Rollup merge of #91008 - Urgau:float-minimum-maximum, r=scottmcm

Adds IEEE 754-2019 minimun and maximum functions for f32/f64

IEEE 754-2019 removed the `minNum` (`min` in Rust) and `maxNum` (`max` in Rust) operations in favor of the newly created `minimum` and `maximum` operations due to their [non-associativity](https://grouper.ieee.org/groups/msc/ANSI_IEEE-Std-754-2019/background/minNum_maxNum_Removal_Demotion_v3.pdf) that cannot be fix in a backwards compatible manner. This PR adds `fN::{minimun,maximum}` functions following the new rules.

### IEEE 754-2019 Rules

> **minimum(x, y)** is x if x < y, y if y < x, and a quiet NaN if either operand is a NaN, according to 6.2.
For this operation, −0 compares less than +0. Otherwise (i.e., when x = y and signs are the same)
it is either x or y.

> **maximum(x, y)** is x if x > y, y if y > x, and a quiet NaN if either operand is a NaN, according to 6.2.
For this operation, +0 compares greater than −0. Otherwise (i.e., when x = y and signs are the
same) it is either x or y.

"IEEE Standard for Floating-Point Arithmetic," in IEEE Std 754-2019 (Revision of IEEE 754-2008) , vol., no., pp.1-84, 22 July 2019, doi: 10.1109/IEEESTD.2019.8766229.

### Implementation

This implementation is inspired by the one in [`glibc` ](https://github.com/bminor/glibc/blob/90f0ac10a74b2d43b5a65aab4be40565e359be43/math/s_fminimum_template.c) (it self derived from the C2X draft) expect that:
 - it doesn't use `copysign` because it's not available in `core` and also because `copysign` is unnecessary (we only want to check the sign, no need to create a new float)
 - it also prefer `other > self` instead of `self < other` like IEEE 754-2019 does

I originally tried to implement them [using intrinsics](Urgau@1d8aa13) but LLVM [error out](https://godbolt.org/z/7sMrxW49a) when trying to lower them to machine intructions, GCC doesn't yet have built-ins for them, only cranelift support them nativelly (as it doesn't support the nativelly the old sementics).

Helps with #83984

library/core/src/num/f32.rs

@@ -673,6 +673,9 @@ impl f32 {
 
     /// Returns the maximum of the two numbers.
     ///
+    /// Follows the IEEE-754 2008 semantics for maxNum, except for handling of signaling NaNs.
+    /// This matches the behavior of libm’s fmin.
+    ///
     /// ```
     /// let x = 1.0f32;
     /// let y = 2.0f32;
@@ -689,6 +692,9 @@ impl f32 {
 
     /// Returns the minimum of the two numbers.
     ///
+    /// Follows the IEEE-754 2008 semantics for minNum, except for handling of signaling NaNs.
+    /// This matches the behavior of libm’s fmin.
+    ///
     /// ```
     /// let x = 1.0f32;
     /// let y = 2.0f32;
@@ -703,6 +709,68 @@ impl f32 {
         intrinsics::minnumf32(self, other)
     }
 
+    /// Returns the maximum of the two numbers, propagating NaNs.
+    ///
+    /// This returns NaN when *either* argument is NaN, as opposed to
+    /// [`f32::max`] which only returns NaN when *both* arguments are NaN.
+    ///
+    /// ```
+    /// #![feature(float_minimum_maximum)]
+    /// let x = 1.0f32;
+    /// let y = 2.0f32;
+    ///
+    /// assert_eq!(x.maximum(y), y);
+    /// assert!(x.maximum(f32::NAN).is_nan());
+    /// ```
+    ///
+    /// If one of the arguments is NaN, then NaN is returned. Otherwise this returns the greater
+    /// of the two numbers. For this operation, -0.0 is considered to be less than +0.0.
+    /// Note that this follows the semantics specified in IEEE 754-2019.
+    #[unstable(feature = "float_minimum_maximum", issue = "91079")]
+    #[inline]
+    pub fn maximum(self, other: f32) -> f32 {
+        if self > other {
+            self
+        } else if other > self {
+            other
+        } else if self == other {
+            if self.is_sign_positive() && other.is_sign_negative() { self } else { other }
+        } else {
+            self + other
+        }
+    }
+
+    /// Returns the minimum of the two numbers, propagating NaNs.
+    ///
+    /// This returns NaN when *either* argument is NaN, as opposed to
+    /// [`f32::min`] which only returns NaN when *both* arguments are NaN.
+    ///
+    /// ```
+    /// #![feature(float_minimum_maximum)]
+    /// let x = 1.0f32;
+    /// let y = 2.0f32;
+    ///
+    /// assert_eq!(x.minimum(y), x);
+    /// assert!(x.minimum(f32::NAN).is_nan());
+    /// ```
+    ///
+    /// If one of the arguments is NaN, then NaN is returned. Otherwise this returns the lesser
+    /// of the two numbers. For this operation, -0.0 is considered to be less than +0.0.
+    /// Note that this follows the semantics specified in IEEE 754-2019.
+    #[unstable(feature = "float_minimum_maximum", issue = "91079")]
+    #[inline]
+    pub fn minimum(self, other: f32) -> f32 {
+        if self < other {
+            self
+        } else if other < self {
+            other
+        } else if self == other {
+            if self.is_sign_negative() && other.is_sign_positive() { self } else { other }
+        } else {
+            self + other
+        }
+    }
+
     /// Rounds toward zero and converts to any primitive integer type,
     /// assuming that the value is finite and fits in that type.
     ///

