stillwater-sc / universal / 28439329267

85%

master: 84%

Build Type

push

github

Committed by web-flow

Commit Message

feat(efloat): implement mathematical minimum, maximum, and difference library (#1128)

* feat(efloat): implement mathematical minimum, maximum, and difference library (#1117)

Implement min(), max(), fmin(), fmax() (with NaN-suppressing capabilities), and
fdim() (positive difference) inside math/minmax.hpp.

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

Create minmax.cpp inside elastic/efloat/math/ to verify all minimum, maximum, and
difference functions, testing boundary cases and NaN suppression.

Resolves #1117
Relates to parent Issue #1092, Epic #1101

* fix(efloat): resolve CodeRabbit review comments on minmax library

1. Implement IEEE-754 signed-zero tie-breaking in fmin() and fmax() inside math/minmax.hpp:
   return negative zero for fmin(-0.0, +0.0), and return positive zero for
   fmax(-0.0, +0.0) by performing an explicit zero-sign check on equal-zero ties.

2. Update regressions: expand minmax.cpp to cover fmin/fmax signed-zero tie-breaking and
   fdim NaN-propagation.

Resolves #1117, PR #1128

Coverage Stats

21 of 23 new or added lines in 1 file covered. (91.3%)

22 existing lines in 4 files now uncovered.

49164 of 58109 relevant lines covered (84.61%)

6684200.88 hits per line