diff --git a/CMakeLists.txt b/CMakeLists.txt
index 0e10b2c..169296d 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -1,6 +1,6 @@
 cmake_minimum_required(VERSION 3.8.2)
 project(cglm
-  VERSION 0.9.4
+  VERSION 0.9.5
   HOMEPAGE_URL https://github.com/recp/cglm
   DESCRIPTION "OpenGL Mathematics (glm) for C"
   LANGUAGES C
diff --git a/cglm.podspec b/cglm.podspec
index 52d6122..85603b7 100644
--- a/cglm.podspec
+++ b/cglm.podspec
@@ -2,7 +2,7 @@ Pod::Spec.new do |s|
 
   # Description
   s.name         = "cglm"
-  s.version      = "0.9.3"
+  s.version      = "0.9.4"
   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.
diff --git a/configure.ac b/configure.ac
index 399aadb..e3daae9 100644
--- a/configure.ac
+++ b/configure.ac
@@ -7,7 +7,7 @@
 #*****************************************************************************
 
 AC_PREREQ([2.69])
-AC_INIT([cglm], [0.9.4], [info@recp.me])
+AC_INIT([cglm], [0.9.5], [info@recp.me])
 AM_INIT_AUTOMAKE([-Wall foreign subdir-objects serial-tests])
 
 # Don't use the default cflags (-O2 -g), we set ours manually in Makefile.am.
diff --git a/docs/source/conf.py b/docs/source/conf.py
index bf1251e..0f5df86 100644
--- a/docs/source/conf.py
+++ b/docs/source/conf.py
@@ -62,9 +62,9 @@ author = u'Recep Aslantas'
 # built documents.
 #
 # The short X.Y version.
-version = u'0.9.4'
+version = u'0.9.5'
 # The full version, including alpha/beta/rc tags.
-release = u'0.9.4'
+release = u'0.9.5'
 
 # The language for content autogenerated by Sphinx. Refer to documentation
 # for a list of supported languages.
@@ -91,6 +91,7 @@ todo_include_todos = False
 # a list of builtin themes.
 #
 html_theme = 'sphinx_rtd_theme'
+pygments_style = 'monokai'
 
 # Theme options are theme-specific and customize the look and feel of a theme
 # further.  For a list of options available for each theme, see the
@@ -111,8 +112,11 @@ 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 = ['sphinx-static']
 
+# Add customm CSS and JS files
+html_css_files = ['theme_overrides.css']
+html_js_files = []
 
 # -- Options for HTMLHelp output ------------------------------------------
 
diff --git a/docs/source/mat2x3.rst b/docs/source/mat2x3.rst
index 433c637..06780d6 100644
--- a/docs/source/mat2x3.rst
+++ b/docs/source/mat2x3.rst
@@ -23,6 +23,16 @@ Functions:
 #. :c:func:`glm_mat2x3_transpose`
 #. :c:func:`glm_mat2x3_scale`
 
+Represented
+~~~~~~~~~~~
+
+.. csv-table:: mat2x3
+   :header: "", "column 1", "column 2"
+
+   "row 1", "m00", "m10"
+   "row 2", "m01", "m11"
+   "row 3", "m02", "m12"
+
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -51,28 +61,68 @@ Functions documentation
       | *[in]*  **src**  pointer to an array of floats
       | *[out]* **dest** destination matrix2x3
 
-.. c:function:: void glm_mat2x3_mul(mat2x3 m1, mat3x2 m2, mat2 dest)
+.. c:function:: void glm_mat2x3_mul(mat2x3 m1, mat3x2 m2, mat3 dest)
 
     multiply m1 and m2 to dest
 
     .. code-block:: c
 
-       glm_mat2x3_mul(mat2x3, mat3x2, mat2);
+       glm_mat2x3_mul(mat2x3, mat3x2, mat3);
 
     Parameters:
       | *[in]*  **m1**    left matrix (mat2x3)
       | *[in]*  **m2**    right matrix (mat3x2)
-      | *[out]* **dest**  destination matrix (mat2)
+      | *[out]* **dest**  destination matrix (mat3)
 
-.. c:function:: void glm_mat2x3_mulv(mat2x3 m, vec3 v, vec2 dest)
+    .. csv-table:: mat2x3
+        :header: "", "column 1", "column 2"
 
-    multiply mat2x3 with vec3 (column vector) and store in dest vector
+        "row 1", "a00", "a10"
+        "row 2", "a01", "a11"
+        "row 3", "a02", "a12"
+
+    .. csv-table:: mat3x2
+        :header: "", "column 1", "column 2", "column 3"
+
+        "row 1", "b00", "b10", "b20"
+        "row 2", "b01", "b11", "b21"
+
+    .. csv-table:: mat3x3
+        :header: "", "column 1", "column 2", "column 3"
+
+        "row 1", "a00 * b00 + a10 * b01", "a00 * b10 + a10 * b11", "a00 * b20 + a10 * b21"
+        "row 2", "a01 * b00 + a11 * b01", "a01 * b10 + a11 * b11", "a01 * b20 + a11 * b21"
+        "row 3", "a02 * b00 + a12 * b01", "a02 * b10 + a12 * b11", "a02 * b20 + a12 * b21"
+
+.. c:function:: void glm_mat2x3_mulv(mat2x3 m, vec2 v, vec3 dest)
+
+    multiply mat2x3 with vec2 (column vector) and store in dest column vector
 
     Parameters:
       | *[in]*  **m**     mat2x3 (left)
       | *[in]*  **v**     vec3 (right, column vector)
       | *[out]* **dest**  destination (result, column vector)
 
+    .. csv-table:: mat2x3
+        :header: "", "column 1", "column 2"
+
+        "row 1", "m00", "m10"
+        "row 2", "m01", "m11"
+        "row 3", "m02", "m12"
+
+    .. csv-table:: column vec2 (1x2)
+        :header: "", "column 1"
+
+        "row 1", "v0"
+        "row 2", "v1"
+
+    .. csv-table:: column vec3 (1x3)
+        :header: "", "column 1"
+
+        "row 1", "m00 * v0 + m10 * v1"
+        "row 2", "m01 * v0 + m11 * v1"
+        "row 3", "m02 * v0 + m12 * v1"
+
 .. c:function:: void glm_mat2x3_transpose(mat2x3 m, mat3x2 dest)
 
     transpose matrix and store in dest
diff --git a/docs/source/mat2x4.rst b/docs/source/mat2x4.rst
index 9a0445f..0acc643 100644
--- a/docs/source/mat2x4.rst
+++ b/docs/source/mat2x4.rst
@@ -23,6 +23,17 @@ Functions:
 #. :c:func:`glm_mat2x4_transpose`
 #. :c:func:`glm_mat2x4_scale`
 
+Represented
+~~~~~~~~~~~
+
+.. csv-table:: mat2x4
+   :header: "", "column 1", "column 2"
+
+   "row 1", "m00", "m10"
+   "row 2", "m01", "m11"
+   "row 3", "m02", "m12"
+   "row 4", "m03", "m13"
+
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -51,28 +62,72 @@ Functions documentation
       | *[in]*  **src**  pointer to an array of floats
       | *[out]* **dest** destination matrix2x4
 
-.. c:function:: void glm_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat2 dest)
+.. c:function:: void glm_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat4 dest)
 
     multiply m1 and m2 to dest
 
     .. code-block:: c
 
-       glm_mat2x4_mul(mat2x4, mat4x2, mat2);
+       glm_mat2x4_mul(mat2x4, mat4x2, mat4);
 
     Parameters:
       | *[in]*  **m1**    left matrix (mat2x4)
       | *[in]*  **m2**    right matrix (mat4x2)
-      | *[out]* **dest**  destination matrix (mat2)
+      | *[out]* **dest**  destination matrix (mat4)
 
-.. c:function:: void glm_mat2x4_mulv(mat2x4 m, vec4 v, vec2 dest)
+    .. csv-table:: mat2x4
+        :header: "", "column 1", "column 2"
 
-    multiply mat2x4 with vec4 (column vector) and store in dest vector
+        "row 1", "a00", "a10"
+        "row 2", "a01", "a11"
+        "row 3", "a02", "a12"
+        "row 4", "a03", "a13"
+
+    .. csv-table:: mat4x2
+        :header: "", "column 1", "column 2", "column 3", "column 4"
+
+        "row 1", "b00", "b10", "b20", "b30"
+        "row 2", "b01", "b11", "b21", "b31"
+
+    .. csv-table:: mat4x4
+        :header: "", "column 1", "column 2", "column 3", "column 4"
+
+        "row 1", "a00 * b00 + a10 * b01", "a00 * b10 + a10 * b11", "a00 * b20 + a10 * b21", "a00 * b30 + a10 * b31"
+        "row 2", "a01 * b00 + a11 * b01", "a01 * b10 + a11 * b11", "a01 * b20 + a11 * b21", "a01 * b30 + a11 * b31"
+        "row 3", "a02 * b00 + a12 * b01", "a02 * b10 + a12 * b11", "a02 * b20 + a12 * b21", "a02 * b30 + a12 * b31"
+        "row 4", "a03 * b00 + a13 * b01", "a03 * b10 + a13 * b11", "a03 * b20 + a13 * b21", "a03 * b30 + a13 * b31"
+
+.. c:function:: void glm_mat2x4_mulv(mat2x4 m, vec2 v, vec4 dest)
+
+    multiply mat2x4 with vec2 (column vector) and store in dest column vector
 
     Parameters:
       | *[in]*  **m**     mat2x4 (left)
-      | *[in]*  **v**     vec4 (right, column vector)
+      | *[in]*  **v**     vec2 (right, column vector)
       | *[out]* **dest**  destination (result, column vector)
 
+    .. csv-table:: mat2x4
+        :header: "", "column 1", "column 2"
+
+        "row 1", "m00", "m10"
+        "row 2", "m01", "m11"
+        "row 3", "m02", "m12"
+        "row 4", "m03", "m13"
+
+    .. csv-table:: column vec2 (1x2)
+        :header: "", "column 1"
+
+        "row 1", "v0"
+        "row 2", "v1"
+
+    .. csv-table:: column vec4 (1x4)
+        :header: "", "column 1"
+
+        "row 1", "m00 * v0 + m10 * v1"
+        "row 2", "m01 * v0 + m11 * v1"
+        "row 3", "m02 * v0 + m12 * v1"
+        "row 4", "m03 * v0 + m13 * v1"
+
 .. c:function:: void glm_mat2x4_transpose(mat2x4 m, mat4x2 dest)
 
     transpose matrix and store in dest
diff --git a/docs/source/mat3x2.rst b/docs/source/mat3x2.rst
index 7290153..7604728 100644
--- a/docs/source/mat3x2.rst
+++ b/docs/source/mat3x2.rst
@@ -23,6 +23,15 @@ Functions:
 #. :c:func:`glm_mat3x2_transpose`
 #. :c:func:`glm_mat3x2_scale`
 
+Represented
+~~~~~~~~~~~
+
+.. csv-table:: mat3x2
+   :header: "", "column 1", "column 2", "column 3"
+
+   "row 1", "m00", "m10", "m20"
+   "row 2", "m01", "m11", "m21"
+
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -51,28 +60,66 @@ Functions documentation
       | *[in]*  **src**  pointer to an array of floats
       | *[out]* **dest** destination matrix3x2
 
-.. c:function:: void glm_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat3 dest)
+.. c:function:: void glm_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat2 dest)
 
     multiply m1 and m2 to dest
 
     .. code-block:: c
 
-       glm_mat3x2_mul(mat3x2, mat2x3, mat3);
+       glm_mat3x2_mul(mat3x2, mat2x3, mat2);
 
     Parameters:
       | *[in]*  **m1**    left matrix (mat3x2)
       | *[in]*  **m2**    right matrix (mat2x3)
-      | *[out]* **dest**  destination matrix (mat3)
+      | *[out]* **dest**  destination matrix (mat2)
 
-.. c:function:: void glm_mat3x2_mulv(mat3x2 m, vec2 v, vec3 dest)
+    .. csv-table:: mat3x2
+        :header: "", "column 1", "column 2", "column 3"
 
-    multiply mat3x2 with vec2 (column vector) and store in dest vector
+        "row 1", "a00", "a10", "a20"
+        "row 2", "a01", "a11", "a21"
+
+    .. csv-table:: mat2x3
+        :header: "", "column 1", "column 2"
+
+        "row 1", "b00", "b10"
+        "row 2", "b01", "b11"
+        "row 3", "b02", "b12"
+
+    .. csv-table:: mat2x2
+        :header: "", "column 1", "column 2"
+
+        "row 1", "a00 * b00 + a10 * b01 + a20 * b02", "a00 * b10 + a10 * b11 + a20 * b12"
+        "row 2", "a01 * b00 + a11 * b01 + a21 * b02", "a01 * b10 + a11 * b11 + a21 * b12"
+
+.. c:function:: void glm_mat3x2_mulv(mat3x2 m, vec3 v, vec2 dest)
+
+    multiply mat3x2 with vec3 (column vector) and store in dest vector
 
     Parameters:
       | *[in]*  **m**     mat3x2 (left)
       | *[in]*  **v**     vec3 (right, column vector)
       | *[out]* **dest**  destination (result, column vector)
 
+    .. csv-table:: mat3x2
+        :header: "", "column 1", "column 2", "column 3"
+
+        "row 1", "m00", "m10", "m20"
+        "row 2", "m01", "m11", "m21"
+
+    .. csv-table:: column vec3 (1x3)
+        :header: "", "column 1"
+
+        "row 1", "v0"
+        "row 2", "v1"
+        "row 3", "v2"
+
+    .. csv-table:: column vec2 (1x2)
+        :header: "", "column 1"
+
+        "row 1", "m00 * v0 + m10 * v1 + m20 * v2"
+        "row 2", "m01 * v0 + m11 * v1 + m21 * v2"
+
 .. c:function:: void glm_mat3x2_transpose(mat3x2 m, mat2x3 dest)
 
     transpose matrix and store in dest
diff --git a/docs/source/mat3x4.rst b/docs/source/mat3x4.rst
index 9c23d06..cddce47 100644
--- a/docs/source/mat3x4.rst
+++ b/docs/source/mat3x4.rst
@@ -23,6 +23,17 @@ Functions:
 #. :c:func:`glm_mat3x4_transpose`
 #. :c:func:`glm_mat3x4_scale`
 
+Represented
+~~~~~~~~~~~
+
+.. csv-table:: mat3x4
+   :header: "", "column 1", "column 2", "column 3"
+
+   "row 1", "m00", "m10", "m20"
+   "row 2", "m01", "m11", "m21"
+   "row 3", "m02", "m12", "m22"
+   "row 4", "m03", "m13", "m23"
+
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -51,28 +62,74 @@ Functions documentation
       | *[in]*  **src**  pointer to an array of floats
       | *[out]* **dest** destination matrix3x4
 
-.. c:function:: void glm_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat3 dest)
+.. c:function:: void glm_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat4 dest)
 
     multiply m1 and m2 to dest
 
     .. code-block:: c
 
-       glm_mat3x4_mul(mat3x4, mat4x3, mat3);
+       glm_mat3x4_mul(mat3x4, mat4x3, mat4);
 
     Parameters:
       | *[in]*  **m1**    left matrix (mat3x4)
       | *[in]*  **m2**    right matrix (mat4x3)
-      | *[out]* **dest**  destination matrix (mat3)
+      | *[out]* **dest**  destination matrix (mat4)
 
-.. c:function:: void glm_mat3x4_mulv(mat3x4 m, vec4 v, vec3 dest)
+    .. csv-table:: mat3x4
+        :header: "", "column 1", "column 2", "column 3"
 
-    multiply mat3x4 with vec4 (column vector) and store in dest vector
+        "row 1", "a00", "a10", "a20"
+        "row 2", "a01", "a11", "a21"
+        "row 3", "a02", "a12", "a22"
+        "row 4", "a03", "a13", "a23"
+
+    .. csv-table:: mat4x3
+        :header: "", "column 1", "column 2", "column 3", "column 4"
+
+        "row 1", "b00", "b10", "b20", "b30"
+        "row 2", "b01", "b11", "b21", "b31"
+        "row 3", "b02", "b12", "b22", "b32"
+
+    .. csv-table:: mat4x4
+        :header: "", "column 1", "column 2", "column 3", "column 4"
+
+        "row 1", "a00 * b00 + a10 * b01 + a20 * b02", "a00 * b10 + a10 * b11 + a20 * b12", "a00 * b20 + a10 * b21 + a20 * b22", "a00 * b30 + a10 * b31 + a20 * b32"
+        "row 2", "a01 * b00 + a11 * b01 + a21 * b02", "a01 * b10 + a11 * b11 + a21 * b12", "a01 * b20 + a11 * b21 + a21 * b22", "a01 * b30 + a11 * b31 + a21 * b32"
+        "row 3", "a02 * b00 + a12 * b01 + a22 * b02", "a02 * b10 + a12 * b11 + a22 * b12", "a02 * b20 + a12 * b21 + a22 * b22", "a02 * b30 + a12 * b31 + a22 * b32"
+        "row 4", "a03 * b00 + a13 * b01 + a23 * b02", "a03 * b10 + a13 * b11 + a23 * b12", "a03 * b20 + a13 * b21 + a23 * b22", "a03 * b30 + a13 * b31 + a23 * b32"
+
+.. c:function:: void glm_mat3x4_mulv(mat3x4 m, vec3 v, vec4 dest)
+
+    multiply mat3x4 with vec3 (column vector) and store in dest vector
 
     Parameters:
       | *[in]*  **m**     mat3x4 (left)
-      | *[in]*  **v**     vec4 (right, column vector)
+      | *[in]*  **v**     vec3 (right, column vector)
       | *[out]* **dest**  destination (result, column vector)
 
+    .. csv-table:: mat3x4
+        :header: "", "column 1", "column 2", "column 3"
+
+        "row 1", "m00", "m10", "m20"
+        "row 2", "m01", "m11", "m21"
+        "row 3", "m02", "m12", "m22"
+        "row 4", "m03", "m13", "m23"
+
+    .. csv-table:: column vec3 (1x3)
+        :header: "", "column 1"
+
+        "row 1", "v0"
+        "row 2", "v1"
+        "row 3", "v2"
+
+    .. csv-table:: column vec4 (1x4)
+        :header: "", "column 1"
+
+        "row 1", "m00 * v0 + m10 * v1 + m20 * v2"
+        "row 2", "m01 * v0 + m11 * v1 + m21 * v2"
+        "row 3", "m02 * v0 + m12 * v1 + m22 * v2"
+        "row 4", "m03 * v0 + m13 * v1 + m23 * v2"
+
 .. c:function:: void glm_mat3x4_transpose(mat3x4 m, mat4x3 dest)
 
     transpose matrix and store in dest
diff --git a/docs/source/mat4x2.rst b/docs/source/mat4x2.rst
index 4b7f291..35ab2b8 100644
--- a/docs/source/mat4x2.rst
+++ b/docs/source/mat4x2.rst
@@ -23,6 +23,15 @@ Functions:
 #. :c:func:`glm_mat4x2_transpose`
 #. :c:func:`glm_mat4x2_scale`
 
