Auto merge of #95317 - Jules-Bertholet:round_ties_to_even, r=pnkfelix,m-ou-se,scottmcm

Add `round_ties_even` to `f32` and `f64`

Tracking issue: #96710

Redux of #82273. See also #55107

Adds a new method, `round_ties_even`, to `f32` and `f64`, that rounds the float to the nearest integer , rounding halfway cases to the number with an even least significant bit. Uses the `roundeven` LLVM intrinsic to do this.

Of the five IEEE 754 rounding modes, this is the only one that doesn't already have a round-to-integer function exposed by Rust (others are `round`, `floor`, `ceil`, and `trunc`).  Ties-to-even is also the rounding mode used for int-to-float and float-to-float `as` casts, as well as float arithmentic operations. So not having an explicit rounding method for it seems like an oversight.

Bikeshed: this PR currently uses `round_ties_even` for the name of the method. But maybe `round_ties_to_even` is better, or `round_even`, or `round_to_even`?

1 files changed, 19 insertions, 0 deletions

diff --git a/library/core/src/intrinsics.rs b/library/core/src/intrinsics.rs
index 18a90599c4d..c6321587adc 100644
--- a/library/core/src/intrinsics.rs
+++ b/library/core/src/intrinsics.rs
@@ -1585,9 +1585,15 @@ extern "rust-intrinsic" {
 
     /// Returns the nearest integer to an `f32`. May raise an inexact floating-point exception
     /// if the argument is not an integer.
+    ///
+    /// The stabilized version of this intrinsic is
+    /// [`f32::round_ties_even`](../../std/primitive.f32.html#method.round_ties_even)
     pub fn rintf32(x: f32) -> f32;
     /// Returns the nearest integer to an `f64`. May raise an inexact floating-point exception
     /// if the argument is not an integer.
+    ///
+    /// The stabilized version of this intrinsic is
+    /// [`f64::round_ties_even`](../../std/primitive.f64.html#method.round_ties_even)
     pub fn rintf64(x: f64) -> f64;
 
     /// Returns the nearest integer to an `f32`.
@@ -1610,6 +1616,19 @@ extern "rust-intrinsic" {
     /// [`f64::round`](../../std/primitive.f64.html#method.round)
     pub fn roundf64(x: f64) -> f64;
 
+    /// Returns the nearest integer to an `f32`. Rounds half-way cases to the number
+    /// with an even least significant digit.
+    ///
+    /// This intrinsic does not have a stable counterpart.
+    #[cfg(not(bootstrap))]
+    pub fn roundevenf32(x: f32) -> f32;
+    /// Returns the nearest integer to an `f64`. Rounds half-way cases to the number
+    /// with an even least significant digit.
+    ///
+    /// This intrinsic does not have a stable counterpart.
+    #[cfg(not(bootstrap))]
+    pub fn roundevenf64(x: f64) -> f64;
+
     /// Float addition that allows optimizations based on algebraic rules.
     /// May assume inputs are finite.
     ///