v1.7.0

What's Changed

• perf: accelerate 32-byte big-endian conversion on ARM64; add ARM CI by @kamilchodola in #102
• Avoid SIMD materialization in UInt256 constructors and factories by @kamilchodola in #105
• Optimize MultiplyMod by the maximum modulus by @kamilchodola in #106
• Optimize narrow 256x256-to-512-bit multiplication by @kamilchodola in #103
• Optimize UInt256.CompareTo with direct limb comparison by @kamilchodola in #107
• Optimize UInt256 division and modulus fast paths by @kamilchodola in #104
• Optimize UInt256 logical shifts with a portable funnel shift by @kamilchodola in #109
• Optimize UInt256 subtraction with speculative borrow propagation and a one-limb scalar path by @kamilchodola in #112
• Optimize UInt256 addition with speculative carry propagation and a one-limb scalar path by @kamilchodola in #108
• perf: spell out the software widening multiply by @LukaszRozmej in #113
• Dispatch UInt256.Multiply on limb shapes with inlined per-shape kernels and one 32-byte result store by @benaadams in #114
• Optimize Int256 negativity-only gates by @kamilchodola in #110
• Answer Mod without computing the quotient by @benaadams in #115
• Cover the remainder-only Mod kernels' rare Knuth paths by @LukaszRozmej in #116
• Bump version prefix from 1.6.0 to 1.7.0 by @kamilchodola in #117

Full Changelog: v1.6.0...v1.7.0

v1.6.0

What's Changed

• Optimize parsing, ExpMod, and MulMod by 64-bit modulus by @kamilchodola in #84
• perf: split SIMD fast paths into inlinable helpers by @maximmenshikov in #95
• ADDMOD: skip wasted DivRems in the 64-bit-modulus reduction for small sums by @kamilchodola in #93
• Int256: top-bit sign test in operator<, limb-based IsZero/IsOne (EVM SLT/SGT path) by @kamilchodola in #97
• LT/GT: route LessThan to scalar on non-AVX-512 (keep the AVX-512 path) by @kamilchodola in #91
• Optimize scalar UInt256 multiplication by @kamilchodola in #99
• Improve UInt256 hash performance by @benaadams in #100
• Optimize MultiplyMod: reuse div remainder in Knuth reduction (-27% to -37%) by @benaadams in #101
• Make (UInt256)BigInteger cast allocation-free by @kamilchodola in #85
• Speed up PaddedBytes with block writes instead of per-byte loop by @kamilchodola in #86
• Vectorize UInt256.ToBigEndian(Span) for the 32-byte word by @kamilchodola in #92
• ADDMOD: skip wasted reductions in the portable 64-bit-modulus remainder (zk guest / ARM path) by @kamilchodola in #96
• Int256: AggressiveInlining hints for hot accessors by @kamilchodola in #90
• Support standard and zkEVM variants in a single NuGet package by @maximmenshikov in #94

New Contributors

• @kamilchodola made their first contribution in #84
• @maximmenshikov made their first contribution in #95

Full Changelog: v1.5.0...v1.6.0

v1.5.0

What's Changed

• Add AppContext switch to allow hash seed to be deterministic by @rubo in #79
• Optimize and harden GetHashCode with Aes by @benaadams in #83

Full Changelog: v1.4.0...v1.5.0

v1.4.0

What's Changed

• Add comprehensive GitHub Copilot instructions for int256 repository onboarding by @Copilot in #54
• Fix avx512 multiply and optimize by @benaadams in #56
• Optimize Lessthan by @benaadams in #57
• Test on all OS by @benaadams in #59
• Optimize add by @benaadams in #58
• Add Vector256 path to lessthan by @benaadams in #60
• Optimize Multiply further by @benaadams in #61
• Remove unused r properties by @benaadams in #63
• Improve benchmarks by @benaadams in #64
• Faster benchmarks, sp fixes by @benaadams in #65
• Use correct Vector acceleration check by @benaadams in #66
• Increase tests by @benaadams in #67
• Improve benchmarks (remove == shortcuts) by @benaadams in #68
• Better divide benchmark by @benaadams in #69
• Benchmarks (better Avx512/Avx2/None) measurements by @benaadams in #70
• Implement INumber interfaces for UInt256 and Int256 by @Copilot in #62
• Revert "Implement INumber interfaces for UInt256 and Int256 (#62)" by @benaadams in #71
• Benchmarks add disable Avx2 by @benaadams in #72
• Improve Testing Matrix by @benaadams in #74
• Improve Divide and AddMod by @benaadams in #73
• Improve Exp performance by @benaadams in #75
• Tweaks by @benaadams in #76
• Faster BigEndian parsing by @benaadams in #77
• CompareTo in overload by @benaadams in #78
• Improve Mulmod perf by @benaadams in #80
• Update version prefix to 1.4.0 by @benaadams in #81

New Contributors

• @Copilot made their first contribution in #54

Full Changelog: v1.3.6...v1.4.0

v1.3.6

What's Changed

• Less branchy Udivrem by @Scooletz in #51
• Use native ulong shifts when custom shift is not needed by @Hyodar in #50
• Increase tests by @benaadams in #52

New Contributors

• @Hyodar made their first contribution in #50

Full Changelog: v1.3.5...v1.3.6

v1.3.5

What's Changed

• Improve vectorization by @benaadams in #49

Full Changelog: v1.3.4...v1.3.5

v1.3.4

What's Changed

• Vectorized Multiply (on Avx512) by @benaadams in #45
• Migrate to .slnx by @rubo in #46
• Migrate to Microsoft.Testing.Platform by @rubo in #47

Full Changelog: v1.3.3...v1.3.4

v1.3.3

What's Changed

• Allow input params as output params by @benaadams in #44

Full Changelog: v1.3.2...v1.3.3

v1.3.2

What's Changed

• Fix Add by @benaadams in #43

Full Changelog: v1.3.1...v1.3.2

v1.3.1

What's Changed

• Improve AddImpl and SubtractImpl with Avx512 by @benaadams in #41
• Add fast Add, Multiply for numbers less than 2^64 by @benaadams in #42

Full Changelog: v1.3.0...v1.3.1