+Represented
+~~~~~~~~~~~
+
+.. csv-table:: mat4x2
+   :header: "", "column 1", "column 2", "column 3", "column4"
+
+   "row 1", "m00", "m10", "m20", "m30"
+   "row 2", "m01", "m11", "m21", "m31"
+
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -51,28 +60,70 @@ Functions documentation
       | *[in]*  **src**  pointer to an array of floats
       | *[out]* **dest** destination matrix4x2
 
-.. c:function:: void glm_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat4 dest)
+.. c:function:: void glm_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat2 dest)
 
     multiply m1 and m2 to dest
 
     .. code-block:: c
 
-       glm_mat4x2_mul(mat4x2, mat2x4, mat4);
+       glm_mat4x2_mul(mat4x2, mat2x4, mat2);
 
     Parameters:
       | *[in]*  **m1**    left matrix (mat4x2)
       | *[in]*  **m2**    right matrix (mat2x4)
-      | *[out]* **dest**  destination matrix (mat4)
+      | *[out]* **dest**  destination matrix (mat2)
 
-.. c:function:: void glm_mat4x2_mulv(mat4x2 m, vec2 v, vec4 dest)
+    .. csv-table:: mat4x2
+        :header: "", "column 1", "column 2", "column 3", "column 4"
 
-    multiply mat4x2 with vec2 (column vector) and store in dest vector
+        "row 1", "a00", "a10", "a20", "a30"
+        "row 2", "a01", "a11", "a21", "a31"
+
+    .. csv-table:: mat2x4
+        :header: "", "column 1", "column 2"
+
+        "row 1", "b00", "b10"
+        "row 2", "b01", "b11"
+        "row 3", "b02", "b12"
+        "row 4", "b03", "b13"
+
+    .. csv-table:: mat2x2
+        :header: "", "column 1", "column 2"
+
+        "row 1", "a00 * b00 + a10 * b01 + a20 * b02 + a30 * b03", "a00 * b10 + a10 * b11 + a20 * b12 + a30 * b13"
+        "row 2", "a01 * b00 + a11 * b01 + a21 * b02 + a31 * b03", "a01 * b10 + a11 * b11 + a21 * b12 + a31 * b13"
+
+.. c:function:: void glm_mat4x2_mulv(mat4x2 m, vec4 v, vec2 dest)
+
+    multiply mat4x2 with vec4 (column vector) and store in dest vector
 
     Parameters:
       | *[in]*  **m**     mat4x2 (left)
-      | *[in]*  **v**     vec2 (right, column vector)
+      | *[in]*  **v**     vec4 (right, column vector)
       | *[out]* **dest**  destination (result, column vector)
 
+    .. csv-table:: mat4x2
+        :header: "", "column 1", "column 2", "column 3"
+
+        "row 1", "m00", "m10", "m20"
+        "row 2", "m01", "m11", "m21"
+        "row 3", "m02", "m12", "m22"
+        "row 4", "m03", "m13", "m23"
+
+    .. csv-table:: column vec4 (1x4)
+        :header: "", "column 1"
+
+        "row 1", "v0"
+        "row 2", "v1"
+        "row 3", "v2"
+        "row 4", "v3"
+
+    .. csv-table:: column vec2 (1x2)
+        :header: "", "column 1"
+
+        "row 1", "m00 * v0 + m10 * v1 + m20 * v2 + m30 * v3"
+        "row 2", "m01 * v0 + m11 * v1 + m21 * v2 + m31 * v3"
+
 .. c:function:: void glm_mat4x2_transpose(mat4x2 m, mat2x4 dest)
 
     transpose matrix and store in dest
diff --git a/docs/source/mat4x3.rst b/docs/source/mat4x3.rst
index 574c09a..5136e88 100644
--- a/docs/source/mat4x3.rst
+++ b/docs/source/mat4x3.rst
@@ -23,6 +23,16 @@ Functions:
 #. :c:func:`glm_mat4x3_transpose`
 #. :c:func:`glm_mat4x3_scale`
 