library/core/src/num/f64.rs

@@ -689,6 +689,9 @@ impl f64 {
 
     /// Returns the maximum of the two numbers.
     ///
+    /// Follows the IEEE-754 2008 semantics for maxNum, except for handling of signaling NaNs.
+    /// This matches the behavior of libm’s fmin.
+    ///
     /// ```
     /// let x = 1.0_f64;
     /// let y = 2.0_f64;
@@ -705,6 +708,9 @@ impl f64 {
 
     /// Returns the minimum of the two numbers.
     ///
+    /// Follows the IEEE-754 2008 semantics for minNum, except for handling of signaling NaNs.
+    /// This matches the behavior of libm’s fmin.
+    ///
     /// ```
     /// let x = 1.0_f64;
     /// let y = 2.0_f64;
@@ -719,6 +725,68 @@ impl f64 {
         intrinsics::minnumf64(self, other)
     }
 
+    /// Returns the maximum of the two numbers, propagating NaNs.
+    ///
+    /// This returns NaN when *either* argument is NaN, as opposed to
+    /// [`f64::max`] which only returns NaN when *both* arguments are NaN.
+    ///
+    /// ```
+    /// #![feature(float_minimum_maximum)]
+    /// let x = 1.0_f64;
+    /// let y = 2.0_f64;
+    ///
+    /// assert_eq!(x.maximum(y), y);
+    /// assert!(x.maximum(f64::NAN).is_nan());
+    /// ```
+    ///
+    /// If one of the arguments is NaN, then NaN is returned. Otherwise this returns the greater
+    /// of the two numbers. For this operation, -0.0 is considered to be less than +0.0.
+    /// Note that this follows the semantics specified in IEEE 754-2019.
+    #[unstable(feature = "float_minimum_maximum", issue = "91079")]
+    #[inline]
+    pub fn maximum(self, other: f64) -> f64 {
+        if self > other {
+            self
+        } else if other > self {
+            other
+        } else if self == other {
+            if self.is_sign_positive() && other.is_sign_negative() { self } else { other }
+        } else {
+            self + other
+        }
+    }
+
+    /// Returns the minimum of the two numbers, propagating NaNs.
+    ///
+    /// This returns NaN when *either* argument is NaN, as opposed to
+    /// [`f64::min`] which only returns NaN when *both* arguments are NaN.
+    ///
+    /// ```
+    /// #![feature(float_minimum_maximum)]
+    /// let x = 1.0_f64;
+    /// let y = 2.0_f64;
+    ///
+    /// assert_eq!(x.minimum(y), x);
+    /// assert!(x.minimum(f64::NAN).is_nan());
+    /// ```
+    ///
+    /// If one of the arguments is NaN, then NaN is returned. Otherwise this returns the lesser
+    /// of the two numbers. For this operation, -0.0 is considered to be less than +0.0.
+    /// Note that this follows the semantics specified in IEEE 754-2019.
+    #[unstable(feature = "float_minimum_maximum", issue = "91079")]
+    #[inline]
+    pub fn minimum(self, other: f64) -> f64 {
+        if self < other {
+            self
+        } else if other < self {
+            other
+        } else if self == other {
+            if self.is_sign_negative() && other.is_sign_positive() { self } else { other }
+        } else {
+            self + other
+        }
+    }
+
     /// Rounds toward zero and converts to any primitive integer type,
     /// assuming that the value is finite and fits in that type.
     ///

library/core/tests/lib.rs

@@ -27,6 +27,7 @@
 #![feature(extern_types)]
 #![feature(flt2dec)]
 #![feature(fmt_internals)]
+#![feature(float_minimum_maximum)]
 #![feature(array_from_fn)]
 #![feature(hashmap_internals)]
 #![feature(try_find)]

