stillwater-sc / universal / 28368322669

85%

master: 84%

Build Type

push

github

Committed by web-flow

Commit Message

feat(efloat): implement mathematical fractional and remainder library (#1116) (#1127)

Implement fmod(), remainder(), drem(), frac(), modf(), copysign(), nextafter(),
and nexttoward() (casting to double to bypass long-double format differences)
inside math/fractional.hpp.

Overhaul comparison logic in efloat_impl.hpp:
1. Fix operator< by adding explicit iszero() checks at the top. This prevents any positive
   sub-unit normal value (scale < 0) from being misclassified as less than zero.
2. Fix compare_limbs() by making it bounds-safe, looping to the maximum size and reading
   0u for missing elements, avoiding all out-of-bounds reads and segfaults.

Register math/fractional.hpp inside our master efloat mathlib.hpp header.

Create fractional.cpp inside elastic/efloat/math/ to verify all fractional, remainder,
and adjacent step functions against exact boundaries and step inequalities.

Resolves #1116
Relates to parent Issue #1092, Epic #1101

Coverage Stats

57 of 89 new or added lines in 2 files covered. (64.04%)

3 existing lines in 1 file now uncovered.

49160 of 58086 relevant lines covered (84.63%)

6690180.75 hits per line