+Represented
+~~~~~~~~~~~
+
+.. csv-table:: mat4x3
+   :header: "", "column 1", "column 2", "column 3", "column4"
+
+   "row 1", "m00", "m10", "m20", "m30"
+   "row 2", "m01", "m11", "m21", "m31"
+   "row 3", "m02", "m12", "m22", "m32"
+
 Functions documentation
 ~~~~~~~~~~~~~~~~~~~~~~~
 
@@ -51,28 +61,72 @@ Functions documentation
       | *[in]*  **src**  pointer to an array of floats
       | *[out]* **dest** destination matrix4x3
 
-.. c:function:: void glm_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat4 dest)
+.. c:function:: void glm_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat3 dest)
 
     multiply m1 and m2 to dest
 
     .. code-block:: c
 
-       glm_mat4x3_mul(mat4x3, mat3x4, mat4);
+       glm_mat4x3_mul(mat4x3, mat3x4, mat3);
 
     Parameters:
       | *[in]*  **m1**    left matrix (mat4x3)
       | *[in]*  **m2**    right matrix (mat3x4)
-      | *[out]* **dest**  destination matrix (mat4)
+      | *[out]* **dest**  destination matrix (mat3)
 
-.. c:function:: void glm_mat4x3_mulv(mat4x3 m, vec3 v, vec4 dest)
+    .. csv-table:: mat4x3
+        :header: "", "column 1", "column 2", "column 3", "column 4"
 
-    multiply mat4x3 with vec3 (column vector) and store in dest vector
+        "row 1", "a00", "a10", "a20", "a30"
+        "row 2", "a01", "a11", "a21", "a31"
+        "row 3", "a02", "a12", "a22", "a32"
+
+    .. csv-table:: mat3x4
+        :header: "", "column 1", "column 2", "column 3"
+
+        "row 1", "b00", "b10", "b20"
+        "row 2", "b01", "b11", "b21"
+        "row 3", "b02", "b12", "b22"
+        "row 4", "b03", "b13", "b23"
+
+    .. csv-table:: mat3x3
+        :header: "", "column 1", "column 2", "column 3"
+
+        "row 1", "a00 * b00 + a10 * b01 + a20 * b02 + a30 * b03", "a00 * b10 + a10 * b11 + a20 * b12 + a30 * b13", "a00 * b20 + a10 * b21 + a20 * b22 + a30 * b23"
+        "row 2", "a01 * b00 + a11 * b01 + a21 * b02 + a31 * b03", "a01 * b10 + a11 * b11 + a21 * b12 + a31 * b13", "a01 * b20 + a11 * b21 + a21 * b22 + a31 * b23"
+        "row 3", "a02 * b00 + a12 * b01 + a22 * b02 + a32 * b03", "a02 * b10 + a12 * b11 + a22 * b12 + a32 * b13", "a02 * b20 + a12 * b21 + a22 * b22 + a32 * b23"
+
+.. c:function:: void glm_mat4x3_mulv(mat4x3 m, vec4 v, vec3 dest)
+
+    multiply mat4x3 with vec4 (column vector) and store in dest column vector
 
     Parameters:
       | *[in]*  **m**     mat4x3 (left)
-      | *[in]*  **v**     vec3 (right, column vector)
+      | *[in]*  **v**     vec4 (right, column vector)
       | *[out]* **dest**  destination (result, column vector)
 
+    .. csv-table:: mat4x3
+        :header: "", "column 1", "column 2", "column 3", "column 4"
+
+        "row 1", "m00", "m10", "m20", "m30"
+        "row 2", "m01", "m11", "m21", "m31"
+        "row 3", "m02", "m12", "m22", "m32"
+
+    .. csv-table:: column vec4 (1x4)
+        :header: "", "column 1"
+
+        "row 1", "v0"
+        "row 2", "v1"
+        "row 3", "v2"
+        "row 4", "v3"
+
+    .. csv-table:: column vec3 (1x3)
+        :header: "", "column 1"
+
+        "row 1", "m00 * v0 + m10 * v1 + m20 * v2 + m30 * v3"
+        "row 2", "m01 * v0 + m11 * v1 + m21 * v2 + m31 * v3"
+        "row 3", "m02 * v0 + m12 * v1 + m22 * v2 + m32 * v3"
+
 .. c:function:: void glm_mat4x3_transpose(mat4x3 m, mat3x4 dest)
 
     transpose matrix and store in dest
diff --git a/docs/source/sphinx-static/theme_overrides.css b/docs/source/sphinx-static/theme_overrides.css
new file mode 100644
index 0000000..0952e2c
--- /dev/null
+++ b/docs/source/sphinx-static/theme_overrides.css
@@ -0,0 +1,12 @@
+@media screen {
+  /* content column
+   *
+   * RTD theme's default is 800px as max width for the content, but we have
+   * tables with tons of columns, which need the full width of the view-port.
+   *
+   * Comment from yocto project theme_overrides.css
+   */
+
+  .wy-nav-content{ max-width: none; }
+
+}
diff --git a/docs/source/vec2.rst b/docs/source/vec2.rst
index 95615d1..e9e38ca 100644
--- a/docs/source/vec2.rst
+++ b/docs/source/vec2.rst
@@ -397,16 +397,16 @@ Functions documentation
       | *[in]*  **src**  pointer to an array of floats
       | *[out]* **dest** destination vector
 
-.. c:function:: void glm_vec2_reflect(vec2 I, vec2 N, vec2 dest) 
+.. c:function:: void glm_vec2_reflect(vec2 v, vec2 n, vec2 dest)
 
     Reflection vector using an incident ray and a surface normal
 
     Parameters:
-      | *[in]*  **I**     incident vector
-      | *[in]*  **N**     *❗️ normalized ❗️* normal vector
+      | *[in]*  **v**     incident vector
+      | *[in]*  **n**     *❗️ normalized ❗️* normal vector
       | *[out]* **dest**  destination: reflection result
 
-.. c:function:: bool glm_vec2_refract(vec2 I, vec2 N, float eta, vec2 dest)
+.. c:function:: bool glm_vec2_refract(vec2 v, vec2 n, float eta, vec2 dest)
 
     Computes refraction vector for an incident vector and a surface normal.
    
@@ -415,8 +415,8 @@ Functions documentation
     Otherwise, computes refraction vector, stores it in dest, and returns true.
 
     Parameters:
-      | *[in]*  **I**     *❗️ normalized ❗️* incident vector
-      | *[in]*  **N**     *❗️ normalized ❗️* normal vector
+      | *[in]*  **v**     *❗️ normalized ❗️* incident vector
+      | *[in]*  **n**     *❗️ normalized ❗️* normal vector
       | *[in]*  **eta**   ratio of indices of refraction (incident/transmitted)
       | *[out]* **dest**  refraction vector if refraction occurs; zero vector otherwise
 
diff --git a/docs/source/vec3.rst b/docs/source/vec3.rst
index 91c16b9..0b989b8 100644
--- a/docs/source/vec3.rst
+++ b/docs/source/vec3.rst
@@ -516,26 +516,26 @@ Functions documentation
       | *[in]*  **src**  pointer to an array of floats
       | *[out]* **dest** destination vector
 
-.. c:function:: void glm_vec3_faceforward(vec3 N, vec3 I, vec3 Nref, vec3 dest)
+.. c:function:: void glm_vec3_faceforward(vec3 n, vec3 v, vec3 nref, vec3 dest)
 
     A vector pointing in the same direction as another
 
     Parameters:
-      | *[in]*  **N**     vector to orient
-      | *[in]*  **I**     incident vector
-      | *[in]*  **Nref**  reference vector
+      | *[in]*  **n**     vector to orient
+      | *[in]*  **v**     incident vector
+      | *[in]*  **nref**  reference vector
       | *[out]* **dest**  destination: oriented vector, pointing away from the surface.
 
-.. c:function:: void glm_vec3_reflect(vec3 I, vec3 N, vec3 dest) 
+.. c:function:: void glm_vec3_reflect(vec3 v, vec3 n, vec3 dest)
 
     Reflection vector using an incident ray and a surface normal
 
     Parameters:
-      | *[in]*  **I**     incident vector
-      | *[in]*  **N**     *❗️ normalized ❗️* normal vector
+      | *[in]*  **v**     incident vector
+      | *[in]*  **n**     *❗️ normalized ❗️* normal vector
       | *[out]* **dest**  destination: reflection result
 
-.. c:function:: bool glm_vec3_refract(vec3 I, vec3 N, float eta, vec3 dest)
+.. c:function:: bool glm_vec3_refract(vec3 v, vec3 n, float eta, vec3 dest)
 
     
     Computes refraction vector for an incident vector and a surface normal.
@@ -545,10 +545,10 @@ Functions documentation
     Otherwise, computes refraction vector, stores it in dest, and returns true.
 
     Parameters:
-      | *[in]*  **I**     *❗️ normalized ❗️* incident vector
-      | *[in]*  **N**     *❗️ normalized ❗️* normal vector
+      | *[in]*  **v**     *❗️ normalized ❗️* incident vector
+      | *[in]*  **n**     *❗️ normalized ❗️* normal vector
       | *[in]*  **eta**   ratio of indices of refraction (incident/transmitted)
       | *[out]* **dest**  refraction vector if refraction occurs; zero vector otherwise
 
     Returns:
-      returns true if refraction occurs; false if total internal reflection occurs.
\ No newline at end of file
+      returns true if refraction occurs; false if total internal reflection occurs.
diff --git a/docs/source/vec4.rst b/docs/source/vec4.rst
index 04c3a82..9035a3d 100644
--- a/docs/source/vec4.rst
+++ b/docs/source/vec4.rst
@@ -427,16 +427,16 @@ Functions documentation
       | *[in]*  **src**  pointer to an array of floats
       | *[out]* **dest** destination vector
 
-.. c:function:: bool glm_vec4_reflect(vec4 I, vec4 N, vec4 dest) 
+.. c:function:: bool glm_vec4_reflect(vec4 v, vec4 n, vec4 dest)
 
     Reflection vector using an incident ray and a surface normal
 
     Parameters:
-      | *[in]*  **I**     incident vector
-      | *[in]*  **N**     *❗️ normalized ❗️* normal vector
+      | *[in]*  **v**     incident vector
+      | *[in]*  **n**     *❗️ normalized ❗️* normal vector
       | *[out]* **dest**  destination: reflection result
 
-.. c:function:: bool glm_vec4_refract(vec4 I, vec4 N, float eta, vec4 dest)
+.. c:function:: bool glm_vec4_refract(vec4 v, vec4 n, float eta, vec4 dest)
 
     computes refraction vector for an incident vector and a surface normal.
    
@@ -449,10 +449,10 @@ Functions documentation
     the 'w' component should manually adjust 'dest' after calling this function.
 
     Parameters:
-      | *[in]*  **I**     *❗️ normalized ❗️* incident vector
-      | *[in]*  **N**     *❗️ normalized ❗️* normal vector
+      | *[in]*  **v**     *❗️ normalized ❗️* incident vector
+      | *[in]*  **n**     *❗️ normalized ❗️* normal vector
       | *[in]*  **eta**   ratio of indices of refraction (incident/transmitted)
       | *[out]* **dest**  refraction vector if refraction occurs; zero vector otherwise
 
     Returns:
-      returns true if refraction occurs; false if total internal reflection occurs.
\ No newline at end of file
+      returns true if refraction occurs; false if total internal reflection occurs.
diff --git a/include/cglm/call/mat2x3.h b/include/cglm/call/mat2x3.h
index 90ff606..59c9ee3 100644
--- a/include/cglm/call/mat2x3.h
+++ b/include/cglm/call/mat2x3.h
@@ -27,11 +27,11 @@ glmc_mat2x3_make(const float * __restrict src, mat2x3 dest);
 
 CGLM_EXPORT
 void
-glmc_mat2x3_mul(mat2x3 m1, mat3x2 m2, mat2 dest);
+glmc_mat2x3_mul(mat2x3 m1, mat3x2 m2, mat3 dest);
 
 CGLM_EXPORT
 void
-glmc_mat2x3_mulv(mat2x3 m, vec3 v, vec2 dest);
+glmc_mat2x3_mulv(mat2x3 m, vec2 v, vec3 dest);
 
 CGLM_EXPORT
 void
diff --git a/include/cglm/call/mat2x4.h b/include/cglm/call/mat2x4.h
index b11fd0e..84e805b 100644
--- a/include/cglm/call/mat2x4.h
+++ b/include/cglm/call/mat2x4.h
@@ -27,11 +27,11 @@ glmc_mat2x4_make(const float * __restrict src, mat2x4 dest);
 
 CGLM_EXPORT
 void
-glmc_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat2 dest);
+glmc_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat4 dest);
 
 CGLM_EXPORT
 void
-glmc_mat2x4_mulv(mat2x4 m, vec4 v, vec2 dest);
+glmc_mat2x4_mulv(mat2x4 m, vec2 v, vec4 dest);
 
 CGLM_EXPORT
 void
diff --git a/include/cglm/call/mat3x2.h b/include/cglm/call/mat3x2.h
index 302ae93..3c0daed 100644
--- a/include/cglm/call/mat3x2.h
+++ b/include/cglm/call/mat3x2.h
@@ -27,11 +27,11 @@ glmc_mat3x2_make(const float * __restrict src, mat3x2 dest);
 
 CGLM_EXPORT
 void
-glmc_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat3 dest);
+glmc_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat2 dest);
 
 CGLM_EXPORT
 void
-glmc_mat3x2_mulv(mat3x2 m, vec2 v, vec3 dest);
+glmc_mat3x2_mulv(mat3x2 m, vec3 v, vec2 dest);
 
 CGLM_EXPORT
 void
diff --git a/include/cglm/call/mat3x4.h b/include/cglm/call/mat3x4.h
index 5339e16..ff23b4c 100644
--- a/include/cglm/call/mat3x4.h
+++ b/include/cglm/call/mat3x4.h
@@ -27,11 +27,11 @@ glmc_mat3x4_make(const float * __restrict src, mat3x4 dest);
 
 CGLM_EXPORT
 void
-glmc_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat3 dest);
+glmc_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat4 dest);
 
 CGLM_EXPORT
 void
-glmc_mat3x4_mulv(mat3x4 m, vec4 v, vec3 dest);
+glmc_mat3x4_mulv(mat3x4 m, vec3 v, vec4 dest);
 
 CGLM_EXPORT
 void
diff --git a/include/cglm/call/mat4x2.h b/include/cglm/call/mat4x2.h
index b60125d..2989124 100644
--- a/include/cglm/call/mat4x2.h
+++ b/include/cglm/call/mat4x2.h
@@ -27,11 +27,11 @@ glmc_mat4x2_make(const float * __restrict src, mat4x2 dest);
 
 CGLM_EXPORT
 void
-glmc_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat4 dest);
+glmc_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat2 dest);
 
 CGLM_EXPORT
 void
-glmc_mat4x2_mulv(mat4x2 m, vec2 v, vec4 dest);
+glmc_mat4x2_mulv(mat4x2 m, vec4 v, vec2 dest);
 
 CGLM_EXPORT
 void
diff --git a/include/cglm/call/mat4x3.h b/include/cglm/call/mat4x3.h
index 131cf51..bb5c0a8 100644
--- a/include/cglm/call/mat4x3.h
+++ b/include/cglm/call/mat4x3.h
@@ -27,11 +27,11 @@ glmc_mat4x3_make(const float * __restrict src, mat4x3 dest);
 
 CGLM_EXPORT
 void
-glmc_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat4 dest);
+glmc_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat3 dest);
 
 CGLM_EXPORT
 void
-glmc_mat4x3_mulv(mat4x3 m, vec3 v, vec4 dest);
+glmc_mat4x3_mulv(mat4x3 m, vec4 v, vec3 dest);
 
 CGLM_EXPORT
 void
diff --git a/include/cglm/call/vec2.h b/include/cglm/call/vec2.h
index 2343d14..507f042 100644
--- a/include/cglm/call/vec2.h
+++ b/include/cglm/call/vec2.h
@@ -199,11 +199,11 @@ glmc_vec2_make(const float * __restrict src, vec2 dest);
 
 CGLM_EXPORT
 void
-glmc_vec2_reflect(vec2 I, vec2 N, vec2 dest);
+glmc_vec2_reflect(vec2 v, vec2 n, vec2 dest);
 
 CGLM_EXPORT
 bool
-glmc_vec2_refract(vec2 I, vec2 N, float eta, vec2 dest);
+glmc_vec2_refract(vec2 v, vec2 n, float eta, vec2 dest);
 
 #ifdef __cplusplus
 }
diff --git a/include/cglm/call/vec3.h b/include/cglm/call/vec3.h
index 2bde3ba..bb5a5d7 100644
--- a/include/cglm/call/vec3.h
+++ b/include/cglm/call/vec3.h
@@ -336,15 +336,15 @@ glmc_vec3_make(const float * __restrict src, vec3 dest);
 
 CGLM_EXPORT
 void
-glmc_vec3_faceforward(vec3 N, vec3 I, vec3 Nref, vec3 dest);
+glmc_vec3_faceforward(vec3 n, vec3 v, vec3 nref, vec3 dest);
 
 CGLM_EXPORT
 void
-glmc_vec3_reflect(vec3 I, vec3 N, vec3 dest);
+glmc_vec3_reflect(vec3 v, vec3 n, vec3 dest);
 
 CGLM_EXPORT
 bool
-glmc_vec3_refract(vec3 I, vec3 N, float eta, vec3 dest);
+glmc_vec3_refract(vec3 v, vec3 n, float eta, vec3 dest);
 
 #ifdef __cplusplus
 }
diff --git a/include/cglm/call/vec4.h b/include/cglm/call/vec4.h
index d121bea..a976e94 100644
--- a/include/cglm/call/vec4.h
+++ b/include/cglm/call/vec4.h
@@ -313,11 +313,11 @@ glmc_vec4_make(const float * __restrict src, vec4 dest);
 
 CGLM_EXPORT
 void
-glmc_vec4_reflect(vec4 I, vec4 N, vec4 dest);
+glmc_vec4_reflect(vec4 v, vec4 n, vec4 dest);
 
 CGLM_EXPORT
 bool
-glmc_vec4_refract(vec4 I, vec4 N, float eta, vec4 dest);
+glmc_vec4_refract(vec4 v, vec4 n, float eta, vec4 dest);
 
 #ifdef __cplusplus
 }
diff --git a/include/cglm/common.h b/include/cglm/common.h
