pub struct DoubleFloat<F>(F, F);
pub type DoubleDouble = DoubleFloat<ieee::Double>;
-// These are legacy semantics for the Fallback, inaccrurate implementation of
+// These are legacy semantics for the Fallback, inaccurate implementation of
// IBM double-double, if the accurate DoubleDouble doesn't handle the
// operation. It's equivalent to having an IEEE number with consecutive 106
// bits of mantissa and 11 bits of exponent.
// \ /
// Normal
//
- // e.g. NaN * NaN = NaN
+ // e.g., NaN * NaN = NaN
// Zero * Inf = NaN
// Normal * Zero = Zero
// Normal * Inf = Inf