// The prover treats a FLOAT as a number: Y == Y, and X < X + 1.0 or not. // Infinity and NaN are not numbers in that sense. A run that let 1.0E308 * // 10.0 become infinity, and X - X become NaN, broke both claims while the // checker proved them. Such a run now stops at the overflow instead. // RUN-STOPS: overflow@7 DECLARE X, Y : FLOAT X <- 1.0E308 * 10.0 Y <- X - X ASSERT Y == Y OUTPUT Y