index d3c3484..af1116f 100644
--- a/include/cglm/common.h
+++ b/include/cglm/common.h
@@ -45,6 +45,10 @@
 #  define CGLM_LIKELY(expr)   (expr)
 #endif
 
+#if defined(_M_FP_FAST) || defined(__FAST_MATH__)
+#  define CGLM_FAST_MATH
+#endif
+
 #define GLM_SHUFFLE4(z, y, x, w) (((z) << 6) | ((y) << 4) | ((x) << 2) | (w))
 #define GLM_SHUFFLE3(z, y, x)    (((z) << 4) | ((y) << 2) | (x))
 
diff --git a/include/cglm/mat2x3.h b/include/cglm/mat2x3.h
index 06fec14..811baaf 100644
--- a/include/cglm/mat2x3.h
+++ b/include/cglm/mat2x3.h
@@ -14,8 +14,8 @@
    CGLM_INLINE void glm_mat2x3_copy(mat2x3 mat, mat2x3 dest);
    CGLM_INLINE void glm_mat2x3_zero(mat2x3 mat);
    CGLM_INLINE void glm_mat2x3_make(const 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_mul(mat2x3 m1, mat3x2 m2, mat3 dest);
+   CGLM_INLINE void glm_mat2x3_mulv(mat2x3 m, vec2 v, vec3 dest);
    CGLM_INLINE void glm_mat2x3_transpose(mat2x3 m, mat3x2 dest);
    CGLM_INLINE void glm_mat2x3_scale(mat2x3 m, float s);
  */
@@ -82,7 +82,7 @@ glm_mat2x3_make(const float * __restrict src, mat2x3 dest) {
  * @brief multiply m1 and m2 to dest
  *
  * @code
- * glm_mat2x3_mul(mat2x3, mat3x2, mat2);
+ * glm_mat2x3_mul(mat2x3, mat3x2, mat3);
  * @endcode
  *
  * @param[in]  m1   left matrix (mat2x3)
@@ -91,19 +91,25 @@ glm_mat2x3_make(const float * __restrict src, mat2x3 dest) {
  */
 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],
+glm_mat2x3_mul(mat2x3 m1, mat3x2 m2, mat3 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[0][0] = a00 * b00 + a10 * b01;
+  dest[0][1] = a01 * b00 + a11 * b01;
+  dest[0][2] = a02 * b00 + a12 * b01;
 
-  dest[1][0] = a10 * b00 + a11 * b10 + a12 * b20;
-  dest[1][1] = a10 * b01 + a11 * b11 + a12 * b21;
+  dest[1][0] = a00 * b10 + a10 * b11;
+  dest[1][1] = a01 * b10 + a11 * b11;
+  dest[1][2] = a02 * b10 + a12 * b11;
+
+  dest[2][0] = a00 * b20 + a10 * b21;
+  dest[2][1] = a01 * b20 + a11 * b21;
+  dest[2][2] = a02 * b20 + a12 * b21;
 }
 
 /*!
@@ -115,11 +121,12 @@ glm_mat2x3_mul(mat2x3 m1, mat3x2 m2, mat2 dest) {
  */
 CGLM_INLINE
 void
-glm_mat2x3_mulv(mat2x3 m, vec3 v, vec2 dest) {
-  float v0 = v[0], v1 = v[1], v2 = v[2];
+glm_mat2x3_mulv(mat2x3 m, vec2 v, vec3 dest) {
+  float v0 = v[0], v1 = v[1];
 
-  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[0] = m[0][0] * v0 + m[1][0] * v1;
+  dest[1] = m[0][1] * v0 + m[1][1] * v1;
+  dest[2] = m[0][2] * v0 + m[1][2] * v1;
 }
 
 /*!
diff --git a/include/cglm/mat2x4.h b/include/cglm/mat2x4.h
index 9b1bd48..e5c1281 100644
--- a/include/cglm/mat2x4.h
+++ b/include/cglm/mat2x4.h
@@ -14,8 +14,8 @@
    CGLM_INLINE void glm_mat2x4_copy(mat2x4 mat, mat2x4 dest);
    CGLM_INLINE void glm_mat2x4_zero(mat2x4 mat);
    CGLM_INLINE void glm_mat2x4_make(const 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_mul(mat2x4 m1, mat4x2 m2, mat4 dest);
+   CGLM_INLINE void glm_mat2x4_mulv(mat2x4 m, vec2 v, vec4 dest);
    CGLM_INLINE void glm_mat2x4_transpose(mat2x4 m, mat4x2 dest);
    CGLM_INLINE void glm_mat2x4_scale(mat2x4 m, float s);
  */
@@ -80,16 +80,16 @@ glm_mat2x4_make(const float * __restrict src, mat2x4 dest) {
  * @brief multiply m1 and m2 to dest
  *
  * @code
- * glm_mat2x4_mul(mat2x4, mat4x2, mat2);
+ * glm_mat2x4_mul(mat2x4, mat4x2, mat4);
  * @endcode
  *
  * @param[in]  m1   left matrix (mat2x4)
  * @param[in]  m2   right matrix (mat4x2)
- * @param[out] dest destination matrix (mat2)
+ * @param[out] dest destination matrix (mat4)
  */
 CGLM_INLINE
 void
-glm_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat2 dest) {
+glm_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat4 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],
 
@@ -98,15 +98,29 @@ glm_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat2 dest) {
         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[0][0] = a00 * b00 + a10 * b01;
+  dest[0][1] = a01 * b00 + a11 * b01;
+  dest[0][2] = a02 * b00 + a12 * b01;
+  dest[0][3] = a03 * b00 + a13 * b01;
 
-  dest[1][0] = a10 * b00 + a11 * b10 + a12 * b20 + a13 * b30;
-  dest[1][1] = a10 * b01 + a11 * b11 + a12 * b21 + a13 * b31;
+  dest[1][0] = a00 * b10 + a10 * b11;
+  dest[1][1] = a01 * b10 + a11 * b11;
+  dest[1][2] = a02 * b10 + a12 * b11;
+  dest[1][3] = a03 * b10 + a13 * b11;
+
+  dest[2][0] = a00 * b20 + a10 * b21;
+  dest[2][1] = a01 * b20 + a11 * b21;
+  dest[2][2] = a02 * b20 + a12 * b21;
+  dest[2][3] = a03 * b20 + a13 * b21;
+
+  dest[3][0] = a00 * b30 + a10 * b31;
+  dest[3][1] = a01 * b30 + a11 * b31;
+  dest[3][2] = a02 * b30 + a12 * b31;
+  dest[3][3] = a03 * b30 + a13 * b31;
 }
 
 /*!
- * @brief multiply matrix with column vector and store in dest vector
+ * @brief multiply matrix with column vector and store in dest column vector
  *
  * @param[in]  m    matrix (left)
  * @param[in]  v    vector (right, column vector)
@@ -114,11 +128,13 @@ glm_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat2 dest) {
  */
 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];
+glm_mat2x4_mulv(mat2x4 m, vec2 v, vec4 dest) {
+  float v0 = v[0], v1 = v[1];
 
-  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[0] = m[0][0] * v0 + m[1][0] * v1;
+  dest[1] = m[0][1] * v0 + m[1][1] * v1;
+  dest[2] = m[0][2] * v0 + m[1][2] * v1;
+  dest[3] = m[0][3] * v0 + m[1][3] * v1;
 }
 
 /*!
diff --git a/include/cglm/mat3x2.h b/include/cglm/mat3x2.h
index 0ed466f..75ed9c2 100644
--- a/include/cglm/mat3x2.h
+++ b/include/cglm/mat3x2.h
@@ -14,8 +14,8 @@
    CGLM_INLINE void glm_mat3x2_copy(mat3x2 mat, mat3x2 dest);
    CGLM_INLINE void glm_mat3x2_zero(mat3x2 mat);
    CGLM_INLINE void glm_mat3x2_make(const 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_mul(mat3x2 m1, mat2x3 m2, mat2 dest);
+   CGLM_INLINE void glm_mat3x2_mulv(mat3x2 m, vec3 v, vec2 dest);
    CGLM_INLINE void glm_mat3x2_transpose(mat3x2 m, mat2x3 dest);
    CGLM_INLINE void glm_mat3x2_scale(mat3x2 m, float s);
  */
@@ -84,16 +84,16 @@ glm_mat3x2_make(const float * __restrict src, mat3x2 dest) {
  * @brief multiply m1 and m2 to dest
  *
  * @code
- * glm_mat3x2_mul(mat3x2, mat2x3, mat3);
+ * glm_mat3x2_mul(mat3x2, mat2x3, mat2);
  * @endcode
  *
  * @param[in]  m1   left matrix (mat3x2)
  * @param[in]  m2   right matrix (mat2x3)
- * @param[out] dest destination matrix (mat3)
+ * @param[out] dest destination matrix (mat2)
  */
 CGLM_INLINE
 void
-glm_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat3 dest) {
+glm_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat2 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],
@@ -101,21 +101,15 @@ glm_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat3 dest) {
         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[0][0] = a00 * b00 + a10 * b01 + a20 * b02;
+  dest[0][1] = a01 * b00 + a11 * b01 + a21 * b02;
 
-  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;
+  dest[1][0] = a00 * b10 + a10 * b11 + a20 * b12;
+  dest[1][1] = a01 * b10 + a11 * b11 + a21 * b12;
 }
 
 /*!
- * @brief multiply matrix with column vector and store in dest vector
+ * @brief multiply matrix with column vector and store in dest column vector
  *
  * @param[in]  m    matrix (left)
  * @param[in]  v    vector (right, column vector)
@@ -123,12 +117,11 @@ glm_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat3 dest) {
  */
 CGLM_INLINE
 void
-glm_mat3x2_mulv(mat3x2 m, vec2 v, vec3 dest) {
-  float v0 = v[0], v1 = v[1];
+glm_mat3x2_mulv(mat3x2 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;
-  dest[1] = m[1][0] * v0 + m[1][1] * v1;
-  dest[2] = m[2][0] * v0 + m[2][1] * v1;
+  dest[0] = m[0][0] * v0 + m[1][0] * v1 + m[2][0] * v2;
+  dest[1] = m[0][1] * v0 + m[1][1] * v1 + m[2][1] * v2;
 }
 
 /*!
diff --git a/include/cglm/mat3x4.h b/include/cglm/mat3x4.h
index aec4295..8d14523 100644
--- a/include/cglm/mat3x4.h
+++ b/include/cglm/mat3x4.h
@@ -14,8 +14,8 @@
    CGLM_INLINE void glm_mat3x4_copy(mat3x4 mat, mat3x4 dest);
    CGLM_INLINE void glm_mat3x4_zero(mat3x4 mat);
    CGLM_INLINE void glm_mat3x4_make(const 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_mul(mat3x4 m1, mat4x3 m2, mat4 dest);
+   CGLM_INLINE void glm_mat3x4_mulv(mat3x4 m, vec3 v, vec4 dest);
    CGLM_INLINE void glm_mat3x4_transpose(mat3x4 m, mat4x3 dest);
    CGLM_INLINE void glm_mat3x4_scale(mat3x4 m, float s);
  */
@@ -87,16 +87,16 @@ glm_mat3x4_make(const float * __restrict src, mat3x4 dest) {
  * @brief multiply m1 and m2 to dest
  *
  * @code
- * glm_mat3x4_mul(mat3x4, mat4x3, mat3);
+ * glm_mat3x4_mul(mat3x4, mat4x3, mat4);
  * @endcode
  *
  * @param[in]  m1   left matrix (mat3x4)
  * @param[in]  m2   right matrix (mat4x3)
- * @param[out] dest destination matrix (mat3)
+ * @param[out] dest destination matrix (mat4)
  */
 CGLM_INLINE
 void
-glm_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat3 dest) {
+glm_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat4 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],
@@ -106,21 +106,29 @@ glm_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat3 dest) {
         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[0][0] = a00 * b00 + a10 * b01 + a20 * b02;
+  dest[0][1] = a01 * b00 + a11 * b01 + a21 * b02;
+  dest[0][2] = a02 * b00 + a12 * b01 + a22 * b02;
+  dest[0][3] = a03 * b00 + a13 * b01 + a23 * b02;
 
-  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[1][0] = a00 * b10 + a10 * b11 + a20 * b12;
+  dest[1][1] = a01 * b10 + a11 * b11 + a21 * b12;
+  dest[1][2] = a02 * b10 + a12 * b11 + a22 * b12;
+  dest[1][3] = a03 * b10 + a13 * b11 + a23 * b12;
 
-  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;
+  dest[2][0] = a00 * b20 + a10 * b21 + a20 * b22;
+  dest[2][1] = a01 * b20 + a11 * b21 + a21 * b22;
+  dest[2][2] = a02 * b20 + a12 * b21 + a22 * b22;
+  dest[2][3] = a03 * b20 + a13 * b21 + a23 * b22;
+
+  dest[3][0] = a00 * b30 + a10 * b31 + a20 * b32;
+  dest[3][1] = a01 * b30 + a11 * b31 + a21 * b32;
+  dest[3][2] = a02 * b30 + a12 * b31 + a22 * b32;
+  dest[3][3] = a03 * b30 + a13 * b31 + a23 * b32;
 }
 
 /*!
- * @brief multiply matrix with column vector and store in dest vector
+ * @brief multiply matrix with column vector and store in dest column vector
  *
  * @param[in]  m    matrix (left)
  * @param[in]  v    vector (right, column vector)
@@ -128,12 +136,13 @@ glm_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat3 dest) {
  */
 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];
+glm_mat3x4_mulv(mat3x4 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 + 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;
+  dest[0] = m[0][0] * v0 + m[1][0] * v1 + m[2][0] * v2;
+  dest[1] = m[0][1] * v0 + m[1][1] * v1 + m[2][1] * v2;
+  dest[2] = m[0][2] * v0 + m[1][2] * v1 + m[2][2] * v2;
+  dest[3] = m[0][3] * v0 + m[1][3] * v1 + m[2][3] * v2;
 }
 
 /*!
@@ -162,10 +171,9 @@ glm_mat3x4_transpose(mat3x4 m, mat4x3 dest) {
 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;
+  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;
+  m[2][0] *= s; m[2][1] *= s; m[2][2] *= s; m[2][3] *= s;
 }
 
 #endif
diff --git a/include/cglm/mat4x2.h b/include/cglm/mat4x2.h
index b1784d4..d3f0ad5 100644
--- a/include/cglm/mat4x2.h
+++ b/include/cglm/mat4x2.h
@@ -14,8 +14,8 @@
    CGLM_INLINE void glm_mat4x2_copy(mat4x2 mat, mat4x2 dest);
    CGLM_INLINE void glm_mat4x2_zero(mat4x2 mat);
    CGLM_INLINE void glm_mat4x2_make(const 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_mul(mat4x2 m1, mat2x4 m2, mat2 dest);
+   CGLM_INLINE void glm_mat4x2_mulv(mat4x2 m, vec4 v, vec2 dest);
    CGLM_INLINE void glm_mat4x2_transpose(mat4x2 m, mat2x4 dest);
    CGLM_INLINE void glm_mat4x2_scale(mat4x2 m, float s);
  */
@@ -90,16 +90,16 @@ glm_mat4x2_make(const float * __restrict src, mat4x2 dest) {
  * @brief multiply m1 and m2 to dest
  *
  * @code
- * glm_mat4x2_mul(mat4x2, mat2x4, mat4);
+ * glm_mat4x2_mul(mat4x2, mat2x4, mat2);
  * @endcode
  *
  * @param[in]  m1   left matrix (mat4x2)
  * @param[in]  m2   right matrix (mat2x4)
- * @param[out] dest destination matrix (mat4)
+ * @param[out] dest destination matrix (mat2)
  */
 CGLM_INLINE
 void
-glm_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat4 dest) {
+glm_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat2 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],
@@ -108,29 +108,15 @@ glm_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat4 dest) {
         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[0][0] = a00 * b00 + a10 * b01 + a20 * b02 + a30 * b03;
+  dest[0][1] = a01 * b00 + a11 * b01 + a21 * b02 + a31 * b03;
 
-  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;
+  dest[1][0] = a00 * b10 + a10 * b11 + a20 * b12 + a30 * b13;
+  dest[1][1] = a01 * b10 + a11 * b11 + a21 * b12 + a31 * b13;
 }
 
 /*!
- * @brief multiply matrix with column vector and store in dest vector
+ * @brief multiply matrix with column vector and store in dest column vector
  *
  * @param[in]  m    matrix (left)
  * @param[in]  v    vector (right, column vector)
@@ -138,13 +124,11 @@ glm_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat4 dest) {
  */
 CGLM_INLINE
 void
-glm_mat4x2_mulv(mat4x2 m, vec2 v, vec4 dest) {
-  float v0 = v[0], v1 = v[1];
+glm_mat4x2_mulv(mat4x2 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;
-  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;
+  dest[0] = m[0][0] * v0 + m[1][0] * v1 + m[2][0] * v2 + m[3][0] * v3;
+  dest[1] = m[0][1] * v0 + m[1][1] * v1 + m[2][1] * v2 + m[3][1] * v3;
 }
 
 /*!
diff --git a/include/cglm/mat4x3.h b/include/cglm/mat4x3.h
index 4266278..2ef05c1 100644
--- a/include/cglm/mat4x3.h
+++ b/include/cglm/mat4x3.h
@@ -14,8 +14,8 @@
    CGLM_INLINE void glm_mat4x3_copy(mat4x3 mat, mat4x3 dest);
    CGLM_INLINE void glm_mat4x3_zero(mat4x3 mat);
    CGLM_INLINE void glm_mat4x3_make(const 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_mul(mat4x3 m1, mat3x4 m2, mat3 dest);
+   CGLM_INLINE void glm_mat4x3_mulv(mat4x3 m, vec4 v, vec3 dest);
    CGLM_INLINE void glm_mat4x3_transpose(mat4x3 m, mat3x4 dest);
    CGLM_INLINE void glm_mat4x3_scale(mat4x3 m, float s);
  */
@@ -99,16 +99,16 @@ glm_mat4x3_make(const float * __restrict src, mat4x3 dest) {
  * @brief multiply m1 and m2 to dest
  *
  * @code
- * glm_mat4x3_mul(mat4x3, mat3x4, mat4);
+ * glm_mat4x3_mul(mat4x3, mat3x4, mat3);
  * @endcode
  *
  * @param[in]  m1   left matrix (mat4x3)
  * @param[in]  m2   right matrix (mat3x4)
- * @param[out] dest destination matrix (mat4)
+ * @param[out] dest destination matrix (mat3)
  */
 CGLM_INLINE
 void
-glm_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat4 dest) {
+glm_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat3 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],
