_EQ(result.X, 2.0f);
@@ -811,6 +842,8 @@ TEST(Multiplication, QuaternionScalar)
        EXPECT_FLOAT_EQ(result.Z, 6.0f);
        EXPECT_FLOAT_EQ(result.W, 8.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Quat result = q * f;
        EXPECT_FLOAT_EQ(result.X, 2.0f);
--- a/test/categories/QuaternionOps.h
+++ b/test/categories/QuaternionOps.h
@@ -22,7 +22,7 @@ TEST(QuaternionOps, Dot)
        float result = HMM_DotQ(q1, q2);
        EXPECT_FLOAT_EQ(result, 70.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        float result = HMM_Dot(q1, q2);
        EXPECT_FLOAT_EQ(result, 70.0f);
@@ -41,7 +41,7 @@ TEST(QuaternionOps, Normalize)
        EXPECT_NEAR(result.Z, 0.5477225575f, 0.001f);
        EXPECT_NEAR(result.W, 0.7302967433f, 0.001f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Quat result = HMM_Norm(q);
        EXPECT_NEAR(result.X, 0.1825741858f, 0.001f);
@@ -97,7 +97,7 @@ TEST(QuaternionOps, SLerp)
        EXPECT_NEAR(result.Z, -0.40824830f, 0.001f);
        EXPECT_NEAR(result.W, 0.70710676f, 0.001f);
    }
-        {
+    {
        HMM_Quat result = HMM_SLerp(from, 1.0f, to);
        EXPECT_NEAR(result.X, 0.5f, 0.001f);
        EXPECT_NEAR(result.Y, 0.5f, 0.001f);
@@ -273,3 +273,25 @@ TEST(QuaternionOps, RotateVectorAxisAngle)
        EXPECT_NEAR(result.Z, 0.707170f, 0.001f);
    }
 }
+
+TEST(QuaternionOps, QuatFromPairs)
+{
+    {
+        HMM_Vec3 n1 = HMM_V3(0.0f, 1.0f, 0.0f);
+        HMM_Vec3 n2 = HMM_V3(0.0f, 0.0f, 1.0f);
+        HMM_Quat q = HMM_QFromNormPair(n1, n2);
+        HMM_Vec3 result = HMM_RotateV3Q(n1, q);
+        EXPECT_NEAR(result.X, n2.X, 0.001f);
+        EXPECT_NEAR(result.Y, n2.Y, 0.001f);
+        EXPECT_NEAR(result.Z, n2.Z, 0.001f);
+    }
+    {
+        HMM_Vec3 v1 = HMM_V3(2.0f, 2.0f, 2.0f);
+        HMM_Vec3 v2 = HMM_V3(3.0f, 0.0f, 0.0f);
+        HMM_Quat q = HMM_QFromVecPair(v1, v2);
+        HMM_Vec3 result = HMM_RotateV3Q(HMM_V3(0.0f, 1.0f, 0.0f), q);
+        EXPECT_NEAR(result.X, 0.577350, 0.001f);
+        EXPECT_NEAR(result.Y, 0.788675, 0.001f);
+        EXPECT_NEAR(result.Z, -0.211325, 0.001f);
+    }
+}
--- a/test/categories/Subtraction.h
+++ b/test/categories/Subtraction.h
@@ -10,12 +10,14 @@ TEST(Subtraction, Vec2)
        EXPECT_FLOAT_EQ(result.X, -2.0f);
        EXPECT_FLOAT_EQ(result.Y, -2.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec2 result = HMM_Sub(v2_1, v2_2);
        EXPECT_FLOAT_EQ(result.X, -2.0f);
        EXPECT_FLOAT_EQ(result.Y, -2.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec2 result = v2_1 - v2_2;
        EXPECT_FLOAT_EQ(result.X, -2.0f);
@@ -39,13 +41,15 @@ TEST(Subtraction, Vec3)
        EXPECT_FLOAT_EQ(result.Y, -3.0f);
        EXPECT_FLOAT_EQ(result.Z, -3.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec3 result = HMM_Sub(v3_1, v3_2);
        EXPECT_FLOAT_EQ(result.X, -3.0f);
        EXPECT_FLOAT_EQ(result.Y, -3.0f);
        EXPECT_FLOAT_EQ(result.Z, -3.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec3 result = v3_1 - v3_2;
        EXPECT_FLOAT_EQ(result.X, -3.0f);
@@ -72,7 +76,7 @@ TEST(Subtraction, Vec4)
        EXPECT_FLOAT_EQ(result.Z, -4.0f);
        EXPECT_FLOAT_EQ(result.W, -4.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec4 result = HMM_Sub(v4_1, v4_2);
        EXPECT_FLOAT_EQ(result.X, -4.0f);
@@ -80,6 +84,8 @@ TEST(Subtraction, Vec4)
        EXPECT_FLOAT_EQ(result.Z, -4.0f);
        EXPECT_FLOAT_EQ(result.W, -4.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Vec4 result = v4_1 - v4_2;
        EXPECT_FLOAT_EQ(result.X, -4.0f);
@@ -124,7 +130,7 @@ TEST(Subtraction, Mat2)
        EXPECT_FLOAT_EQ(result.Elements[1][0], 4.0);
        EXPECT_FLOAT_EQ(result.Elements[1][1], 4.0);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat2 result = HMM_Sub(b,a);
        EXPECT_FLOAT_EQ(result.Elements[0][0], 4.0);
@@ -132,7 +138,8 @@ TEST(Subtraction, Mat2)
        EXPECT_FLOAT_EQ(result.Elements[1][0], 4.0);
        EXPECT_FLOAT_EQ(result.Elements[1][1], 4.0);
    }
-
+#endif
+#ifdef __cplusplus
    {
        HMM_Mat2 result = b - a;
        EXPECT_FLOAT_EQ(result.Elements[0][0], 4.0);
@@ -183,7 +190,7 @@ TEST(Subtraction, Mat3)
        EXPECT_FLOAT_EQ(result.Elements[2][1], 9.0);
        EXPECT_FLOAT_EQ(result.Elements[2][2], 9.0);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat3 result = HMM_Sub(b,a);
        EXPECT_FLOAT_EQ(result.Elements[0][0], 9.0);
@@ -196,7 +203,9 @@ TEST(Subtraction, Mat3)
        EXPECT_FLOAT_EQ(result.Elements[2][1], 9.0);
        EXPECT_FLOAT_EQ(result.Elements[2][2], 9.0);
    }
+#endif

