win: improve timestamp precision in TIME_T_TO_FILETIME

The TIME_T_TO_FILETIME macro currently truncates the fractional part of
the tick count when converting from double. This leads to precision
errors similar to those fixed in the Unix implementation.

This commit adds rounding to the macro to ensure better consistency and
pass the new nanosecond precision tests.
This commit is contained in:
Bitshifter-9 2025-11-27 09:39:57 +05:30
parent 19d7a319ba
commit 862794da51

File diff suppressed because it is too large Load Diff