@@ -118,29 +118,21 @@ glm_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat4 dest) {
         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[0][0] = a00 * b00 + a10 * b01 + a20 * b02 + a30 * b03;
+  dest[0][1] = a01 * b00 + a11 * b01 + a21 * b02 + a31 * b03;
+  dest[0][2] = a02 * b00 + a12 * b01 + a22 * b02 + a32 * b03;
 
-  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[1][0] = a00 * b10 + a10 * b11 + a20 * b12 + a30 * b13;
+  dest[1][1] = a01 * b10 + a11 * b11 + a21 * b12 + a31 * b13;
+  dest[1][2] = a02 * b10 + a12 * b11 + a22 * b12 + a32 * b13;
 
-  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;
+  dest[2][0] = a00 * b20 + a10 * b21 + a20 * b22 + a30 * b23;
+  dest[2][1] = a01 * b20 + a11 * b21 + a21 * b22 + a31 * b23;
+  dest[2][2] = a02 * b20 + a12 * b21 + a22 * b22 + a32 * b23;
 }
 
 /*!
- * @brief multiply matrix with column vector and store in dest vector
+ * @brief multiply matrix with column vector and store in dest column vector
  *
  * @param[in]  m    matrix (left)
  * @param[in]  v    vector (right, column vector)
@@ -148,13 +140,12 @@ glm_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat4 dest) {
  */
 CGLM_INLINE
 void
-glm_mat4x3_mulv(mat4x3 m, vec3 v, vec4 dest) {
-  float v0 = v[0], v1 = v[1], v2 = v[2];
+glm_mat4x3_mulv(mat4x3 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;
-  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;
+  dest[0] = m[0][0] * v0 + m[1][0] * v1 + m[2][0] * v2 + m[3][0] * v3;
+  dest[1] = m[0][1] * v0 + m[1][1] * v1 + m[2][1] * v2 + m[3][1] * v3;
+  dest[2] = m[0][2] * v0 + m[1][2] * v1 + m[2][2] * v2 + m[3][2] * v3;
 }
 
 /*!
diff --git a/include/cglm/simd/intrin.h b/include/cglm/simd/intrin.h
index 137df65..11c46e5 100644
--- a/include/cglm/simd/intrin.h
+++ b/include/cglm/simd/intrin.h
@@ -10,6 +10,9 @@
 
 #if defined( _MSC_VER )
 #  if (defined(_M_AMD64) || defined(_M_X64)) || _M_IX86_FP == 2
+#    ifndef __SSE__
+#      define __SSE__
+#    endif
 #    ifndef __SSE2__
 #      define __SSE2__
 #    endif
@@ -24,15 +27,22 @@
 #  endif
 #endif
 
-#if defined( __SSE__ ) || defined( __SSE2__ )
+#if defined(__SSE__)
 #  include <xmmintrin.h>
-#  include <emmintrin.h>
 #  define CGLM_SSE_FP 1
 #  ifndef CGLM_SIMD_x86
 #    define CGLM_SIMD_x86
 #  endif
 #endif
 
+#if defined(__SSE2__)
+#  include <emmintrin.h>
+#  define CGLM_SSE2_FP 1
+#  ifndef CGLM_SIMD_x86
+#    define CGLM_SIMD_x86
+#  endif
+#endif
+
 #if defined(__SSE3__)
 #  include <pmmintrin.h>
 #  ifndef CGLM_SIMD_x86
diff --git a/include/cglm/simd/x86.h b/include/cglm/simd/x86.h
index 8fd5a72..81081dc 100644
--- a/include/cglm/simd/x86.h
+++ b/include/cglm/simd/x86.h
@@ -21,7 +21,7 @@
 #define glmm_set1(x) _mm_set1_ps(x)
 #define glmm_128     __m128
 
-#ifdef CGLM_USE_INT_DOMAIN
+#if defined(CGLM_USE_INT_DOMAIN) && defined(__SSE2__)
 #  define glmm_shuff1(xmm, z, y, x, w)                                        \
      _mm_castsi128_ps(_mm_shuffle_epi32(_mm_castps_si128(xmm),                \
                                         _MM_SHUFFLE(z, y, x, w)))
@@ -55,17 +55,40 @@
 #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)                                           \
+#if defined(__SSE2__)
+#  define GLMM_NEGZEROf ((int)0x80000000) /*  0x80000000 ---> -0.0f  */
+#  define GLMM_POSZEROf ((int)0x00000000) /*  0x00000000 ---> +0.0f  */
+#else
+#  ifdef CGLM_FAST_MATH
+     union { int i; float f; } static GLMM_NEGZEROf_TU = { .i = (int)0x80000000 };
+#    define GLMM_NEGZEROf GLMM_NEGZEROf_TU.f
+#    define GLMM_POSZEROf 0.0f
+#  else
+#    define GLMM_NEGZEROf -0.0f
+#    define GLMM_POSZEROf  0.0f
+#  endif
+#endif
+
+#if defined(__SSE2__)
+#  define GLMM__SIGNMASKf(X, Y, Z, W)                                         \
    _mm_castsi128_ps(_mm_set_epi32(X, Y, Z, W))
   /* _mm_set_ps(X, Y, Z, W); */
+#else
+#  define GLMM__SIGNMASKf(X, Y, Z, W)  _mm_set_ps(X, Y, Z, W)
+#endif
 
-#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_PNPN GLMM__SIGNMASKf(GLMM_POSZEROf, GLMM_NEGZEROf, GLMM_POSZEROf, GLMM_NEGZEROf)
+#define glmm_float32x4_SIGNMASK_NPNP GLMM__SIGNMASKf(GLMM_NEGZEROf, GLMM_POSZEROf, GLMM_NEGZEROf, GLMM_POSZEROf)
+#define glmm_float32x4_SIGNMASK_NPPN GLMM__SIGNMASKf(GLMM_NEGZEROf, GLMM_POSZEROf, GLMM_POSZEROf, GLMM_NEGZEROf)
+
+/* fasth math prevents -0.0f to work */
+#if defined(__SSE2__)
+#  define glmm_float32x4_SIGNMASK_NEG _mm_castsi128_ps(_mm_set1_epi32(GLMM_NEGZEROf)) /* _mm_set1_ps(-0.0f) */
+#else
+#  define glmm_float32x4_SIGNMASK_NEG _mm_set1_ps(GLMM_NEGZEROf)
+#endif
 
-#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
@@ -207,6 +230,7 @@ glmm_norm_inf(__m128 a) {
   return _mm_cvtss_f32(glmm_vhmax(glmm_abs(a)));
 }
 
+#if defined(__SSE2__)
 static inline
 __m128
 glmm_load3(float v[3]) {
@@ -225,6 +249,7 @@ glmm_store3(float v[3], __m128 vx) {
   _mm_storel_pi(CGLM_CASTPTR_ASSUME_ALIGNED(v, __m64), vx);
   _mm_store_ss(&v[2], glmm_shuff1(vx, 2, 2, 2, 2));
 }
+#endif
 
 static inline
 __m128
diff --git a/include/cglm/struct/mat2x3.h b/include/cglm/struct/mat2x3.h
index 77084a7..3798893 100644
--- a/include/cglm/struct/mat2x3.h
+++ b/include/cglm/struct/mat2x3.h
@@ -14,7 +14,7 @@
    CGLM_INLINE mat2x3s glms_mat2x3_zero(void);
    CGLM_INLINE mat2x3s glms_mat2x3_make(const float * __restrict src);
    CGLM_INLINE mat2s   glms_mat2x3_mul(mat2x3s m1, mat3x2s m2);
-   CGLM_INLINE vec2s   glms_mat2x3_mulv(mat2x3s m, vec3s v);
+   CGLM_INLINE vec3s   glms_mat2x3_mulv(mat2x3s m, vec2s v);
    CGLM_INLINE mat3x2s glms_mat2x3_transpose(mat2x3s m);
    CGLM_INLINE mat2x3s glms_mat2x3_scale(mat2x3s m, float s);
  */
@@ -73,9 +73,9 @@ glms_mat2x3_(make)(const float * __restrict src) {
  * @returns destination matrix (mat2s)
  */
 CGLM_INLINE
-mat2s
+mat3s
 glms_mat2x3_(mul)(mat2x3s m1, mat3x2s m2) {
-  mat2s r;
+  mat3s r;
   glm_mat2x3_mul(m1.raw, m2.raw, r.raw);
   return r;
 }
@@ -85,12 +85,12 @@ glms_mat2x3_(mul)(mat2x3s m1, mat3x2s m2) {
  *
  * @param[in]  m    matrix (left)
  * @param[in]  v    vector (right, column vector)
- * @param[out] dest result vector
+ * @returns destination vector (vec3s)
  */
 CGLM_INLINE
-vec2s
-glms_mat2x3_(mulv)(mat2x3s m, vec3s v) {
-  vec2s r;
+vec3s
+glms_mat2x3_(mulv)(mat2x3s m, vec2s v) {
+  vec3s r;
   glm_mat2x3_mulv(m.raw, v.raw, r.raw);
   return r;
 }
diff --git a/include/cglm/struct/mat2x4.h b/include/cglm/struct/mat2x4.h
index fbb63e0..afeedc1 100644
--- a/include/cglm/struct/mat2x4.h
+++ b/include/cglm/struct/mat2x4.h
@@ -14,7 +14,7 @@
    CGLM_INLINE mat2x4s glms_mat2x4_zero(void);
    CGLM_INLINE mat2x4s glms_mat2x4_make(const float * __restrict src);
    CGLM_INLINE mat2s   glms_mat2x4_mul(mat2x4s m1, mat4x2s m2);
-   CGLM_INLINE vec2s   glms_mat2x4_mulv(mat2x4s m, vec4s v);
+   CGLM_INLINE vec4s   glms_mat2x4_mulv(mat2x4s m, vec2s v);
    CGLM_INLINE mat4x2s glms_mat2x4_transpose(mat2x4s m);
    CGLM_INLINE mat2x4s glms_mat2x4_scale(mat2x4s m, float s);
  */
@@ -73,24 +73,24 @@ glms_mat2x4_(make)(const float * __restrict src) {
  * @returns destination matrix (mat2s)
  */
 CGLM_INLINE
-mat2s
+mat4s
 glms_mat2x4_(mul)(mat2x4s m1, mat4x2s m2) {
-  mat2s r;
+  mat4s r;
   glm_mat2x4_mul(m1.raw, m2.raw, r.raw);
   return r;
 }
 
 /*!
- * @brief multiply matrix with column vector and store in dest vector
+ * @brief multiply matrix with column vector and store in dest column vector
  *
  * @param[in]  m    matrix (left)
  * @param[in]  v    vector (right, column vector)
  * @param[out] dest result vector
  */
 CGLM_INLINE
-vec2s
-glms_mat2x4_(mulv)(mat2x4s m, vec4s v) {
-  vec2s r;
+vec4s
+glms_mat2x4_(mulv)(mat2x4s m, vec2s v) {
+  vec4s r;
   glm_mat2x4_mulv(m.raw, v.raw, r.raw);
   return r;
 }
diff --git a/include/cglm/struct/mat3x2.h b/include/cglm/struct/mat3x2.h
index e9d077b..a95d5e1 100644
--- a/include/cglm/struct/mat3x2.h
+++ b/include/cglm/struct/mat3x2.h
@@ -13,8 +13,8 @@
  Functions:
    CGLM_INLINE mat3x2s glms_mat3x2_zero(void);
    CGLM_INLINE mat3x2s glms_mat3x2_make(const float * __restrict src);
-   CGLM_INLINE mat3s   glms_mat3x2_mul(mat3x2s m1, mat2x3s m2);
-   CGLM_INLINE vec3s   glms_mat3x2_mulv(mat3x2s m, vec2s v);
+   CGLM_INLINE mat2s   glms_mat3x2_mul(mat3x2s m1, mat2x3s m2);
+   CGLM_INLINE vec2s   glms_mat3x2_mulv(mat3x2s m, vec3s v);
    CGLM_INLINE mat2x3s glms_mat3x2_transpose(mat3x2s m);
    CGLM_INLINE mat3x2s glms_mat3x2_scale(mat3x2s m, float s);
  */
@@ -73,9 +73,9 @@ glms_mat3x2_(make)(const float * __restrict src) {
  * @returns destination matrix (mat3s)
  */
 CGLM_INLINE
-mat3s
+mat2s
 glms_mat3x2_(mul)(mat3x2s m1, mat2x3s m2) {
-  mat3s r;
+  mat2s r;
   glm_mat3x2_mul(m1.raw, m2.raw, r.raw);
   return r;
 }
@@ -88,9 +88,9 @@ glms_mat3x2_(mul)(mat3x2s m1, mat2x3s m2) {
  * @param[out] dest result vector
  */
 CGLM_INLINE
-vec3s
-glms_mat3x2_(mulv)(mat3x2s m, vec2s v) {
-  vec3s r;
+vec2s
+glms_mat3x2_(mulv)(mat3x2s m, vec3s v) {
+  vec2s r;
   glm_mat3x2_mulv(m.raw, v.raw, r.raw);
   return r;
 }
diff --git a/include/cglm/struct/mat3x4.h b/include/cglm/struct/mat3x4.h
index 29dd870..ee688f0 100644
--- a/include/cglm/struct/mat3x4.h
+++ b/include/cglm/struct/mat3x4.h
@@ -13,8 +13,8 @@
  Functions:
    CGLM_INLINE mat3x4s glms_mat3x4_zero(void);
    CGLM_INLINE mat3x4s glms_mat3x4_make(const float * __restrict src);
-   CGLM_INLINE mat3s   glms_mat3x4_mul(mat3x4s m1, mat4x3s m2);
-   CGLM_INLINE vec3s   glms_mat3x4_mulv(mat3x4s m, vec4s v);
+   CGLM_INLINE mat4s   glms_mat3x4_mul(mat3x4s m1, mat4x3s m2);
+   CGLM_INLINE vec4s   glms_mat3x4_mulv(mat3x4s m, vec3s v);
    CGLM_INLINE mat4x3s glms_mat3x4_transpose(mat3x4s m);
    CGLM_INLINE mat3x4s glms_mat3x4_scale(mat3x4s m, float s);
  */
@@ -70,27 +70,27 @@ glms_mat3x4_(make)(const float * __restrict src) {
  *
  * @param[in]  m1   left matrix (mat3x4s)
  * @param[in]  m2   right matrix (mat4x3s)
- * @returns destination matrix (mat3s)
+ * @returns destination matrix (mat4s)
  */
 CGLM_INLINE
-mat3s
+mat4s
 glms_mat3x4_(mul)(mat3x4s m1, mat4x3s m2) {
-  mat3s r;
+  mat4s r;
   glm_mat3x4_mul(m1.raw, m2.raw, r.raw);
   return r;
 }
 
 /*!
- * @brief multiply matrix with column vector and store in dest vector
+ * @brief multiply matrix with column vector and store in dest column vector
  *
  * @param[in]  m    matrix (left)
  * @param[in]  v    vector (right, column vector)
- * @param[out] dest result vector
+ * @returns destination vector (vec4s)
  */
 CGLM_INLINE
-vec3s
-glms_mat3x4_(mulv)(mat3x4s m, vec4s v) {
-  vec3s r;
+vec4s
+glms_mat3x4_(mulv)(mat3x4s m, vec3s v) {
+  vec4s r;
   glm_mat3x4_mulv(m.raw, v.raw, r.raw);
   return r;
 }
diff --git a/include/cglm/struct/mat4x2.h b/include/cglm/struct/mat4x2.h
index 3a0db6f..17589bc 100644
--- a/include/cglm/struct/mat4x2.h
+++ b/include/cglm/struct/mat4x2.h
@@ -13,8 +13,8 @@
  Functions:
    CGLM_INLINE mat4x2s glms_mat4x2_zero(void);
    CGLM_INLINE mat4x2s glms_mat4x2_make(const float * __restrict src);
-   CGLM_INLINE mat4s   glms_mat4x2_mul(mat4x2s m1, mat2x4s m2);
-   CGLM_INLINE vec4s   glms_mat4x2_mulv(mat4x2s m, vec2s v);
+   CGLM_INLINE mat2s   glms_mat4x2_mul(mat4x2s m1, mat2x4s m2);
+   CGLM_INLINE vec2s   glms_mat4x2_mulv(mat4x2s m, vec4s v);
    CGLM_INLINE mat2x4s glms_mat4x2_transpose(mat4x2s m);
    CGLM_INLINE mat4x2s glms_mat4x2_scale(mat4x2s m, float s);
  */
@@ -71,27 +71,27 @@ glms_mat4x2_(make)(const float * __restrict src) {
  *
  * @param[in]  m1   left matrix (mat4x2s)
  * @param[in]  m2   right matrix (mat2x4s)
- * @returns destination matrix (mat4s)
+ * @returns destination matrix (mat2s)
  */
 CGLM_INLINE
-mat4s
+mat2s
 glms_mat4x2_(mul)(mat4x2s m1, mat2x4s m2) {
-  mat4s r;
+  mat2s r;
   glm_mat4x2_mul(m1.raw, m2.raw, r.raw);
   return r;
 }
 
 /*!
- * @brief multiply matrix with column vector and store in dest vector
+ * @brief multiply matrix with column vector and store in dest column vector
  *
  * @param[in]  m    matrix (left)
  * @param[in]  v    vector (right, column vector)
- * @param[out] dest result vector
+ * @returns destination vector (vec2s)
  */
 CGLM_INLINE
-vec4s
-glms_mat4x2_(mulv)(mat4x2s m, vec2s v) {
-  vec4s r;
+vec2s
+glms_mat4x2_(mulv)(mat4x2s m, vec4s v) {
+  vec2s r;
   glm_mat4x2_mulv(m.raw, v.raw, r.raw);
   return r;
 }
diff --git a/include/cglm/struct/mat4x3.h b/include/cglm/struct/mat4x3.h
index f2cc7bb..37a68e1 100644
--- a/include/cglm/struct/mat4x3.h
+++ b/include/cglm/struct/mat4x3.h
@@ -13,8 +13,8 @@
  Functions:
    CGLM_INLINE mat4x3s glms_mat4x3_zero(void);
    CGLM_INLINE mat4x3s glms_mat4x3_make(const float * __restrict src);
-   CGLM_INLINE mat4s   glms_mat4x3_mul(mat4x3s m1, mat3x4s m2);
-   CGLM_INLINE vec4s   glms_mat4x3_mulv(mat4x3s m, vec3s v);
+   CGLM_INLINE mat3s   glms_mat4x3_mul(mat4x3s m1, mat3x4s m2);
+   CGLM_INLINE vec3s   glms_mat4x3_mulv(mat4x3s m, vec4s v);
    CGLM_INLINE mat3x4s glms_mat4x3_transpose(mat4x3s m);
    CGLM_INLINE mat4x3s glms_mat4x3_scale(mat4x3s m, float s);
  */
@@ -70,12 +70,12 @@ glms_mat4x3_(make)(const float * __restrict src) {
  *
  * @param[in]  m1   left matrix (mat4x3s)
  * @param[in]  m2   right matrix (mat3x4s)
- * @returns destination matrix (mat4s)
+ * @returns destination matrix (mat3s)
  */
 CGLM_INLINE
-mat4s
+mat3s
 glms_mat4x3_(mul)(mat4x3s m1, mat3x4s m2) {
-  mat4s r;
+  mat3s r;
   glm_mat4x3_mul(m1.raw, m2.raw, r.raw);
   return r;
 }
@@ -85,12 +85,12 @@ glms_mat4x3_(mul)(mat4x3s m1, mat3x4s m2) {
  *
  * @param[in]  m    matrix (left)
  * @param[in]  v    vector (right, column vector)
- * @param[out] dest result vector
+ * @returns destination vector (vec3s)
  */
 CGLM_INLINE
-vec4s
-glms_mat4x3_(mulv)(mat4x3s m, vec3s v) {
-  vec4s r;
+vec3s
+glms_mat4x3_(mulv)(mat4x3s m, vec4s v) {
+  vec3s r;
   glm_mat4x3_mulv(m.raw, v.raw, r.raw);
   return r;
 }
diff --git a/include/cglm/struct/vec2.h b/include/cglm/struct/vec2.h
index 3540d32..e1190e2 100644
--- a/include/cglm/struct/vec2.h
+++ b/include/cglm/struct/vec2.h
@@ -54,8 +54,8 @@
    CGLM_INLINE vec2s glms_vec2_clamp(vec2s v, float minVal, float maxVal)
    CGLM_INLINE vec2s glms_vec2_lerp(vec2s from, vec2s to, float t)
    CGLM_INLINE vec2s glms_vec2_make(float * restrict src)
-   CGLM_INLINE vec2s glms_vec2_reflect(vec2s I, vec2s N)
-   CGLM_INLINE bool  glms_vec2_refract(vec2s I, vec2s N, float eta, vec2s *dest)
+   CGLM_INLINE vec2s glms_vec2_reflect(vec2s v, vec2s n)
+   CGLM_INLINE bool  glms_vec2_refract(vec2s v, vec2s n, float eta, vec2s *dest)
  */
 
 #ifndef cglms_vec2s_h
@@ -702,9 +702,9 @@ glms_vec2_(make)(const float * __restrict src) {
  */
 CGLM_INLINE
 vec2s
-glms_vec2_(reflect)(vec2s I, vec2s N) {
+glms_vec2_(reflect)(vec2s v, vec2s n) {