+#ifdef __cplusplus
    b -= a;
    EXPECT_FLOAT_EQ(b.Elements[0][0], 9.0);
    EXPECT_FLOAT_EQ(b.Elements[0][1], 9.0);
@@ -245,7 +254,7 @@ TEST(Subtraction, Mat4)
            }
        }
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Mat4 result = HMM_Sub(m4_1, m4_2);
        for (int Column = 0; Column < 4; ++Column)
@@ -256,6 +265,8 @@ TEST(Subtraction, Mat4)
            }
        }
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Mat4 result = m4_1 - m4_2;
        for (int Column = 0; Column < 4; ++Column)
@@ -290,7 +301,7 @@ TEST(Subtraction, Quaternion)
        EXPECT_FLOAT_EQ(result.Z, -4.0f);
        EXPECT_FLOAT_EQ(result.W, -4.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Quat result = HMM_Sub(q1, q2);
        EXPECT_FLOAT_EQ(result.X, -4.0f);
@@ -298,6 +309,8 @@ TEST(Subtraction, Quaternion)
        EXPECT_FLOAT_EQ(result.Z, -4.0f);
        EXPECT_FLOAT_EQ(result.W, -4.0f);
    }
+#endif
+#ifdef __cplusplus
    {
        HMM_Quat result = q1 - q2;
        EXPECT_FLOAT_EQ(result.X, -4.0f);
--- a/test/categories/VectorOps.h
+++ b/test/categories/VectorOps.h
@@ -10,7 +10,7 @@ TEST(VectorOps, LengthSquared)
    EXPECT_FLOAT_EQ(HMM_LenSqrV3(v3), 14.0f);
    EXPECT_FLOAT_EQ(HMM_LenSqrV4(v4), 15.0f);

