| Name: David M. Gay's floating point routines |
| URL: http://www.netlib.org/fp/ |
| License: MIT-like |
| |
| Original dtoa.c file can be found at <http://www.netlib.org/fp/dtoa.c>. |
| Original g_fmt.c file can be found at <http://www.netlib.org/fp/g_fmt.c>. |
| |
| List of changes made to original code: |
| - wrapped functions in dmg_fp namespace |
| - renamed .c files to .cc |
| - added dmg_fp.h header |
| - added #define IEEE_8087 to dtoa.cc |
| - added #define NO_HEX_FP to dtoa.cc |
| - made some minor changes to allow clean compilation under g++ -Wall, see |
| gcc_warnings.patch. |
| - made some minor changes to build on 64-bit, see gcc_64_bit.patch. |
| - made minor changes for -Wextra for Mac build, see mac_wextra.patch |
| - crash fix for running with reduced CPU float precision, see |
| float_precision_crash.patch and crbug.com/123157 |
| - fixed warnings under msvc, see msvc_warnings.patch |
| - fixed -Wchar-subscripts warning in "if (!hexdig['0'])" in dtoa.cc |
| - removed the use of 'register' keyword from g_fmt.cc |