   vec2s dest;
-  glm_vec2_reflect(I.raw, N.raw, dest.raw);
+  glm_vec2_reflect(v.raw, n.raw, dest.raw);
   return dest;
 }
 
@@ -715,8 +715,8 @@ glms_vec2_(reflect)(vec2s I, vec2s N) {
  * occurs (angle too great given eta), dest is set to zero and returns false.
  * Otherwise, computes refraction vector, stores it in dest, and returns true.
  *
- * @param[in]  I    normalized incident vector
- * @param[in]  N    normalized normal vector
+ * @param[in]  v    normalized incident vector
+ * @param[in]  n    normalized normal vector
  * @param[in]  eta  ratio of indices of refraction (incident/transmitted)
  * @param[out] dest refraction vector if refraction occurs; zero vector otherwise
  *
@@ -724,8 +724,8 @@ glms_vec2_(reflect)(vec2s I, vec2s N) {
  */
 CGLM_INLINE
 bool
-glms_vec2_(refract)(vec2s I, vec2s N, float eta, vec2s * __restrict dest) {
-  return glm_vec2_refract(I.raw, N.raw, eta, dest->raw);
+glms_vec2_(refract)(vec2s v, vec2s n, float eta, vec2s * __restrict dest) {
+  return glm_vec2_refract(v.raw, n.raw, eta, dest->raw);
 }
 
 #endif /* cglms_vec2s_h */
diff --git a/include/cglm/struct/vec3.h b/include/cglm/struct/vec3.h
index e1e0a29..536af03 100644
--- a/include/cglm/struct/vec3.h
+++ b/include/cglm/struct/vec3.h
@@ -76,9 +76,9 @@
    CGLM_INLINE vec3s glms_vec3_smoothinterpc(vec3s from, vec3s to, float t);
    CGLM_INLINE vec3s glms_vec3_swizzle(vec3s v, int mask);
    CGLM_INLINE vec3s glms_vec3_make(float * restrict src);
-   CGLM_INLINE vec3s glms_vec3_faceforward(vec3s N, vec3s I, vec3s Nref);
-   CGLM_INLINE vec3s glms_vec3_reflect(vec3s I, vec3s N);
-   CGLM_INLINE bool  glms_vec3_refract(vec3s I, vec3s N, float eta, vec3s *dest)
+   CGLM_INLINE vec3s glms_vec3_faceforward(vec3s n, vec3s v, vec3s nref);
+   CGLM_INLINE vec3s glms_vec3_reflect(vec3s v, vec3s n);
+   CGLM_INLINE bool  glms_vec3_refract(vec3s v, vec3s n, float eta, vec3s *dest)
 
  Convenient:
    CGLM_INLINE vec3s glms_cross(vec3s a, vec3s b);
@@ -1091,16 +1091,16 @@ glms_vec3_(make)(const float * __restrict src) {
  *
  * orients a vector to point away from a surface as defined by its normal
  *
- * @param[in] N      vector to orient.
- * @param[in] I      incident vector
- * @param[in] Nref   reference vector
+ * @param[in] n      vector to orient.
+ * @param[in] v      incident vector
+ * @param[in] nref   reference vector
  * @returns oriented vector, pointing away from the surface.
  */
 CGLM_INLINE
 vec3s
-glms_vec3_(faceforward)(vec3s N, vec3s I, vec3s Nref) {
+glms_vec3_(faceforward)(vec3s n, vec3s v, vec3s nref) {
   vec3s dest;
-  glm_vec3_faceforward(N.raw, I.raw, Nref.raw, dest.raw);
+  glm_vec3_faceforward(n.raw, v.raw, nref.raw, dest.raw);
   return dest;
 }
 
@@ -1113,9 +1113,9 @@ glms_vec3_(faceforward)(vec3s N, vec3s I, vec3s Nref) {
  */
 CGLM_INLINE
 vec3s
-glms_vec3_(reflect)(vec3s I, vec3s N) {
+glms_vec3_(reflect)(vec3s v, vec3s n) {
   vec3s dest;
-  glm_vec3_reflect(I.raw, N.raw, dest.raw);
+  glm_vec3_reflect(v.raw, n.raw, dest.raw);
   return dest;
 }
 
@@ -1126,8 +1126,8 @@ glms_vec3_(reflect)(vec3s I, vec3s N) {
  * occurs (angle too great given eta), dest is set to zero and returns false.
  * Otherwise, computes refraction vector, stores it in dest, and returns true.
  *
- * @param[in]  I    normalized incident vector
- * @param[in]  N    normalized normal vector
+ * @param[in]  v    normalized incident vector
+ * @param[in]  n    normalized normal vector
  * @param[in]  eta  ratio of indices of refraction (incident/transmitted)
  * @param[out] dest refraction vector if refraction occurs; zero vector otherwise
  *
@@ -1135,8 +1135,8 @@ glms_vec3_(reflect)(vec3s I, vec3s N) {
  */
 CGLM_INLINE
 bool
-glms_vec3_(refract)(vec3s I, vec3s N, float eta, vec3s * __restrict dest) {
-  return glm_vec3_refract(I.raw, N.raw, eta, dest->raw);
+glms_vec3_(refract)(vec3s v, vec3s n, float eta, vec3s * __restrict dest) {
+  return glm_vec3_refract(v.raw, n.raw, eta, dest->raw);
 }
 
 #endif /* cglms_vec3s_h */
diff --git a/include/cglm/struct/vec4.h b/include/cglm/struct/vec4.h
index 4431460..8b2ef00 100644
--- a/include/cglm/struct/vec4.h
+++ b/include/cglm/struct/vec4.h
@@ -67,8 +67,8 @@
    CGLM_INLINE vec4s glms_vec4_cubic(float s);
    CGLM_INLINE vec4s glms_vec4_swizzle(vec4s v, int mask);
    CGLM_INLINE vec4s glms_vec4_make(float * restrict src);
-   CGLM_INLINE vec4s glms_vec4_reflect(vec4s I, vec4s N);
-   CGLM_INLINE bool  glms_vec4_refract(vec4s I, vec4s N, float eta, vec4s *dest)
+   CGLM_INLINE vec4s glms_vec4_reflect(vec4s v, vec4s n);
+   CGLM_INLINE bool  glms_vec4_refract(vec4s v, vec4s n, float eta, vec4s *dest)
  */
 
 #ifndef cglms_vec4s_h
@@ -932,15 +932,15 @@ glms_vec4_(make)(const float * __restrict src) {
 /*!
  * @brief reflection vector using an incident ray and a surface normal
  *
- * @param[in]  I    incident vector
- * @param[in]  N    normalized normal vector
+ * @param[in]  v    incident vector
+ * @param[in]  n    normalized normal vector
  * @returns reflection result
  */
 CGLM_INLINE
 vec4s
-glms_vec4_(reflect)(vec4s I, vec4s N) {
+glms_vec4_(reflect)(vec4s v, vec4s n) {
   vec4s dest;
-  glm_vec4_reflect(I.raw, N.raw, dest.raw);
+  glm_vec4_reflect(v.raw, n.raw, dest.raw);
   return dest;
 }
 
@@ -955,8 +955,8 @@ glms_vec4_(reflect)(vec4s I, vec4s N) {
  * incident vector 'I' in the output 'dest', users requiring the preservation of
  * the 'w' component should manually adjust 'dest' after calling this function.
  *
- * @param[in]  I    normalized incident vector
- * @param[in]  N    normalized normal vector
+ * @param[in]  v    normalized incident vector
+ * @param[in]  n    normalized normal vector
  * @param[in]  eta  ratio of indices of refraction (incident/transmitted)
  * @param[out] dest refraction vector if refraction occurs; zero vector otherwise
  *
@@ -964,8 +964,8 @@ glms_vec4_(reflect)(vec4s I, vec4s N) {
  */
 CGLM_INLINE
 bool
-glms_vec4_(refract)(vec4s I, vec4s N, float eta, vec4s * __restrict dest) {
-  return glm_vec4_refract(I.raw, N.raw, eta, dest->raw);
+glms_vec4_(refract)(vec4s v, vec4s n, float eta, vec4s * __restrict dest) {
+  return glm_vec4_refract(v.raw, n.raw, eta, dest->raw);
 }
 
 #endif /* cglms_vec4s_h */
diff --git a/include/cglm/vec2.h b/include/cglm/vec2.h
index aeb9e91..aaac759 100644
--- a/include/cglm/vec2.h
+++ b/include/cglm/vec2.h
@@ -55,8 +55,8 @@
    CGLM_INLINE void  glm_vec2_clamp(vec2 v, float minVal, float maxVal)
    CGLM_INLINE void  glm_vec2_lerp(vec2 from, vec2 to, float t, vec2 dest)
    CGLM_INLINE void  glm_vec2_make(float * restrict src, vec2 dest)
-   CGLM_INLINE void  glm_vec2_reflect(vec2 I, vec2 N, vec2 dest)
-   CGLM_INLINE void  glm_vec2_refract(vec2 I, vec2 N, float eta, vec2 dest)
+   CGLM_INLINE void  glm_vec2_reflect(vec2 v, vec2 n, vec2 dest)
+   CGLM_INLINE void  glm_vec2_refract(vec2 v, vec2 n, float eta, vec2 dest)
  */
 
 #ifndef cglm_vec2_h
@@ -716,16 +716,16 @@ glm_vec2_make(const float * __restrict src, vec2 dest) {
 /*!
  * @brief reflection vector using an incident ray and a surface normal
  *
- * @param[in]  I    incident vector
- * @param[in]  N    normalized normal vector
+ * @param[in]  v    incident vector
+ * @param[in]  n    normalized normal vector
  * @param[out] dest destination vector for the reflection result
  */
 CGLM_INLINE
 void
-glm_vec2_reflect(vec2 I, vec2 N, vec2 dest) {
+glm_vec2_reflect(vec2 v, vec2 n, vec2 dest) {
   vec2 temp;
-  glm_vec2_scale(N, 2.0f * glm_vec2_dot(I, N), temp);
-  glm_vec2_sub(I, temp, dest);
+  glm_vec2_scale(n, 2.0f * glm_vec2_dot(v, n), temp);
+  glm_vec2_sub(v, temp, dest);
 }
 
 /*!
@@ -735,8 +735,8 @@ glm_vec2_reflect(vec2 I, vec2 N, vec2 dest) {
  * occurs (angle too great given eta), dest is set to zero and returns false.
  * Otherwise, computes refraction vector, stores it in dest, and returns true.
  *
- * @param[in]  I    normalized incident vector
- * @param[in]  N    normalized normal vector
+ * @param[in]  v    normalized incident vector
+ * @param[in]  n    normalized normal vector
  * @param[in]  eta  ratio of indices of refraction (incident/transmitted)
  * @param[out] dest refraction vector if refraction occurs; zero vector otherwise
  *
@@ -744,10 +744,10 @@ glm_vec2_reflect(vec2 I, vec2 N, vec2 dest) {
  */
 CGLM_INLINE
 bool
-glm_vec2_refract(vec2 I, vec2 N, float eta, vec2 dest) {
+glm_vec2_refract(vec2 v, vec2 n, float eta, vec2 dest) {
   float ndi, eni, k;
 
-  ndi = glm_vec2_dot(N, I);
+  ndi = glm_vec2_dot(n, v);
   eni = eta * ndi;
   k   = 1.0f + eta * eta - eni * eni;
 
@@ -756,8 +756,8 @@ glm_vec2_refract(vec2 I, vec2 N, float eta, vec2 dest) {
     return false;
   }
 
-  glm_vec2_scale(I, eta, dest);
-  glm_vec2_mulsubs(N, eni + sqrtf(k), dest);
+  glm_vec2_scale(v, eta, dest);
+  glm_vec2_mulsubs(n, eni + sqrtf(k), dest);
   return true;
 }
 
diff --git a/include/cglm/vec3.h b/include/cglm/vec3.h
index 72bd371..be4029b 100644
--- a/include/cglm/vec3.h
+++ b/include/cglm/vec3.h
@@ -80,9 +80,9 @@
    CGLM_INLINE void  glm_vec3_smoothinterpc(vec3 from, vec3 to, float t, vec3 dest);
    CGLM_INLINE void  glm_vec3_swizzle(vec3 v, int mask, vec3 dest);
    CGLM_INLINE void  glm_vec3_make(float * restrict src, vec3 dest);
-   CGLM_INLINE void  glm_vec3_faceforward(vec3 N, vec3 I, vec3 Nref, vec3 dest);
-   CGLM_INLINE void  glm_vec3_reflect(vec3 I, vec3 N, vec3 dest);
-   CGLM_INLINE void  glm_vec3_refract(vec3 I, vec3 N, float eta, vec3 dest);
+   CGLM_INLINE void  glm_vec3_faceforward(vec3 n, vec3 v, vec3 nref, vec3 dest);
+   CGLM_INLINE void  glm_vec3_reflect(vec3 v, vec3 n, vec3 dest);
+   CGLM_INLINE void  glm_vec3_refract(vec3 v, vec3 n, float eta, vec3 dest);
 
  Convenient:
    CGLM_INLINE void  glm_cross(vec3 a, vec3 b, vec3 d);
@@ -1210,36 +1210,36 @@ glm_vec3_make(const float * __restrict src, vec3 dest) {
  *
  * orients a vector to point away from a surface as defined by its normal
  *
- * @param[in] N      vector to orient
- * @param[in] I      incident vector
- * @param[in] Nref   reference vector
+ * @param[in] n      vector to orient
+ * @param[in] v      incident vector
+ * @param[in] nref   reference vector
  * @param[out] dest  oriented vector, pointing away from the surface
  */
 CGLM_INLINE
 void
-glm_vec3_faceforward(vec3 N, vec3 I, vec3 Nref, vec3 dest) {
-  if (glm_vec3_dot(I, Nref) < 0.0f) {
+glm_vec3_faceforward(vec3 n, vec3 v, vec3 nref, vec3 dest) {
+  if (glm_vec3_dot(v, nref) < 0.0f) {
     /* N is facing away from I */
-    glm_vec3_copy(N, dest);
+    glm_vec3_copy(n, dest);
   } else {
     /* N is facing towards I, negate it */
-    glm_vec3_negate_to(N, dest);
+    glm_vec3_negate_to(n, dest);
   }
 }
 
 /*!
  * @brief reflection vector using an incident ray and a surface normal
  *
- * @param[in]  I    incident vector
- * @param[in]  N    normalized normal vector
+ * @param[in]  v    incident vector
+ * @param[in]  n    normalized normal vector
  * @param[out] dest reflection result
  */
 CGLM_INLINE
 void
-glm_vec3_reflect(vec3 I, vec3 N, vec3 dest) {
+glm_vec3_reflect(vec3 v, vec3 n, vec3 dest) {
   vec3 temp;
-  glm_vec3_scale(N, 2.0f * glm_vec3_dot(I, N), temp);
-  glm_vec3_sub(I, temp, dest);
+  glm_vec3_scale(n, 2.0f * glm_vec3_dot(v, n), temp);
+  glm_vec3_sub(v, temp, dest);
 }
 
 /*!
@@ -1249,8 +1249,8 @@ glm_vec3_reflect(vec3 I, vec3 N, vec3 dest) {
  * occurs (angle too great given eta), dest is set to zero and returns false.
  * Otherwise, computes refraction vector, stores it in dest, and returns true.
  *
- * @param[in]  I    normalized incident vector
- * @param[in]  N    normalized normal vector
+ * @param[in]  v    normalized incident vector
+ * @param[in]  n    normalized normal vector
  * @param[in]  eta  ratio of indices of refraction (incident/transmitted)
  * @param[out] dest refraction vector if refraction occurs; zero vector otherwise
  *
@@ -1258,10 +1258,10 @@ glm_vec3_reflect(vec3 I, vec3 N, vec3 dest) {
  */
 CGLM_INLINE
 bool
-glm_vec3_refract(vec3 I, vec3 N, float eta, vec3 dest) {
+glm_vec3_refract(vec3 v, vec3 n, float eta, vec3 dest) {
   float ndi, eni, k;
 
-  ndi = glm_vec3_dot(N, I);
+  ndi = glm_vec3_dot(n, v);
   eni = eta * ndi;
   k   = 1.0f + eta * eta - eni * eni;
 
@@ -1270,8 +1270,8 @@ glm_vec3_refract(vec3 I, vec3 N, float eta, vec3 dest) {
     return false;
   }
 
-  glm_vec3_scale(I, eta, dest);
-  glm_vec3_mulsubs(N, eni + sqrtf(k), dest);
+  glm_vec3_scale(v, eta, dest);
+  glm_vec3_mulsubs(n, eni + sqrtf(k), dest);
   return true;
 }
 
diff --git a/include/cglm/vec4.h b/include/cglm/vec4.h
index 4053b9f..e24675f 100644
--- a/include/cglm/vec4.h
+++ b/include/cglm/vec4.h
@@ -65,8 +65,8 @@
    CGLM_INLINE void  glm_vec4_smoothinterpc(vec4 from, vec4 to, float t, vec4 dest);
    CGLM_INLINE void  glm_vec4_swizzle(vec4 v, int mask, vec4 dest);
    CGLM_INLINE void  glm_vec4_make(float * restrict src, vec4 dest);
-   CGLM_INLINE void  glm_vec4_reflect(vec4 I, vec4 N, vec4 dest);
-   CGLM_INLINE void  glm_vec4_refract(vec4 I, vec4 N, float eta, vec4 dest);
+   CGLM_INLINE void  glm_vec4_reflect(vec4 v, vec4 n, vec4 dest);
+   CGLM_INLINE void  glm_vec4_refract(vec4 v, vec4 n, float eta, vec4 dest);
 
  DEPRECATED:
    glm_vec4_dup
@@ -1309,20 +1309,20 @@ glm_vec4_make(const float * __restrict src, vec4 dest) {
 /*!
  * @brief reflection vector using an incident ray and a surface normal
  *
- * @param[in]  I    incident vector
- * @param[in]  N    normalized normal vector
+ * @param[in]  v    incident vector
+ * @param[in]  n    normalized normal vector
  * @param[out] dest destination vector for the reflection result
  */
 CGLM_INLINE
 void
-glm_vec4_reflect(vec4 I, vec4 N, vec4 dest) {
+glm_vec4_reflect(vec4 v, vec4 n, vec4 dest) {
   vec4 temp;
 
   /* TODO: direct simd touch */
-  glm_vec4_scale(N, 2.0f * glm_vec4_dot(I, N), temp);
-  glm_vec4_sub(I, temp, dest);
+  glm_vec4_scale(n, 2.0f * glm_vec4_dot(v, n), temp);
+  glm_vec4_sub(v, temp, dest);
 
-  dest[3] = I[3];
+  dest[3] = v[3];
 }
 
 /*!
@@ -1336,8 +1336,8 @@ glm_vec4_reflect(vec4 I, vec4 N, vec4 dest) {
  * incident vector 'I' in the output 'dest', users requiring the preservation of
  * the 'w' component should manually adjust 'dest' after calling this function.
  *
- * @param[in]  I    normalized incident vector
- * @param[in]  N    normalized normal vector
+ * @param[in]  v    normalized incident vector
+ * @param[in]  n    normalized normal vector
  * @param[in]  eta  ratio of indices of refraction (incident/transmitted)
  * @param[out] dest refraction vector if refraction occurs; zero vector otherwise
  *
@@ -1345,10 +1345,10 @@ glm_vec4_reflect(vec4 I, vec4 N, vec4 dest) {
  */
 CGLM_INLINE