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    EXPECT_FLOAT_EQ(HMM_LenSqr(v2), 5.0f);
    EXPECT_FLOAT_EQ(HMM_LenSqr(v3), 14.0f);
    EXPECT_FLOAT_EQ(HMM_LenSqr(v4), 15.0f);
@@ -27,7 +27,7 @@ TEST(VectorOps, Length)
    EXPECT_FLOAT_EQ(HMM_LenV3(v3), 7.0f);
    EXPECT_FLOAT_EQ(HMM_LenV4(v4), 13.892444f);

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    EXPECT_FLOAT_EQ(HMM_Len(v2), 9.0553856f);
    EXPECT_FLOAT_EQ(HMM_Len(v3), 7.0f);
    EXPECT_FLOAT_EQ(HMM_Len(v4), 13.892444f);
@@ -62,7 +62,7 @@ TEST(VectorOps, Normalize)
        EXPECT_LT(result.W, 0.0f);
    }

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec2 result = HMM_Norm(v2);
        EXPECT_NEAR(HMM_LenV2(result), 1.0f, 0.001f);
@@ -112,7 +112,7 @@ TEST(VectorOps, NormalizeZero)
        EXPECT_FLOAT_EQ(result.W, 0.0f);
    }

-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec2 result = HMM_Norm(v2);
        EXPECT_FLOAT_EQ(result.X, 0.0f);
@@ -152,7 +152,7 @@ TEST(VectorOps, DotVec2)
    HMM_Vec2 v2 = HMM_V2(3.0f, 4.0f);

    EXPECT_FLOAT_EQ(HMM_DotV2(v1, v2), 11.0f);
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    EXPECT_FLOAT_EQ(HMM_Dot(v1, v2), 11.0f);
 #endif
 }
@@ -163,7 +163,7 @@ TEST(VectorOps, DotVec3)
    HMM_Vec3 v2 = HMM_V3(4.0f, 5.0f, 6.0f);

    EXPECT_FLOAT_EQ(HMM_DotV3(v1, v2), 32.0f);
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    EXPECT_FLOAT_EQ(HMM_Dot(v1, v2), 32.0f);
 #endif
 }
@@ -174,7 +174,7 @@ TEST(VectorOps, DotVec4)
    HMM_Vec4 v2 = HMM_V4(5.0f, 6.0f, 7.0f, 8.0f);

    EXPECT_FLOAT_EQ(HMM_DotV4(v1, v2), 70.0f);
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    EXPECT_FLOAT_EQ(HMM_Dot(v1, v2), 70.0f);
 #endif
 }
@@ -189,7 +189,7 @@ TEST(VectorOps, LerpV2)
        EXPECT_FLOAT_EQ(result.X, 0.5f);
        EXPECT_FLOAT_EQ(result.Y, 0.5f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec2 result = HMM_Lerp(v1, 0.5, v2);
        EXPECT_FLOAT_EQ(result.X, 0.5f);
@@ -209,7 +209,7 @@ TEST(VectorOps, LerpV3)
        EXPECT_FLOAT_EQ(result.Y, 1.0f);
        EXPECT_FLOAT_EQ(result.Z, 0.5f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec3 result = HMM_Lerp(v1, 0.5, v2);
        EXPECT_FLOAT_EQ(result.X, 0.5f);
@@ -231,7 +231,7 @@ TEST(VectorOps, LerpV4)
        EXPECT_FLOAT_EQ(result.Z, 0.5f);
        EXPECT_FLOAT_EQ(result.W, 1.0f);
    }
-#ifdef __cplusplus
+#if HANDMADE_MATH__USE_C11_GENERICS || defined(__cplusplus)
    {
        HMM_Vec4 result = HMM_Lerp(v1, 0.5, v2);
        EXPECT_FLOAT_EQ(result.X, 0.5f);
--- a/test/run_test_clang.bat
+++ b/test/run_test_clang.bat
@@ -3,16 +3,25 @@
 if not exist "build" mkdir build
 pushd build

