Class description
Symbols of this type represent floating-point rounding modes. They are passed to the with-rounding-mode word to control how inexact values are calculated when exact results cannot fit in a floating-point type. The following symbols are defined:
+round-nearest+ rounds the exact result to the nearest representable value, using the even value when the result is halfway between its two nearest values.
+round-zero+ rounds the exact result toward zero, that is, down for positive values, and up for negative values.
+round-down+ always rounds the exact result down.
+round-up+ always rounds the exact result up.