library/core/tests/num/mod.rs

@@ -715,6 +715,67 @@ macro_rules! test_float {
                 assert!(($nan as $fty).max($nan).is_nan());
             }
             #[test]
+            fn minimum() {
+                assert_eq!((0.0 as $fty).minimum(0.0), 0.0);
+                assert!((0.0 as $fty).minimum(0.0).is_sign_positive());
+                assert_eq!((-0.0 as $fty).minimum(0.0), -0.0);
+                assert!((-0.0 as $fty).minimum(0.0).is_sign_negative());
+                assert_eq!((-0.0 as $fty).minimum(-0.0), -0.0);
+                assert!((-0.0 as $fty).minimum(-0.0).is_sign_negative());
+                assert_eq!((9.0 as $fty).minimum(9.0), 9.0);
+                assert_eq!((-9.0 as $fty).minimum(0.0), -9.0);
+                assert_eq!((0.0 as $fty).minimum(9.0), 0.0);
+                assert!((0.0 as $fty).minimum(9.0).is_sign_positive());
+                assert_eq!((-0.0 as $fty).minimum(9.0), -0.0);
+                assert!((-0.0 as $fty).minimum(9.0).is_sign_negative());
+                assert_eq!((-0.0 as $fty).minimum(-9.0), -9.0);
+                assert_eq!(($inf as $fty).minimum(9.0), 9.0);
+                assert_eq!((9.0 as $fty).minimum($inf), 9.0);
+                assert_eq!(($inf as $fty).minimum(-9.0), -9.0);
+                assert_eq!((-9.0 as $fty).minimum($inf), -9.0);
+                assert_eq!(($neginf as $fty).minimum(9.0), $neginf);
+                assert_eq!((9.0 as $fty).minimum($neginf), $neginf);
+                assert_eq!(($neginf as $fty).minimum(-9.0), $neginf);
+                assert_eq!((-9.0 as $fty).minimum($neginf), $neginf);
+                assert!(($nan as $fty).minimum(9.0).is_nan());
+                assert!(($nan as $fty).minimum(-9.0).is_nan());
+                assert!((9.0 as $fty).minimum($nan).is_nan());
+                assert!((-9.0 as $fty).minimum($nan).is_nan());
+                assert!(($nan as $fty).minimum($nan).is_nan());
+            }
+            #[test]
+            fn maximum() {
+                assert_eq!((0.0 as $fty).maximum(0.0), 0.0);
+                assert!((0.0 as $fty).maximum(0.0).is_sign_positive());
+                assert_eq!((-0.0 as $fty).maximum(0.0), 0.0);
+                assert!((-0.0 as $fty).maximum(0.0).is_sign_positive());
+                assert_eq!((-0.0 as $fty).maximum(-0.0), -0.0);
+                assert!((-0.0 as $fty).maximum(-0.0).is_sign_negative());
+                assert_eq!((9.0 as $fty).maximum(9.0), 9.0);
+                assert_eq!((-9.0 as $fty).maximum(0.0), 0.0);
+                assert!((-9.0 as $fty).maximum(0.0).is_sign_positive());
+                assert_eq!((-9.0 as $fty).maximum(-0.0), -0.0);
+                assert!((-9.0 as $fty).maximum(-0.0).is_sign_negative());
+                assert_eq!((0.0 as $fty).maximum(9.0), 9.0);
+                assert_eq!((0.0 as $fty).maximum(-9.0), 0.0);
+                assert!((0.0 as $fty).maximum(-9.0).is_sign_positive());
+                assert_eq!((-0.0 as $fty).maximum(-9.0), -0.0);
+                assert!((-0.0 as $fty).maximum(-9.0).is_sign_negative());
+                assert_eq!(($inf as $fty).maximum(9.0), $inf);
+                assert_eq!((9.0 as $fty).maximum($inf), $inf);
+                assert_eq!(($inf as $fty).maximum(-9.0), $inf);
+                assert_eq!((-9.0 as $fty).maximum($inf), $inf);
+                assert_eq!(($neginf as $fty).maximum(9.0), 9.0);
+                assert_eq!((9.0 as $fty).maximum($neginf), 9.0);
+                assert_eq!(($neginf as $fty).maximum(-9.0), -9.0);
+                assert_eq!((-9.0 as $fty).maximum($neginf), -9.0);
+                assert!(($nan as $fty).maximum(9.0).is_nan());
+                assert!(($nan as $fty).maximum(-9.0).is_nan());
+                assert!((9.0 as $fty).maximum($nan).is_nan());
+                assert!((-9.0 as $fty).maximum($nan).is_nan());
+                assert!(($nan as $fty).maximum($nan).is_nan());
+            }
+            #[test]
             fn rem_euclid() {
                 let a: $fty = 42.0;
                 assert!($inf.rem_euclid(a).is_nan());

library/std/src/f32/tests.rs

@@ -19,6 +19,18 @@ fn test_max_nan() {
     assert_eq!(2.0f32.max(f32::NAN), 2.0);
 }
 
+#[test]
+fn test_minimum() {
+    assert!(f32::NAN.minimum(2.0).is_nan());
+    assert!(2.0f32.minimum(f32::NAN).is_nan());
+}
+
+#[test]
+fn test_maximum() {
+    assert!(f32::NAN.maximum(2.0).is_nan());
+    assert!(2.0f32.maximum(f32::NAN).is_nan());
+}
+
 #[test]
 fn test_nan() {
     let nan: f32 = f32::NAN;

library/std/src/lib.rs

@@ -287,6 +287,7 @@
 #![feature(exhaustive_patterns)]
 #![feature(extend_one)]
 #![feature(fn_traits)]
+#![feature(float_minimum_maximum)]
 #![feature(format_args_nl)]
 #![feature(gen_future)]
 #![feature(generator_trait)]