-clang-cl /Fehmm_test_c.exe ..\HandmadeMath.c ..\hmm_test.c
-hmm_test_c
+clang-cl /std:c11 /Fehmm_test_c11.exe ..\HandmadeMath.c ..\hmm_test.c || exit /b 1
+hmm_test_c11 || exit /b 1

-clang-cl /Fehmm_test_c_no_sse.exe /DHANDMADE_MATH_NO_SSE ..\HandmadeMath.c ..\hmm_test.c
-hmm_test_c_no_sse
+clang-cl /std:c11 /Fehmm_test_c11_no_simd.exe /DHANDMADE_MATH_NO_SIMD ..\HandmadeMath.c ..\hmm_test.c || exit /b 1
+hmm_test_c11_no_simd || exit /b 1

-clang-cl /Fehmm_test_cpp.exe ..\HandmadeMath.cpp ..\hmm_test.cpp
-hmm_test_cpp
+clang-cl /std:c17 /Fehmm_test_c17.exe ..\HandmadeMath.c ..\hmm_test.c || exit /b 1
+hmm_test_c17 || exit /b 1

-clang-cl /Fehmm_test_cpp_no_sse.exe /DHANDMADE_MATH_NO_SSE ..\HandmadeMath.cpp ..\hmm_test.cpp
-hmm_test_cpp_no_sse
+clang-cl /std:c++14 /Fehmm_test_cpp14.exe ..\HandmadeMath.cpp ..\hmm_test.cpp || exit /b 1
+hmm_test_cpp14 || exit /b 1
+
+clang-cl /std:c++14 /Fehmm_test_cpp14_no_simd.exe /DHANDMADE_MATH_NO_SIMD ..\HandmadeMath.cpp ..\hmm_test.cpp || exit /b 1
+hmm_test_cpp14_no_simd || exit /b 1
+
+clang-cl /std:c++17 /Fehmm_test_cpp17.exe ..\HandmadeMath.cpp ..\hmm_test.cpp || exit /b 1
+hmm_test_cpp17 || exit /b 1
+
+clang-cl /std:c++20 /Fehmm_test_cpp20.exe ..\HandmadeMath.cpp ..\hmm_test.cpp || exit /b 1
+hmm_test_cpp20 || exit /b 1

 popd
--- a/test/run_test_msvc.bat
+++ b/test/run_test_msvc.bat
@@ -1,27 +1,32 @@
@echo off

-if "%1%"=="travis" (
-    call "C:\Program Files (x86)\Microsoft Visual Studio\2017\BuildTools\Common7\Tools\VsDevCmd.bat" -host_arch=amd64 -arch=amd64
-) else (
-    where /q cl
-    if ERRORLEVEL 1 (
-        for /f "delims=" %%a in ('"%ProgramFiles(x86)%\Microsoft Visual Studio\Installer\vswhere.exe" -find VC\Auxiliary\Build\vcvarsall.bat') do (%%a x64)
-    )
+where /q cl
+if ERRORLEVEL 1 (
+    for /f "delims=" %%a in ('"%ProgramFiles(x86)%\Microsoft Visual Studio\Installer\vswhere.exe" -find VC\Auxiliary\Build\vcvarsall.bat') do (%%a x64)
 )

 if not exist "build" mkdir build
 pushd build

-cl /Fehmm_test_c.exe ..\HandmadeMath.c ..\hmm_test.c
-hmm_test_c
+cl /std:c11 /Fehmm_test_c11.exe ..\HandmadeMath.c ..\hmm_test.c || exit /b 1
+hmm_test_c11 || exit /b 1

-cl /Fehmm_test_c_no_sse.exe /DHANDMADE_MATH_NO_SSE ..\HandmadeMath.c ..\hmm_test.c
-hmm_test_c_no_sse
+cl /std:c11 /Fehmm_test_c11_no_simd.exe /DHANDMADE_MATH_NO_SIMD ..\HandmadeMath.c ..\hmm_test.c || exit /b 1
+hmm_test_c11_no_simd || exit /b 1

