simd intrinsics: add simd_shuffle_generic

Author: RalfJung

library/core/src/intrinsics/simd.rs

@@ -190,14 +190,27 @@ extern "platform-intrinsic" {
     ///
     /// `T` must be a vector.
     ///
-    /// `U` must be a const array of `i32`s.
+    /// `U` must be a **const** array of `i32`s. This means it must either refer to a named
+    /// const or be given as an inline const expression (`const { ... }`).
     ///
     /// `V` must be a vector with the same element type as `T` and the same length as `U`.
     ///
-    /// Concatenates `x` and `y`, then returns a new vector such that each element is selected from
-    /// the concatenation by the matching index in `idx`.
+    /// Returns a new vector such that element `i` is selected from `xy[idx[i]]`, where `xy`
+    /// is the concatenation of `x` and `y`. It is a compile-time error if `idx[i]` is out-of-bounds
+    /// of `xy`.
     pub fn simd_shuffle<T, U, V>(x: T, y: T, idx: U) -> V;
 
+    /// Shuffle two vectors by const indices.
+    ///
+    /// `T` must be a vector.
+    ///
+    /// `V` must be a vector with the same element type as `T` and the same length as `IDX`.
+    ///
+    /// Returns a new vector such that element `i` is selected from `xy[IDX[i]]`, where `xy`
+    /// is the concatenation of `x` and `y`. It is a compile-time error if `IDX[i]` is out-of-bounds
+    /// of `xy`.
+    pub fn simd_shuffle_generic<T, U, const IDX: &'static [u32]>(x: T, y: T) -> U;
+
     /// Read a vector of pointers.
     ///
     /// `T` must be a vector.