Rollup merge of rust-lang#153675 - RalfJung:simd-overflow, r=scottmcm

simd_add/sub/mul/neg: document overflow behavior

`simd_neg` had an odd comment about overflow not being UB, without saying what the behavior is instead. Replace that by just saying this uses wrapping arithmetic, and add the same for add/sub/mul. div/rem are already documented to cause UB on div-by-zero and min-div-by-minus-one, and shl/shr cause UB on too large shift amounts.

Author: jhpratt

library/core/src/intrinsics/simd.rs

@@ -62,20 +62,23 @@ pub const unsafe fn simd_splat<T, U>(value: U) -> T;
 /// Adds two simd vectors elementwise.
 ///
 /// `T` must be a vector of integers or floats.
+/// For integers, wrapping arithmetic is used.
 #[rustc_intrinsic]
 #[rustc_nounwind]
 pub const unsafe fn simd_add<T>(x: T, y: T) -> T;
 
 /// Subtracts `rhs` from `lhs` elementwise.
 ///
 /// `T` must be a vector of integers or floats.
+/// For integers, wrapping arithmetic is used.
 #[rustc_intrinsic]
 #[rustc_nounwind]
 pub const unsafe fn simd_sub<T>(lhs: T, rhs: T) -> T;
 
 /// Multiplies two simd vectors elementwise.
 ///
 /// `T` must be a vector of integers or floats.
+/// For integers, wrapping arithmetic is used.
 #[rustc_intrinsic]
 #[rustc_nounwind]
 pub const unsafe fn simd_mul<T>(x: T, y: T) -> T;
@@ -233,8 +236,7 @@ pub const unsafe fn simd_as<T, U>(x: T) -> U;
 /// Negates a vector elementwise.
 ///
 /// `T` must be a vector of integers or floats.
-///
-/// Rust panics for `-<int>::Min` due to overflow, but it is not UB with this intrinsic.
+/// For integers, wrapping arithmetic is used.
 #[rustc_intrinsic]
 #[rustc_nounwind]
 pub const unsafe fn simd_neg<T>(x: T) -> T;