propose sane arithmetic and shift operation semantics

Author: gnzlbg

text/0000-ppv.md

@@ -512,6 +512,48 @@ pub unsafe fn load_unaligned_unchecked(slice: &[element_type]) -> Self;
 }
 ```
 
+#### Vertical arithmetic operations
+
+Vertical (lane-wise) arithmetic operations are provided by the following trait
+implementations:
+
+* All signed integer, unsigned integer, and floating point vector types implement:
+
+    * `{Add,Sub,Mul,Div,Rem}<RHS=Self,Output=Self>`
+    * `{Add,Sub,Mul,Div,Rem}Assign<RHS=Self>`
+
+* All signed and unsigned integer vectors also implement:
+
+    * `{Shl,Shr}<RHS=Self,Output=Self>`, `{Shl,Shr}Assign<RHS=Self>`: vertical
+      (lane-wise) bit-shift operations.
+
+##### Integer vector semantics
+
+The behavior of these operations for integer vectors is the same as that of the
+scalar integer types. That is: `panic!` on both overflow and division by zero.
+  
+##### Floating-point semantics
+
+The behavior of these operations for floating-point numbers is the same as that
+of the scalar floating point types, that is, `+-INFINITY` on overflow, `NaN` on
+division by zero, etc.
+
+#### Wrapping arithmetic operations
+
+All signed and unsigned integer vector types implement the whole set of `pub fn
+wrapping_{add,sub,mul,div,rem}(self, Self) -> Self` methods which, on overflow,
+produce the correct mathematical result modulo `2^n`.
+
+The `div` and `rem` method `panic!` on division by zero.
+
+#### Unsafe wrapping arithmetic operations
+
+All signed and unsigned integer vectors implement
+`pub unsafe fn wrapping_{div,rem}_unchecked(self, Self) -> Self`
+methods which, on overflow, produce the correct mathematical result modulo `2^n`.
+
+If any of the vector elements is divided by zero the behavior is undefined.
+
 #### Binary minmax vertical operations
 
 All portable signed integer, unsigned integer, and floating-point vectors