1. f11ebf8 corrected to the best known range (fixed after X.696 Corrigendum 1) by Lev Walkin · 7 years ago
  2. c44edc0 take into account X696 Corrigendum 1 (05/2017) by Lev Walkin · 7 years ago
  3. b545070 generate float instead of double if constraints suggest this by Lev Walkin · 7 years ago
  4. ab1d1e1 OER support and randomized testing for the REAL (floating point) type by Lev Walkin · 7 years ago