3 History of Major Changes to SoftFloat, up to Release 2a
9 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
10 Release 2a (1998 December)
12 -- Added functions to convert between 64-bit integers (int64) and all
13 supported floating-point formats.
15 -- Fixed a bug in all 64-bit-version square root functions except
16 `float32_sqrt' that caused the result sometimes to be off by 1 unit in
17 the last place (1 ulp) from what it should be. (Bug discovered by Paul
20 -- Improved the makefiles.
22 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
25 -- Created the 64-bit (bits64) version, adding the floatx80 and float128
28 -- Changed the source directory structure, splitting the sources into a
29 `bits32' and a `bits64' version. Renamed `environment.h' to `milieu.h'
30 (to avoid confusion with environment variables).
32 -- Fixed a small error that caused `float64_round_to_int' often to round the
33 wrong way in nearest/even mode when the operand was between 2^20 and 2^21
34 and halfway between two integers.
36 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
37 Release 1a (1996 July)
39 -- Corrected a mistake that caused borderline underflow cases not to raise
40 the underflow flag when they should have. (Problem reported by Doug
43 -- Added the `float_detect_tininess' variable to control whether tininess is
44 detected before or after rounding.
46 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
51 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -