Rename EXP_MAX to EXP_SAT

"Maximum" is technically correct here with regards to what the
bitpattern can represent, but it is not the numeric maximum value of the
exponent which has a relationship with the bias. So, replace the maximum
terminology with "saturated" to indicate it only means the full
bitpattern.

This change is more relevant to `libm` than `compiler-builtins`.

Author: tgross35

src/float/add.rs

@@ -14,7 +14,7 @@ where
 
     let bits = F::BITS.cast();
     let significand_bits = F::SIG_BITS;
-    let max_exponent = F::EXP_MAX;
+    let max_exponent = F::EXP_SAT;
 
     let implicit_bit = F::IMPLICIT_BIT;
     let significand_mask = F::SIG_MASK;

src/float/div.rs

@@ -107,7 +107,7 @@ where
 
     let significand_bits = F::SIG_BITS;
     // Saturated exponent, representing infinity
-    let exponent_sat: F::Int = F::EXP_MAX.cast();
+    let exponent_sat: F::Int = F::EXP_SAT.cast();
 
     let exponent_bias = F::EXP_BIAS;
     let implicit_bit = F::IMPLICIT_BIT;

src/float/extend.rs

@@ -26,7 +26,7 @@ where
 
     let dst_bits = R::BITS;
     let dst_sign_bits = R::SIG_BITS;
-    let dst_inf_exp = R::EXP_MAX;
+    let dst_inf_exp = R::EXP_SAT;
     let dst_exp_bias = R::EXP_BIAS;
     let dst_min_normal = R::IMPLICIT_BIT;
 

src/float/mod.rs

@@ -51,11 +51,14 @@ pub(crate) trait Float:
     /// The bitwidth of the exponent.
     const EXP_BITS: u32 = Self::BITS - Self::SIG_BITS - 1;
 
-    /// The saturated value of the exponent (infinite representation), in the rightmost postiion.
-    const EXP_MAX: u32 = (1 << Self::EXP_BITS) - 1;
+    /// The saturated (maximum bitpattern) value of the exponent, i.e. the infinite
+    /// representation.
+    ///
+    /// This is in the rightmost position, use `EXP_MASK` for the shifted value.
+    const EXP_SAT: u32 = (1 << Self::EXP_BITS) - 1;
 
     /// The exponent bias value.
-    const EXP_BIAS: u32 = Self::EXP_MAX >> 1;
+    const EXP_BIAS: u32 = Self::EXP_SAT >> 1;
 
     /// A mask for the sign bit.
     const SIGN_MASK: Self::Int;

src/float/mul.rs

@@ -14,7 +14,7 @@ where
 
     let bits = F::BITS;
     let significand_bits = F::SIG_BITS;
-    let max_exponent = F::EXP_MAX;
+    let max_exponent = F::EXP_SAT;
 
     let exponent_bias = F::EXP_BIAS;
 

src/float/trunc.rs

@@ -29,7 +29,7 @@ where
     let dst_zero = R::Int::ZERO;
     let dst_one = R::Int::ONE;
     let dst_bits = R::BITS;
-    let dst_inf_exp = R::EXP_MAX;
+    let dst_inf_exp = R::EXP_SAT;
     let dst_exp_bias = R::EXP_BIAS;
 
     let underflow_exponent: F::Int = (src_exp_bias + 1 - dst_exp_bias).cast();