 bool
-glm_vec4_refract(vec4 I, vec4 N, float eta, vec4 dest) {
+glm_vec4_refract(vec4 v, vec4 n, float eta, vec4 dest) {
   float ndi, eni, k;
 
-  ndi = glm_vec4_dot(N, I);
+  ndi = glm_vec4_dot(n, v);
   eni = eta * ndi;
   k   = 1.0f + eta * eta - eni * eni;
 
@@ -1357,8 +1357,8 @@ glm_vec4_refract(vec4 I, vec4 N, float eta, vec4 dest) {
     return false;
   }
 
-  glm_vec4_scale(I, eta, dest);
-  glm_vec4_mulsubs(N, eni + sqrtf(k), dest);
+  glm_vec4_scale(v, eta, dest);
+  glm_vec4_mulsubs(n, eni + sqrtf(k), dest);
   return true;
 }
 
diff --git a/include/cglm/version.h b/include/cglm/version.h
index d8a1111..704fea0 100644
--- a/include/cglm/version.h
+++ b/include/cglm/version.h
@@ -10,6 +10,6 @@
 
 #define CGLM_VERSION_MAJOR 0
 #define CGLM_VERSION_MINOR 9
-#define CGLM_VERSION_PATCH 4
+#define CGLM_VERSION_PATCH 5
 
 #endif /* cglm_version_h */
diff --git a/meson.build b/meson.build
index 29ec0c0..3afdc4f 100644
--- a/meson.build
+++ b/meson.build
@@ -1,5 +1,5 @@
 project('cglm', 'c',
-    version : '0.9.4',
+    version : '0.9.5',
     license : 'mit',
     default_options : [
         'c_std=c11',
diff --git a/src/mat2x3.c b/src/mat2x3.c
index dc5fd30..3397bf2 100644
--- a/src/mat2x3.c
+++ b/src/mat2x3.c
@@ -28,13 +28,13 @@ glmc_mat2x3_make(const float * __restrict src, mat2x3 dest) {
 
 CGLM_EXPORT
 void
-glmc_mat2x3_mul(mat2x3 m1, mat3x2 m2, mat2 dest) {
+glmc_mat2x3_mul(mat2x3 m1, mat3x2 m2, mat3 dest) {
   glm_mat2x3_mul(m1, m2, dest);
 }
 
 CGLM_EXPORT
 void
-glmc_mat2x3_mulv(mat2x3 m, vec3 v, vec2 dest) {
+glmc_mat2x3_mulv(mat2x3 m, vec2 v, vec3 dest) {
   glm_mat2x3_mulv(m, v, dest);
 }
 
diff --git a/src/mat2x4.c b/src/mat2x4.c
index 73db032..a221ed3 100644
--- a/src/mat2x4.c
+++ b/src/mat2x4.c
@@ -28,13 +28,13 @@ glmc_mat2x4_make(const float * __restrict src, mat2x4 dest) {
 
 CGLM_EXPORT
 void
-glmc_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat2 dest) {
+glmc_mat2x4_mul(mat2x4 m1, mat4x2 m2, mat4 dest) {
   glm_mat2x4_mul(m1, m2, dest);
 }
 
 CGLM_EXPORT
 void
-glmc_mat2x4_mulv(mat2x4 m, vec4 v, vec2 dest) {
+glmc_mat2x4_mulv(mat2x4 m, vec2 v, vec4 dest) {
   glm_mat2x4_mulv(m, v, dest);
 }
 
diff --git a/src/mat3x2.c b/src/mat3x2.c
index 8517a9e..5ade303 100644
--- a/src/mat3x2.c
+++ b/src/mat3x2.c
@@ -28,13 +28,13 @@ glmc_mat3x2_make(const float * __restrict src, mat3x2 dest) {
 
 CGLM_EXPORT
 void
-glmc_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat3 dest) {
+glmc_mat3x2_mul(mat3x2 m1, mat2x3 m2, mat2 dest) {
   glm_mat3x2_mul(m1, m2, dest);
 }
 
 CGLM_EXPORT
 void
-glmc_mat3x2_mulv(mat3x2 m, vec2 v, vec3 dest) {
+glmc_mat3x2_mulv(mat3x2 m, vec3 v, vec2 dest) {
   glm_mat3x2_mulv(m, v, dest);
 }
 
diff --git a/src/mat3x4.c b/src/mat3x4.c
index efc9062..75963c9 100644
--- a/src/mat3x4.c
+++ b/src/mat3x4.c
@@ -28,13 +28,13 @@ glmc_mat3x4_make(const float * __restrict src, mat3x4 dest) {
 
 CGLM_EXPORT
 void
-glmc_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat3 dest) {
+glmc_mat3x4_mul(mat3x4 m1, mat4x3 m2, mat4 dest) {
   glm_mat3x4_mul(m1, m2, dest);
 }
 
 CGLM_EXPORT
 void
-glmc_mat3x4_mulv(mat3x4 m, vec4 v, vec3 dest) {
+glmc_mat3x4_mulv(mat3x4 m, vec3 v, vec4 dest) {
   glm_mat3x4_mulv(m, v, dest);
 }
 
diff --git a/src/mat4x2.c b/src/mat4x2.c
index 2edd2e0..32015f2 100644
--- a/src/mat4x2.c
+++ b/src/mat4x2.c
@@ -28,13 +28,13 @@ glmc_mat4x2_make(const float * __restrict src, mat4x2 dest) {
 
 CGLM_EXPORT
 void
-glmc_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat4 dest) {
+glmc_mat4x2_mul(mat4x2 m1, mat2x4 m2, mat2 dest) {
   glm_mat4x2_mul(m1, m2, dest);
 }
 
 CGLM_EXPORT
 void
-glmc_mat4x2_mulv(mat4x2 m, vec2 v, vec4 dest) {
+glmc_mat4x2_mulv(mat4x2 m, vec4 v, vec2 dest) {
   glm_mat4x2_mulv(m, v, dest);
 }
 
diff --git a/src/mat4x3.c b/src/mat4x3.c
index a50b983..8285424 100644
--- a/src/mat4x3.c
+++ b/src/mat4x3.c
@@ -28,13 +28,13 @@ glmc_mat4x3_make(const float * __restrict src, mat4x3 dest) {
 
 CGLM_EXPORT
 void
-glmc_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat4 dest) {
+glmc_mat4x3_mul(mat4x3 m1, mat3x4 m2, mat3 dest) {
   glm_mat4x3_mul(m1, m2, dest);
 }
 
 CGLM_EXPORT
 void
-glmc_mat4x3_mulv(mat4x3 m, vec3 v, vec4 dest) {
+glmc_mat4x3_mulv(mat4x3 m, vec4 v, vec3 dest) {
   glm_mat4x3_mulv(m, v, dest);
 }
 
diff --git a/src/vec2.c b/src/vec2.c
index 4f638a1..b124f86 100644
--- a/src/vec2.c
+++ b/src/vec2.c
@@ -305,12 +305,12 @@ glmc_vec2_make(const float * __restrict src, vec2 dest) {
 
 CGLM_EXPORT
 void
-glmc_vec2_reflect(vec2 I, vec2 N, vec2 dest) {
-  glm_vec2_reflect(I, N, dest);
+glmc_vec2_reflect(vec2 v, vec2 n, vec2 dest) {
+  glm_vec2_reflect(v, n, dest);
 }
 
 CGLM_EXPORT
 bool
-glmc_vec2_refract(vec2 I, vec2 N, float eta, vec2 dest) {
-  return glm_vec2_refract(I, N, eta, dest);
+glmc_vec2_refract(vec2 v, vec2 n, float eta, vec2 dest) {
+  return glm_vec2_refract(v, n, eta, dest);
 }
diff --git a/src/vec3.c b/src/vec3.c
index a31ce3f..c1316dc 100644
--- a/src/vec3.c
+++ b/src/vec3.c
@@ -462,18 +462,18 @@ glmc_vec3_make(const float * __restrict src, vec3 dest) {
 
 CGLM_EXPORT
 void
-glmc_vec3_faceforward(vec3 N, vec3 I, vec3 Nref, vec3 dest) {
-  glm_vec3_faceforward(N, I, Nref, dest);
+glmc_vec3_faceforward(vec3 n, vec3 v, vec3 nref, vec3 dest) {
+  glm_vec3_faceforward(n, v, nref, dest);
 }
 
 CGLM_EXPORT
 void
-glmc_vec3_reflect(vec3 I, vec3 N, vec3 dest) {
-  glm_vec3_reflect(I, N, dest);
+glmc_vec3_reflect(vec3 v, vec3 n, vec3 dest) {
+  glm_vec3_reflect(v, n, dest);
 }
 
 CGLM_EXPORT
 bool
-glmc_vec3_refract(vec3 I, vec3 N, float eta, vec3 dest) {
-  return glm_vec3_refract(I, N, eta, dest);
+glmc_vec3_refract(vec3 v, vec3 n, float eta, vec3 dest) {
+  return glm_vec3_refract(v, n, eta, dest);
 }
diff --git a/src/vec4.c b/src/vec4.c
index de23462..cac6606 100644
--- a/src/vec4.c
+++ b/src/vec4.c
@@ -426,12 +426,12 @@ glmc_vec4_make(const float * __restrict src, vec4 dest) {
 
 CGLM_EXPORT
 void
-glmc_vec4_reflect(vec4 I, vec4 N, vec4 dest) {
-  glm_vec4_reflect(I, N, dest);
+glmc_vec4_reflect(vec4 v, vec4 n, vec4 dest) {
+  glm_vec4_reflect(v, n, dest);
 }
 
 CGLM_EXPORT
 bool
-glmc_vec4_refract(vec4 I, vec4 N, float eta, vec4 dest) {
-  return glm_vec4_refract(I, N, eta, dest);
+glmc_vec4_refract(vec4 v, vec4 n, float eta, vec4 dest) {
+  return glm_vec4_refract(v, n, eta, dest);
 }
diff --git a/test/src/test_mat2x3.h b/test/src/test_mat2x3.h
index 7d615e4..8eb8144 100644
--- a/test/src/test_mat2x3.h
+++ b/test/src/test_mat2x3.h
@@ -83,42 +83,42 @@ TEST_IMPL(GLM_PREFIX, mat2x3_mul) {
   mat2x3 m1 = GLM_MAT2X3_ZERO_INIT;
   mat3x2 m2 = GLM_MAT3X2_ZERO_INIT;
 
-  mat2 m3 = GLM_MAT2_ZERO_INIT;
-  mat2 m4 = GLM_MAT2_ZERO_INIT;
+  mat3 m3 = GLM_MAT3_ZERO_INIT;
+  mat3 m4 = GLM_MAT3_ZERO_INIT;
 
-  int i, j, k;
+  int c, r, k;
 
   /* test random matrices */
   /* random matrices */
   test_rand_mat2x3(m1);
   test_rand_mat3x2(m2);
 
-  for (i = 0; i < 2; i++) {
-    for (j = 0; j < 2; j++) {
-      for (k = 0; k < 3; k++) {
-        m4[i][j] += m1[i][k] * m2[k][j];
+  for (r = 0; r < 3; r++) {
+    for (c = 0; c < 3; c++) {
+      for (k = 0; k < 2; k++) {
+        m4[c][r] += m1[k][r] * m2[c][k];
       }
     }
   }
 
   GLM(mat2x3_mul)(m1, m2, m3);
-  ASSERTIFY(test_assert_mat2_eq(m3, m4))
+  ASSERTIFY(test_assert_mat3_eq(m3, m4))
 
   TEST_SUCCESS
 }
 
 TEST_IMPL(GLM_PREFIX, mat2x3_mulv) {
   mat2x3 mat = A_MATRIX2X3;
-  vec3 v = {11.0f, 21.0f, 31.0f};
+  vec2 v = {11.0f, 21.0f};
 
   int  i;
-  vec2 dest;
+  vec3 dest;
   float res = 0.0;
 
   GLM(mat2x3_mulv)(mat, v, dest);
 
-  for (i = 0; i < 2; i++) {
-    res = mat[i][0] * v[0] + mat[i][1] * v[1] + mat[i][2] * v[2];
+  for (i = 0; i < 3; i++) {
+    res = mat[0][i] * v[0] + mat[1][i] * v[1];
     ASSERT(test_eq(dest[i], res))
   }
 
diff --git a/test/src/test_mat2x4.h b/test/src/test_mat2x4.h
index ea594d1..17ee35f 100644
--- a/test/src/test_mat2x4.h
+++ b/test/src/test_mat2x4.h
@@ -86,42 +86,42 @@ TEST_IMPL(GLM_PREFIX, mat2x4_mul) {
   mat2x4 m1 = GLM_MAT2X4_ZERO_INIT;
   mat4x2 m2 = GLM_MAT4X2_ZERO_INIT;
 
-  mat2 m3 = GLM_MAT2_ZERO_INIT;
-  mat2 m4 = GLM_MAT2_ZERO_INIT;
+  mat4 m3 = GLM_MAT4_ZERO_INIT;
+  mat4 m4 = GLM_MAT4_ZERO_INIT;
 
-  int i, j, k;
+  int c, r, k;
 
   /* test random matrices */
   /* random matrices */
   test_rand_mat2x4(m1);
   test_rand_mat4x2(m2);
 
-  for (i = 0; i < 2; i++) {
-    for (j = 0; j < 2; j++) {
-      for (k = 0; k < 4; k++) {
-        m4[i][j] += m1[i][k] * m2[k][j];
+  for (r = 0; r < 4; r++) {
+    for (c = 0; c < 4; c++) {
+      for (k = 0; k < 2; k++) {
+        m4[c][r] += m1[k][r] * m2[c][k];
       }
     }
   }
 
   GLM(mat2x4_mul)(m1, m2, m3);
-  ASSERTIFY(test_assert_mat2_eq(m3, m4))
+  ASSERTIFY(test_assert_mat4_eq(m3, m4))
 
   TEST_SUCCESS
 }
 
 TEST_IMPL(GLM_PREFIX, mat2x4_mulv) {
   mat2x4 mat = A_MATRIX2X4;
-  vec4 v = {11.0f, 21.0f, 31.0f, 41.0f};
+  vec2 v = {11.0f, 21.0f};
 
   int  i;
-  vec2 dest;
+  vec4 dest;
   float res = 0.0;
 
   GLM(mat2x4_mulv)(mat, v, dest);
 
-  for (i = 0; i < 2; i++) {
-    res = mat[i][0] * v[0] + mat[i][1] * v[1] + mat[i][2] * v[2] + mat[i][3] * v[3];
+  for (i = 0; i < 4; i++) {
+    res = mat[0][i] * v[0] + mat[1][i] * v[1];
     ASSERT(test_eq(dest[i], res))
   }
 
diff --git a/test/src/test_mat3x2.h b/test/src/test_mat3x2.h
index a825a2b..397879b 100644
--- a/test/src/test_mat3x2.h
+++ b/test/src/test_mat3x2.h
@@ -84,40 +84,40 @@ TEST_IMPL(GLM_PREFIX, mat3x2_mul) {
   mat3x2 m1 = GLM_MAT3X2_ZERO_INIT;
   mat2x3 m2 = GLM_MAT2X3_ZERO_INIT;
 
-  mat3 m3 = GLM_MAT3_ZERO_INIT;
-  mat3 m4 = GLM_MAT3_ZERO_INIT;
+  mat2 m3 = GLM_MAT2_ZERO_INIT;
+  mat2 m4 = GLM_MAT2_ZERO_INIT;
 
-  int i, j, k;
+  int c, r, k;
 
   test_rand_mat3x2(m1);
   test_rand_mat2x3(m2);
 
-  for (i = 0; i < 3; i++) {
-    for (j = 0; j < 3; j++) {
-      for (k = 0; k < 2; k++) {
-        m4[i][j] += m1[i][k] * m2[k][j];
+  for (r = 0; r < 2; r++) {
+    for (c = 0; c < 2; c++) {
+      for (k = 0; k < 3; k++) {
+        m4[c][r] += m1[k][r] * m2[c][k];
       }
     }
   }
 
   GLM(mat3x2_mul)(m1, m2, m3);
-  ASSERTIFY(test_assert_mat3_eq(m3, m4))
+  ASSERTIFY(test_assert_mat2_eq(m3, m4))
 
   TEST_SUCCESS
 }
 
 TEST_IMPL(GLM_PREFIX, mat3x2_mulv) {
   mat3x2 mat = A_MATRIX3X2;
-  vec2 v = {11.0f, 21.0f};
+  vec3 v = {11.0f, 21.0f, 31.0f};
 
   int  i;
-  vec3 dest;
+  vec2 dest;
   float res = 0.0;
 
   GLM(mat3x2_mulv)(mat, v, dest);
 
-  for (i = 0; i < 3; i++) {
-    res = mat[i][0] * v[0] + mat[i][1] * v[1];
+  for (i = 0; i < 2; i++) {
+    res = mat[0][i] * v[0] + mat[1][i] * v[1] + mat[2][i] * v[2];
     ASSERT(test_eq(dest[i], res))
   }
 
diff --git a/test/src/test_mat3x4.h b/test/src/test_mat3x4.h
index c7d704c..16520ac 100644
--- a/test/src/test_mat3x4.h
+++ b/test/src/test_mat3x4.h
@@ -90,40 +90,40 @@ TEST_IMPL(GLM_PREFIX, mat3x4_mul) {
   mat3x4 m1 = GLM_MAT3X4_ZERO_INIT;
   mat4x3 m2 = GLM_MAT4X3_ZERO_INIT;
 
-  mat3 m3 = GLM_MAT3_ZERO_INIT;
-  mat3 m4 = GLM_MAT3_ZERO_INIT;
+  mat4 m3 = GLM_MAT4_ZERO_INIT;
+  mat4 m4 = GLM_MAT4_ZERO_INIT;
 
-  int i, j, k;
+  int c, r, k;
 
   test_rand_mat3x4(m1);
   test_rand_mat4x3(m2);
 
-  for (i = 0; i < 3; i++) {
-    for (j = 0; j < 3; j++) {
-      for (k = 0; k < 4; k++) {
-        m4[i][j] += m1[i][k] * m2[k][j];
+  for (r = 0; r < 4; r++) {
+    for (c = 0; c < 4; c++) {
+      for (k = 0; k < 3; k++) {
+        m4[c][r] += m1[k][r] * m2[c][k];
       }
     }
   }
 
   GLM(mat3x4_mul)(m1, m2, m3);
-  ASSERTIFY(test_assert_mat3_eq(m3, m4))
+  ASSERTIFY(test_assert_mat4_eq(m3, m4))
 
   TEST_SUCCESS
 }
 
 TEST_IMPL(GLM_PREFIX, mat3x4_mulv) {
   mat3x4 mat = A_MATRIX3X4;
-  vec4 v = {11.0f, 21.0f, 31.0f, 41.0f};
+  vec4 v = {11.0f, 21.0f, 31.0f};
 
   int  i;
-  vec3 dest;
+  vec4 dest;
   float res = 0.0;
 
   GLM(mat3x4_mulv)(mat, v, dest);
 
-  for (i = 0; i < 3; i++) {
-    res = mat[i][0] * v[0] + mat[i][1] * v[1] + mat[i][2] * v[2];
+  for (i = 0; i < 4; i++) {
+    res = mat[0][i] * v[0] + mat[1][i] * v[1] + mat[2][i] * v[2];
     ASSERT(test_eq(dest[i], res))
   }
 
diff --git a/test/src/test_mat4x2.h b/test/src/test_mat4x2.h
index fd1ac48..b58d974 100644
--- a/test/src/test_mat4x2.h
+++ b/test/src/test_mat4x2.h
@@ -87,40 +87,40 @@ TEST_IMPL(GLM_PREFIX, mat4x2_mul) {
   mat4x2 m1 = GLM_MAT4X2_ZERO_INIT;
   mat2x4 m2 = GLM_MAT2X4_ZERO_INIT;
 
-  mat4 m3 = GLM_MAT4_ZERO_INIT;
-  mat4 m4 = GLM_MAT4_ZERO_INIT;
+  mat2 m3 = GLM_MAT2_ZERO_INIT;
+  mat2 m4 = GLM_MAT2_ZERO_INIT;
 
-  int i, j, k;
+  int c, r, k;
 
   test_rand_mat4x2(m1);
   test_rand_mat2x4(m2);
 
-  for (i = 0; i < 4; i++) {
-    for (j = 0; j < 4; j++) {
-      for (k = 0; k < 2; k++) {
-        m4[i][j] += m1[i][k] * m2[k][j];
+  for (r = 0; r < 2; r++) {
+    for (c = 0; c < 2; c++) {
+      for (k = 0; k < 4; k++) {
+	m4[c][r] += m1[k][r] * m2[c][k];
       }
     }
   }
 
   GLM(mat4x2_mul)(m1, m2, m3);
-  ASSERTIFY(test_assert_mat4_eq(m3, m4))
+  ASSERTIFY(test_assert_mat2_eq(m3, m4))
 
   TEST_SUCCESS
 }
 
 TEST_IMPL(GLM_PREFIX, mat4x2_mulv) {
   mat4x2 mat = A_MATRIX4X2;
-  vec2 v = {11.0f, 21.0f};
+  vec4 v = {11.0f, 21.0f, 31.0f, 41.0f};
 
   int  i;
-  vec4 dest;
+  vec2 dest;
   float res = 0.0;
 
   GLM(mat4x2_mulv)(mat, v, dest);
 
-  for (i = 0; i < 4; i++) {
-    res = mat[i][0] * v[0] + mat[i][1] * v[1];
+  for (i = 0; i < 2; i++) {
+    res = mat[0][i] * v[0] + mat[1][i] * v[1] + mat[2][i] * v[2] + mat[3][i] * v[3];
     ASSERT(test_eq(dest[i], res))
   }
 
diff --git a/test/src/test_mat4x3.h b/test/src/test_mat4x3.h
index b760405..dc7c28e 100644
--- a/test/src/test_mat4x3.h
+++ b/test/src/test_mat4x3.h
@@ -91,40 +91,40 @@ TEST_IMPL(GLM_PREFIX, mat4x3_mul) {
   mat4x3 m1 = GLM_MAT4X3_ZERO_INIT;
   mat3x4 m2 = GLM_MAT3X4_ZERO_INIT;
 
-  mat4 m3 = GLM_MAT4_ZERO_INIT;
-  mat4 m4 = GLM_MAT4_ZERO_INIT;
+  mat3 m3 = GLM_MAT3_ZERO_INIT;
+  mat3 m4 = GLM_MAT3_ZERO_INIT;
 
-  int i, j, k;
+  int c, r, k;
 
   test_rand_mat4x3(m1);
   test_rand_mat3x4(m2);
 
-  for (i = 0; i < 4; i++) {
-    for (j = 0; j < 4; j++) {
-      for (k = 0; k < 3; k++) {
-        m4[i][j] += m1[i][k] * m2[k][j];
+  for (r = 0; r < 3; r++) {
+    for (c = 0; c < 3; c++) {
+      for (k = 0; k < 4; k++) {
+        m4[c][r] += m1[k][r] * m2[c][k];
       }
     }
   }
 
   GLM(mat4x3_mul)(m1, m2, m3);
-  ASSERTIFY(test_assert_mat4_eq(m3, m4))
+  ASSERTIFY(test_assert_mat3_eq(m3, m4))
 
   TEST_SUCCESS
 }
 
 TEST_IMPL(GLM_PREFIX, mat4x3_mulv) {
   mat4x3 mat = A_MATRIX4X3;
-  vec3 v = {11.0f, 21.0f, 31.0f};
+  vec4 v = {11.0f, 21.0f, 31.0f, 41.0f};
 
   int  i;
-  vec4 dest;
+  vec3 dest;
   float res = 0.0;
 
   GLM(mat4x3_mulv)(mat, v, dest);
 
-  for (i = 0; i < 4; i++) {