-cl /Fehmm_test_cpp.exe ..\HandmadeMath.cpp ..\hmm_test.cpp
-hmm_test_cpp
+cl /std:c17 /Fehmm_test_c17.exe ..\HandmadeMath.c ..\hmm_test.c || exit /b 1
+hmm_test_c17 || exit /b 1

-cl /Fehmm_test_cpp_no_sse.exe /DHANDMADE_MATH_NO_SSE ..\HandmadeMath.cpp ..\hmm_test.cpp
-hmm_test_cpp_no_sse
+cl /std:c++14 /Fehmm_test_cpp14.exe ..\HandmadeMath.cpp ..\hmm_test.cpp || exit /b 1
+hmm_test_cpp14 || exit /b 1
+
+cl /std:c++14 /Fehmm_test_cpp14_no_simd.exe /DHANDMADE_MATH_NO_SIMD ..\HandmadeMath.cpp ..\hmm_test.cpp || exit /b 1
+hmm_test_cpp14_no_simd || exit /b 1
+
+cl /std:c++17 /Fehmm_test_cpp17.exe ..\HandmadeMath.cpp ..\hmm_test.cpp || exit /b 1
+hmm_test_cpp17 || exit /b 1
+
+cl /std:c++20 /Fehmm_test_cpp20.exe ..\HandmadeMath.cpp ..\hmm_test.cpp || exit /b 1
+hmm_test_cpp20 || exit /b 1

 popd
Author	SHA1	Message	Date
Ben Visness	142ba3cd9d	Align backslashes for great aesthetics (very important)	2025-02-24 17:45:35 -06:00
Ben Visness	8c2ac269ba	Redo test scripts	2025-02-23 20:33:06 -06:00
Ben Visness	736ebaaf23	Test and fix _Generic	2025-02-23 20:33:06 -06:00
Ben Visness	bdc7dd2a51	Merge pull request #172 from jonasgf/whitespace Remove trailing whitespace and extra newlines at end of file	2024-05-28 09:46:45 -05:00
Ben Visness	b4ce44823b	Merge pull request #171 from jonasgf/missing-braces Ignore missing braces warning for all GCC versions	2024-05-28 09:46:22 -05:00
Jonas	2901e70ca3	Remove trailing whitespace and extra newlines at end of file	2024-04-27 22:56:11 +02:00
Jonas	21b4a0d209	Ignore missing braces warning for all GCC versions	2024-04-27 22:46:15 +02:00
Ben Visness	43836aa3a6	Merge pull request #170 from B-Y-P/master Add HMM_QFromNormPair() and HMM_QFromVecPair()	2024-04-11 11:08:32 -05:00
BYP	bdc04881de	Add HMM_QFromNormPair() and HMM_QFromVecPair()	2024-04-08 21:21:54 -04:00
Ben Visness	322826bcee	Merge pull request #169 from Jack-Punter/neon_support Neon support	2023-12-15 20:14:20 -06:00
jack punter	4e1e97522d	Address Ben's changes	2023-12-14 21:39:50 +00:00
Jack Punter	6603c796cc	Update test Makefile to use NO_SIMD over NO_SSE	2023-12-12 01:53:43 +00:00
Jack Punter	623215b228	NEON Quaternion Multiply	2023-12-12 01:47:01 +00:00
Jack Punter	fcc510f767	Fix scalar	2023-12-12 01:42:37 +00:00
Jack Punter	133a595b6f	Better Neon tranpose vld can de-interleave up to 4-float element structures in one go into multiple registers. This allows us to load each column of the matrix and put each element into a corresponding vector, essentially doing the entire transpose in 1 instruction Fixed it	2023-12-12 01:42:37 +00:00
Jack Punter	5f20d693c9	Add Neon implementations for all but Quaternion multiply	2023-12-11 22:57:19 +00:00