v2.4.0

What's Changed

• Improve comparisons for decimal32 by @mborland in #729
• Improve u128 x u128 by @mborland in #731
• Fix clang19 by @mborland in #732
• Improved comparisons for decimal64 by @mborland in #730
• Faster dec128 mul by @mborland in #733
• Add support for b2 modular builds and boostlook by @mborland in #734
• Add in the initial support for DPD conversions by @mborland in #738
• Reduce language requirement for constexpr BID conversions by @mborland in #739
• Implement 64 bit DPD conversions by @mborland in #742
• Add 128-bit DPD conversions by @mborland in #743

Full Changelog: v2.3.1...v2.4.0

v2.3.1

What's Changed

• Remove u128 x64 intrinsic pessimization by @mborland in #725

Full Changelog: v2.3.0...v2.3.1

v2.3.0

What's Changed

• Remove optional dependency and overloads of boost.type_traits by @mborland in #721
• Add intel compiler to CI by @mborland in #722
• Fix a typo by @jk-jeon in #723

New Contributors

• @jk-jeon made their first contribution in #723

Full Changelog: v2.2.0...v2.3.0

v2.2.0

What's Changed

• Simplify non-finite number functions and handling by @mborland in #699
• Add ppc64le runs to CI by @mborland in #700
• Improve comparison operators for decimal32 and decimal32_fast by @mborland in #703
• Better 64 bit comparisons by @mborland in #705
• Improve 128 bit comparisons by @mborland in #709
• decimal32 operator< without normalization in the general case by @mborland in #711
• Decimal32 fast comps v2 by @mborland in #714

Full Changelog: v2.1.0...v2.2.0

v2.1.0

What's Changed

• elliptic integral improvements by @ckormanyos in #624
• Improve multiplication for decimal128_fast by @mborland in #662
• Ellint coverage by @ckormanyos in #666
• Remove unneeded calculation of modulus from division by @mborland in #667
• Replace NaN types for invalid arguments by @mborland in #671
• Improve fast type coverage by @mborland in #669
• Funcs and cover by @ckormanyos in #670
• Improved basic operators for 32 and 64 bit types by @mborland in #677
• Fix drone ASAN runs by @mborland in #682
• Refactor decimal128 and decimal128_fast by @mborland in #683
• Improve testing coverage by @mborland in #685
• Improve 32-bit addition implementation by @mborland in #689
• Improve 64 and 128 bit addition implementations by @mborland in #691
• Remove recursive call from 32 bit operator- by @mborland in #693
• Improve 64 and 128 bit subtraction by @mborland in #696

Full Changelog: v2.0.0...v2.1.0

v2.0.0

This release adds the types decimal32_fast, decimal64_fast and decimal128_fast. They offer the same numerical results as their related non-fast type, but without the layout constraints of IEEE 754. See: https://cppalliance.org/decimal/decimal.html#Benchmarks

What's Changed

• Fix promotion system with inclusion of decimal32_fast by @mborland in #589
• Start tuning cmath functions by @ckormanyos in #580
• Add implementations of for decimal32_fast by @mborland in #590
• Fixed width rounding by @mborland in #586
• Drone: divide GCC 12 ASAN 32 jobs by @sdarwin in #597
• Riemann zeta by @ckormanyos in #593
• Improve comparison functions by @mborland in #598
• decimal64_fast basis by @mborland in #599
• Add support for mixed decimal64_fast and Integer operations by @mborland in #606
• decimal64_fast cmath additions by @mborland in #609
• Add bindings and testing of decimal64_fast STL like functions by @mborland in #611
• Add Decimal64 fast to benchmarks by @mborland in #615
• More func by @ckormanyos in #607
• Add to and from BID for all types by @mborland in #625
• Make dec32_fast even faster by @mborland in #620
• Offer fast math macro by @mborland in #627
• Begin implementing decimal128_fast by @mborland in #631
• Implement basic arithmetic for decimal128_fast by @mborland in #635
• Implement mixed arithmetic and comparisions for decimal128_fast by @mborland in #636
• Test and fix discrepency between decimal128 and decimal128_fast by @mborland in #642
• Add cmath support to decimal128_fast by @mborland in #638
• Different improvement of counting methods by @mborland in #647
• Implement beta by @mborland in #649
• Improvements to decimal64_fast by @mborland in #644
• Finish decimal128_fast by @mborland in #654
• Improve dec64 multiplication by @mborland in #655
• Faster dec128 fast by @mborland in #658
• Improve num_digits() for uint256_t by @mborland in #660

Full Changelog: v1.2.0...v2.0.0

v1.2.0

What's Changed

• Additional examples and documentation by @mborland in #553
• Refactor decimal32 FMA implementation by @mborland in #555
• Fix potential overflows in conversions to integral types by @mborland in #561
• Improve some basic 128-bit operations by @mborland in #560
• Add buffer sizing for charconv by @mborland in #562
• Add GCC/libdfp benchmarks by @mborland in #564
• Delete operator"" (long double) by @mborland in #569
• Start finish gammas but lgamma-128 still needed by @ckormanyos in #551
• Study mul256 by @ckormanyos in #574
• Update docs by @ckormanyos in #579
• Remove loop from constructors by @mborland in #577
• Reduce decoding branches by @mborland in #583
• Fix decimal64 fma double rounding by @mborland in #584

Full Changelog: v1.1.0...v1.2.0

v1.1.0

What's Changed

• Impl of cosh/exp at 128 bits and a few tests by @ckormanyos in #527
• Sinh128 by @ckormanyos in #529
• Tanh128 by @ckormanyos in #532
• Logarithmic funcs by @ckormanyos in #536
• Faster zero removal by @mborland in #526
• Drone: divide GCC 7 jobs by @sdarwin in #542
• Restore mistakenly removed promotions by @ckormanyos in #540
• Log10 by @ckormanyos in #544
• Add mingw to CI by @mborland in #545
• Handle pure powers of 10 in pow() function by @ckormanyos in #546
• Impl expm1 64/128 and other small repairs by @ckormanyos in #548
• Preliminary step simplify atan() by @ckormanyos in #549

Full Changelog: v1.0.1...v1.1.0

v1.0.1

What's Changed

• Improve operations expanding or shrinking significand by @mborland in #518
• Fix decimal128 tgamma by @ckormanyos in #521
• Fix decimal128 fma by @mborland in #523

Full Changelog: v1.0.0...v1.0.1