[PATCH v2 0/4] Optimize flt-32/dbl-64 remainder implementation

[Adhemerval Zanella]

The commit 34b9f8bc17 provides an optimized fmod implementation; use
the same strategy used for remainderf and implement the double variant
on top of fmod.

Also optimized the flt-32 variant with some micro-optimizations.

Adhemerval Zanella (4):
  benchtests: Add remainder benchtest
  benchtests: Add remainderf benchtest
  math: Optimize dbl-64 remainder implementation
  math: Optimize flt-32 remainder implementation

 benchtests/Makefile                   |    4 +
 benchtests/remainder-inputs           | 2182 +++++++++++++++++++++++++
 benchtests/remainderf-inputs          | 2182 +++++++++++++++++++++++++
 sysdeps/ieee754/dbl-64/e_remainder.c  |  194 +--
 sysdeps/ieee754/dbl-64/urem.h         |   45 -
 sysdeps/ieee754/flt-32/e_remainderf.c |  103 +-
 6 files changed, 4474 insertions(+), 236 deletions(-)
 create mode 100644 benchtests/remainder-inputs
 create mode 100644 benchtests/remainderf-inputs
 delete mode 100644 sysdeps/ieee754/dbl-64/urem.h

-- 
2.43.0