-    res = mat[i][0] * v[0] + mat[i][1] * v[1] + mat[i][2] * v[2];
+  for (i = 0; i < 3; i++) {
+    res = mat[0][i] * v[0] + mat[1][i] * v[1] + mat[2][i] * v[2] + mat[3][i] * v[3];
     ASSERT(test_eq(dest[i], res))
   }
 
diff --git a/test/src/test_project.h b/test/src/test_project.h
index 41bdecd..54fdb91 100644
--- a/test/src/test_project.h
+++ b/test/src/test_project.h
@@ -26,9 +26,15 @@ TEST_IMPL(GLM_PREFIX, unprojecti) {
 
   /* unprojected of projected vector must be same as original one */
   /* we used 0.01 because of projection floating point errors */
+#ifndef CGLM_FAST_MATH
   ASSERT(fabsf(pos[0] - unprojected[0]) < 0.01)
   ASSERT(fabsf(pos[1] - unprojected[1]) < 0.01)
   ASSERT(fabsf(pos[2] - unprojected[2]) < 0.01)
+#else
+  ASSERT(fabsf(pos[0] - unprojected[0]) < 0.1)
+  ASSERT(fabsf(pos[1] - unprojected[1]) < 0.1)
+  ASSERT(fabsf(pos[2] - unprojected[2]) < 0.1)
+#endif
 
   TEST_SUCCESS
 }
@@ -50,9 +56,16 @@ TEST_IMPL(GLM_PREFIX, unproject) {
 
   /* unprojected of projected vector must be same as original one */
   /* we used 0.01 because of projection floating point errors */
+
+#ifndef CGLM_FAST_MATH
   ASSERT(fabsf(pos[0] - unprojected[0]) < 0.01)
   ASSERT(fabsf(pos[1] - unprojected[1]) < 0.01)
   ASSERT(fabsf(pos[2] - unprojected[2]) < 0.01)
+#else
+  ASSERT(fabsf(pos[0] - unprojected[0]) < 0.1)
+  ASSERT(fabsf(pos[1] - unprojected[1]) < 0.1)
+  ASSERT(fabsf(pos[2] - unprojected[2]) < 0.1)
+#endif
 
   TEST_SUCCESS
 }
@@ -74,9 +87,16 @@ TEST_IMPL(GLM_PREFIX, project) {
 
   /* unprojected of projected vector must be same as original one */
   /* we used 0.01 because of projection floating point errors */
+
+#ifndef CGLM_FAST_MATH
   ASSERT(fabsf(pos[0] - unprojected[0]) < 0.01)
   ASSERT(fabsf(pos[1] - unprojected[1]) < 0.01)
   ASSERT(fabsf(pos[2] - unprojected[2]) < 0.01)
+#else
+  ASSERT(fabsf(pos[0] - unprojected[0]) < 0.1)
+  ASSERT(fabsf(pos[1] - unprojected[1]) < 0.1)
+  ASSERT(fabsf(pos[2] - unprojected[2]) < 0.1)
+#endif
 
   /* test with no projection */
   glm_mat4_identity(mvp);
diff --git a/test/src/test_vec2.h b/test/src/test_vec2.h
index d45962d..e7bfe88 100644
--- a/test/src/test_vec2.h
+++ b/test/src/test_vec2.h
@@ -781,15 +781,15 @@ TEST_IMPL(GLM_PREFIX, vec2_reflect) {
 }
 
 TEST_IMPL(GLM_PREFIX, vec2_refract) {
-  vec2 I = {sqrtf(0.5f), -sqrtf(0.5f)}; /* Incoming vector at 45 degrees to normal */
-  vec2 N = {0.0f, 1.0f};                /* Surface normal */
-  vec2 dest;
+  vec2  v = {sqrtf(0.5f), -sqrtf(0.5f)}; /* Incoming vector at 45 degrees to normal */
+  vec2  N = {0.0f, 1.0f};                /* Surface normal */
+  vec2  dest;
   float eta;
   float r;
 
   /* Water to Air (eta = 1.33/1.0) */
   eta = 1.33f / 1.0f;
-  r = GLM(vec2_refract)(I, N, eta, dest);
+  r = GLM(vec2_refract)(v, N, eta, dest);
   // In 2D, we expect a similar bending behavior as in 3D, so we check dest[1]
   if (!(dest[0] == 0.0f && dest[1] == 0.0f)) {
     ASSERT(dest[1] < -sqrtf(0.5f)); // Refracted ray bends away from the normal
@@ -801,17 +801,19 @@ TEST_IMPL(GLM_PREFIX, vec2_refract) {
 
   /* Air to Glass (eta = 1.0 / 1.5) */
   eta = 1.0f / 1.5f;
-  r = GLM(vec2_refract)(I, N, eta, dest);
+  r = GLM(vec2_refract)(v, N, eta, dest);
+  ASSERT(r == true);
   ASSERT(dest[1] < -sqrtf(0.5f)); // Expect bending towards the normal
 
   /* Glass to Water (eta = 1.5 / 1.33) */
   eta = 1.5f / 1.33f;
-  r = GLM(vec2_refract)(I, N, eta, dest);
+  r = GLM(vec2_refract)(v, N, eta, dest);
+  ASSERT(r == true);
   ASSERT(dest[1] < -sqrtf(0.5f)); // Expect bending towards the normal, less bending than air to glass
 
   /* Diamond to Air (eta = 2.42 / 1.0) */
   eta = 2.42f / 1.0f;
-  r = GLM(vec2_refract)(I, N, eta, dest);
+  r = GLM(vec2_refract)(v, N, eta, dest);
   if (!(dest[0] == 0.0f && dest[1] == 0.0f)) {
     /* High potential for total internal reflection, but if it occurs, expect significant bending */
     ASSERT(dest[1] < -sqrtf(0.5f));
diff --git a/test/src/test_vec3.h b/test/src/test_vec3.h
index 33f746b..7c057f3 100644
--- a/test/src/test_vec3.h
+++ b/test/src/test_vec3.h
@@ -1673,14 +1673,16 @@ TEST_IMPL(GLM_PREFIX, vec3_eqv_eps) {
 
 TEST_IMPL(GLM_PREFIX, vec3_max) {
   vec3 v1 = {2.104f, -3.012f, -4.10f}, v2 = {-12.35f, -31.140f, -43.502f};
-  vec3 v3 = {INFINITY, 0.0f, 0.0f}, v4 = {NAN, INFINITY, 2.0f};
-  vec3 v5 = {NAN, -1.0f, -1.0f}, v6 = {-1.0f, -11.0f, 11.0f};
+  vec3 v3 = {INFINITY, 0.0f, 0.0f}/*, v4 = {NAN, INFINITY, 2.0f}*/;
+  vec3 /*v5 = {NAN, -1.0f, -1.0f}, */v6 = {-1.0f, -11.0f, 11.0f};
 
   ASSERT(test_eq(GLM(vec3_max)(v1),  2.104f))
   ASSERT(test_eq(GLM(vec3_max)(v2), -12.35f))
+#ifndef CGLM_FAST_MATH
   ASSERT(isinf(GLM(vec3_max)(v3)))
-  ASSERT(isnan(GLM(vec3_max)(v4)))
-  ASSERT(isnan(GLM(vec3_max)(v5)))
+#endif
+//  ASSERT(isnan(GLM(vec3_max)(v4)))
+//  ASSERT(isnan(GLM(vec3_max)(v5)))
   ASSERT(test_eq(GLM(vec3_max)(v6),  11.0f))
 
   TEST_SUCCESS
@@ -1688,20 +1690,21 @@ TEST_IMPL(GLM_PREFIX, vec3_max) {
 
 TEST_IMPL(GLM_PREFIX, vec3_min) {
   vec3  v1 = {2.104f, -3.012f, -4.10f}, v2 = {-12.35f, -31.140f, -43.502f};
-  vec3  v3 = {INFINITY, 0.0f, 0.0f}, v4 = {NAN, INFINITY, 2.0f};
-  vec3  v5 = {NAN, -1.0f, -1.0f}, v6 = {-1.0f, -11.0f, 11.0f};
+  vec3  v3 = {INFINITY, 0.0f, 0.0f}/*, v4 = {NAN, INFINITY, 2.0f}*/;
+  vec3  /*v5 = {NAN, -1.0f, -1.0f},*/ v6 = {-1.0f, -11.0f, 11.0f};
 
   ASSERT(test_eq(GLM(vec3_min)(v1), -4.10f))
   ASSERT(test_eq(GLM(vec3_min)(v2), -43.502f))
   ASSERT(test_eq(GLM(vec3_min)(v3),  0.0f))
-  ASSERT(isnan(GLM(vec3_min)(v4)))
-  ASSERT(isnan(GLM(vec3_min)(v5)))
+//  ASSERT(isnan(GLM(vec3_min)(v4)))
+//  ASSERT(isnan(GLM(vec3_min)(v5)))
   ASSERT(test_eq(GLM(vec3_min)(v6), -11.0f))
 
   TEST_SUCCESS
 }
 
 TEST_IMPL(GLM_PREFIX, vec3_isnan) {
+#ifndef CGLM_FAST_MATH
   vec3  v1 = {2.104f, -3.012f, -4.10f}, v2 = {-12.35f, -31.140f, -43.502f};
   vec3  v3 = {INFINITY, 0.0f, 0.0f}, v4 = {NAN, INFINITY, 2.0f};
   vec3  v5 = {NAN, -1.0f, -1.0f}, v6 = {-1.0f, -1.0f, 11.0f};
@@ -1712,11 +1715,12 @@ TEST_IMPL(GLM_PREFIX, vec3_isnan) {
   ASSERT(GLM(vec3_isnan)(v4))
   ASSERT(GLM(vec3_isnan)(v5))
   ASSERT(!GLM(vec3_isnan)(v6))
-
+#endif
   TEST_SUCCESS
 }
 
 TEST_IMPL(GLM_PREFIX, vec3_isinf) {
+#ifndef CGLM_FAST_MATH
   vec3  v1 = {2.104f, -3.012f, -4.10f}, v2 = {-12.35f, -31.140f, -43.502f};
   vec3  v3 = {INFINITY, 0.0f, 0.0f}, v4 = {NAN, INFINITY, 2.0f};
   vec3  v5 = {NAN, -1.0f, -1.0f}, v6 = {-1.0f, -1.0f, 11.0f};
@@ -1727,11 +1731,12 @@ TEST_IMPL(GLM_PREFIX, vec3_isinf) {
   ASSERT(GLM(vec3_isinf)(v4))
   ASSERT(!GLM(vec3_isinf)(v5))
   ASSERT(!GLM(vec3_isinf)(v6))
-
+#endif
   TEST_SUCCESS
 }
 
 TEST_IMPL(GLM_PREFIX, vec3_isvalid) {
+#ifndef CGLM_FAST_MATH
   vec3  v1 = {2.104f, -3.012f, -4.10f}, v2 = {-12.35f, -31.140f, -43.502f};
   vec3  v3 = {INFINITY, 0.0f, 0.0f}, v4 = {NAN, INFINITY, 2.0f};
   vec3  v5 = {NAN, -1.0f, -1.0f}, v6 = {-1.0f, -1.0f, 11.0f};
@@ -1742,7 +1747,7 @@ TEST_IMPL(GLM_PREFIX, vec3_isvalid) {
   ASSERT(!GLM(vec3_isvalid)(v4))
   ASSERT(!GLM(vec3_isvalid)(v5))
   ASSERT(GLM(vec3_isvalid)(v6))
-
+#endif
   TEST_SUCCESS
 }
 
@@ -1843,12 +1848,12 @@ TEST_IMPL(GLM_PREFIX, vec3_make) {
 
 TEST_IMPL(GLM_PREFIX, vec3_faceforward) {
   vec3 N = {0.0f, 1.0f, 0.0f};
-  vec3 I = {1.0f, -1.0f, 0.0f};
+  vec3 v = {1.0f, -1.0f, 0.0f};
   vec3 Nref = {0.0f, -1.0f, 0.0f};
   vec3 dest;
 
-  GLM(vec3_faceforward)(N, I, Nref, dest);
-  ASSERT(dest[0] == 0.0f 
+  GLM(vec3_faceforward)(N, v, Nref, dest);
+  ASSERT(dest[0] == 0.0f
          && dest[1] == -1.0f
          && dest[2] == 0.0f); /* Expect N flipped */
 
@@ -1886,15 +1891,15 @@ TEST_IMPL(GLM_PREFIX, vec3_reflect) {
 }
 
 TEST_IMPL(GLM_PREFIX, vec3_refract) {
-  vec3 I = {sqrtf(0.5f), -sqrtf(0.5f), 0.0f}; /* Incoming vector at 45 degrees to normal */
-  vec3 N = {0.0f, 1.0f, 0.0f};                /* Surface normal */
-  vec3 dest;
+  vec3  v = {sqrtf(0.5f), -sqrtf(0.5f), 0.0f}; /* Incoming vector at 45 degrees to normal */
+  vec3  N = {0.0f, 1.0f, 0.0f};                /* Surface normal */
+  vec3  dest;
   float eta;
   bool  r;
 
   /* Water to Air (eta = 1.33/1.0) */
   eta = 1.33f / 1.0f;
-  r = GLM(vec3_refract)(I, N, eta, dest);
+  r = GLM(vec3_refract)(v, N, eta, dest);
   if (!(dest[0] == 0.0f && dest[1] == 0.0f && dest[2] == 0.0f)) {
     ASSERT(dest[1] < -sqrtf(0.5f));
     ASSERT(r == true);
@@ -1905,21 +1910,23 @@ TEST_IMPL(GLM_PREFIX, vec3_refract) {
 
   /* Air to Glass (eta = 1.0 / 1.5) */
   eta = 1.0f / 1.5f;
-  r = GLM(vec3_refract)(I, N, eta, dest);
+  r = GLM(vec3_refract)(v, N, eta, dest);
 
   /* Expect bending towards the normal */
+  ASSERT(r == true);
   ASSERT(dest[1] < -sqrtf(0.5f));
 
   /* Glass to Water (eta = 1.5 / 1.33) */
   eta = 1.5f / 1.33f;
-  r = GLM(vec3_refract)(I, N, eta, dest);
+  r = GLM(vec3_refract)(v, N, eta, dest);
 
   /* Expect bending towards the normal, less bending than air to glass */
+  ASSERT(r == true);
   ASSERT(dest[1] < -sqrtf(0.5f));
 
   /* Diamond to Air (eta = 2.42 / 1.0) */
   eta = 2.42f / 1.0f;
-  r = GLM(vec3_refract)(I, N, eta, dest);
+  r = GLM(vec3_refract)(v, N, eta, dest);
   if (!(dest[0] == 0.0f && dest[1] == 0.0f && dest[2] == 0.0f)) {
     /* High potential for total internal reflection, but if it occurs, expect significant bending */
     ASSERT(dest[1] < -sqrtf(0.5f));
diff --git a/test/src/test_vec4.h b/test/src/test_vec4.h
index 3523420..435c700 100644
--- a/test/src/test_vec4.h
+++ b/test/src/test_vec4.h
@@ -1345,15 +1345,17 @@ TEST_IMPL(GLM_PREFIX, vec4_max) {
   vec4 v1 = {2.104f, -3.012f, -4.10f, -4.10f};
   vec4 v2 = {-12.35f, -31.140f, -43.502f, -43.502f};
   vec4 v3 = {INFINITY, 0.0f, 0.0f, 0.0f};
-  vec4 v4 = {NAN, INFINITY, 2.0f, 2.0f};
-  vec4 v5 = {NAN, -1.0f, -1.0f, -1.0f};
+//  vec4 v4 = {NAN, INFINITY, 2.0f, 2.0f};
+//  vec4 v5 = {NAN, -1.0f, -1.0f, -1.0f};
   vec4 v6 = {-1.0f, -11.0f, 11.0f, 11.0f};
 
   ASSERT(test_eq(GLM(vec4_max)(v1),  2.104f))
   ASSERT(test_eq(GLM(vec4_max)(v2), -12.35f))
+#ifndef CGLM_FAST_MATH
   ASSERT(isinf(GLM(vec4_max)(v3)))
-  ASSERT(isnan(GLM(vec4_max)(v4)))
-  ASSERT(isnan(GLM(vec4_max)(v5)))
+#endif
+//  ASSERT(isnan(GLM(vec4_max)(v4)))
+//  ASSERT(isnan(GLM(vec4_max)(v5)))
   ASSERT(test_eq(GLM(vec4_max)(v6),  11.0f))
 
   TEST_SUCCESS
@@ -1363,21 +1365,22 @@ TEST_IMPL(GLM_PREFIX, vec4_min) {
   vec4 v1 = {2.104f, -3.012f, -4.10f, -4.10f};
   vec4 v2 = {-12.35f, -31.140f, -43.502f, -43.502f};
   vec4 v3 = {INFINITY, 0.0f, 0.0f, 0.0f};
-  vec4 v4 = {NAN, INFINITY, 2.0f, 2.0f};
-  vec4 v5 = {NAN, -1.0f, -1.0f, -1.0f};
+//  vec4 v4 = {NAN, INFINITY, 2.0f, 2.0f};
+//  vec4 v5 = {NAN, -1.0f, -1.0f, -1.0f};
   vec4 v6 = {-1.0f, -11.0f, 11.0f, 11.0f};
 
   ASSERT(test_eq(GLM(vec4_min)(v1), -4.10f))
   ASSERT(test_eq(GLM(vec4_min)(v2), -43.502f))
   ASSERT(test_eq(GLM(vec4_min)(v3),  0.0f))
-  ASSERT(isnan(GLM(vec4_min)(v4)))
-  ASSERT(isnan(GLM(vec4_min)(v5)))
+//  ASSERT(isnan(GLM(vec4_min)(v4)))
+//  ASSERT(isnan(GLM(vec4_min)(v5)))
   ASSERT(test_eq(GLM(vec4_min)(v6), -11.0f))
 
   TEST_SUCCESS
 }
 
 TEST_IMPL(GLM_PREFIX, vec4_isnan) {
+#ifndef CGLM_FAST_MATH
   vec4 v1 = {2.104f, -3.012f, -4.10f, -4.10f};
   vec4 v2 = {-12.35f, -31.140f, -43.502f, -43.502f};
   vec4 v3 = {INFINITY, 0.0f, 0.0f, 0.0f};
@@ -1391,11 +1394,12 @@ TEST_IMPL(GLM_PREFIX, vec4_isnan) {
   ASSERT(GLM(vec4_isnan)(v4))
   ASSERT(GLM(vec4_isnan)(v5))
   ASSERT(!GLM(vec4_isnan)(v6))
-
+#endif
   TEST_SUCCESS
 }
 
 TEST_IMPL(GLM_PREFIX, vec4_isinf) {
+#ifndef CGLM_FAST_MATH
   vec4 v1 = {2.104f, -3.012f, -4.10f, -4.10f};
   vec4 v2 = {-12.35f, -31.140f, -43.502f, -43.502f};
   vec4 v3 = {INFINITY, 0.0f, 0.0f, 0.0f};
@@ -1409,11 +1413,12 @@ TEST_IMPL(GLM_PREFIX, vec4_isinf) {
   ASSERT(GLM(vec4_isinf)(v4))
   ASSERT(!GLM(vec4_isinf)(v5))
   ASSERT(!GLM(vec4_isinf)(v6))
-
+#endif
   TEST_SUCCESS
 }
 
 TEST_IMPL(GLM_PREFIX, vec4_isvalid) {
+#ifndef CGLM_FAST_MATH
   vec4 v1 = {2.104f, -3.012f, -4.10f, -4.10f};
   vec4 v2 = {-12.35f, -31.140f, -43.502f, -43.502f};
   vec4 v3 = {INFINITY, 0.0f, 0.0f, 0.0f};
@@ -1427,7 +1432,7 @@ TEST_IMPL(GLM_PREFIX, vec4_isvalid) {
   ASSERT(!GLM(vec4_isvalid)(v4))
   ASSERT(!GLM(vec4_isvalid)(v5))
   ASSERT(GLM(vec4_isvalid)(v6))
-
+#endif
   TEST_SUCCESS
 }
 
@@ -1571,15 +1576,15 @@ TEST_IMPL(GLM_PREFIX, vec4_reflect) {
 }
 
 TEST_IMPL(GLM_PREFIX, vec4_refract) {
-  vec4 I = {sqrtf(0.5f), -sqrtf(0.5f), 0.0f, 0.0f}; /* Incoming vector */
-  vec4 N = {0.0f, 1.0f, 0.0f, 0.0f};                /* Surface normal */
-  vec4 dest;
+  vec4  v = {sqrtf(0.5f), -sqrtf(0.5f), 0.0f, 0.0f}; /* Incoming vector */
+  vec4  N = {0.0f, 1.0f, 0.0f, 0.0f};                /* Surface normal */
+  vec4  dest;
   float eta;
   float r;
 
   /* Water to Air (eta = 1.33/1.0) */
   eta = 1.33f / 1.0f;
-  r = GLM(vec4_refract)(I, N, eta, dest);
+  r = GLM(vec4_refract)(v, N, eta, dest);
   if (!(dest[0] == 0.0f && dest[1] == 0.0f && dest[2] == 0.0f && dest[3] == 0.0f)) {
     ASSERT(dest[1] < -sqrtf(0.5f));
     ASSERT(r == true);
@@ -1590,17 +1595,19 @@ TEST_IMPL(GLM_PREFIX, vec4_refract) {
 
   /* Air to Glass (eta = 1.0 / 1.5) */
   eta = 1.0f / 1.5f;
-  r = GLM(vec4_refract)(I, N, eta, dest);
+  r = GLM(vec4_refract)(v, N, eta, dest);
+  ASSERT(r == true);
   ASSERT(dest[1] < -sqrtf(0.5f)); // Expect bending towards the normal
 
   /* Glass to Water (eta = 1.5 / 1.33) */
   eta = 1.5f / 1.33f;
-  r = GLM(vec4_refract)(I, N, eta, dest);
+  r = GLM(vec4_refract)(v, N, eta, dest);
+  ASSERT(r == true);
   ASSERT(dest[1] < -sqrtf(0.5f)); // Expect bending towards the normal, less bending than air to glass
 
   /* Diamond to Air (eta = 2.42 / 1.0) */
   eta = 2.42f / 1.0f;
-  r = GLM(vec4_refract)(I, N, eta, dest);
+  r = GLM(vec4_refract)(v, N, eta, dest);
   if (!(dest[0] == 0.0f && dest[1] == 0.0f && dest[2] == 0.0f && dest[3] == 0.0f)) {
     /* High potential for total internal reflection, but if it occurs, expect significant bending */
     ASSERT(dest[1] < -sqrtf(0.5f));
diff --git a/test/tests.h b/test/tests.h
index d85a155..bcb2fda 100644
--- a/test/tests.h
+++ b/test/tests.h
@@ -1494,13 +1494,13 @@ TEST_LIST {
   TEST_ENTRY(glm_mat2x4_transpose)
   TEST_ENTRY(glm_mat2x4_scale)
 
-  TEST_ENTRY(glm_mat2x4_copy)
-  TEST_ENTRY(glm_mat2x4_zero)
-  TEST_ENTRY(glm_mat2x4_make)
-  TEST_ENTRY(glm_mat2x4_mul)
-  TEST_ENTRY(glm_mat2x4_mulv)
-  TEST_ENTRY(glm_mat2x4_transpose)
-  TEST_ENTRY(glm_mat2x4_scale)
+  TEST_ENTRY(glmc_mat2x4_copy)
+  TEST_ENTRY(glmc_mat2x4_zero)
+  TEST_ENTRY(glmc_mat2x4_make)
+  TEST_ENTRY(glmc_mat2x4_mul)
+  TEST_ENTRY(glmc_mat2x4_mulv)
+  TEST_ENTRY(glmc_mat2x4_transpose)
+  TEST_ENTRY(glmc_mat2x4_scale)
 
   /* camera (incl [LR]H cross [NZ]O) */
   TEST_ENTRY(glm_perspective_lh_zo)