v3.7.0

SIMD v3.7.0

Diff since v3.6.0

Merged pull requests:

• fix IEEE compliancy in floating point min/max reductions on LLVM verson that supports it (#130) (@KristofferC)

Closed issues:

• Support for complex number functions for currently supported types (Ex: real,conj) (#126)
• reduce(min, Vec) doesn't respect IEEE 754 (#129)
• Extract array value from a one-hot Vec Bool mask (#132)

v3.6.0

SIMD v3.6.0

Diff since v3.5.0

Merged pull requests:

• Define real, imag, conj, angle, and abs2 for Vec{N,<:Real} (#127) (@favba)

Closed issues:

• @fastmath not applying to additions sometimes (#108)
• No type promotion for arithmetics of mixed types (#128)

v3.5.0

SIMD v3.5.0

Diff since v3.4.7

v3.4.7

SIMD v3.4.7

Diff since v3.4.6

v3.4.6

SIMD v3.4.6

Diff since v3.4.5

Merged pull requests:

• Fix uneven trailing whitespace in strings (#114) (@c42f)
• fix some shift operations to agree with Base (and prevent undefined behavior) (#119) (@KristofferC)

Closed issues:

• vload on Matrix{Float64} (#116)

v3.4.5

SIMD v3.4.5

Diff since v3.4.4

Closed issues:

• any function is slow (#109)

Merged pull requests:

• NFC remove deprecated Vararg{<:Any} (#105) (@oscardssmith)
• Migrate from SnoopPrecompile to PrecompileTools (#113) (@timholy)

v3.4.4

SIMD v3.4.4

Diff since v3.4.3

Closed issues:

• Support @fastmath min (#103)

Merged pull requests:

• Add @propagate_inbounds to getindex/setindexi! on Vec (#102) (@Seelengrab)
• add fastmath for min and max (#104) (@KristofferC)

v3.4.3

SIMD v3.4.3

Diff since v3.4.2

Merged pull requests:

• add some very basic precompile workload (#101) (@KristofferC)

v3.4.2

SIMD v3.4.2

Diff since v3.4.1

Merged pull requests:

• fix llvm intrinsic function for pow, powi (#100) (@KristofferC)

v3.4.1

SIMD v3.4.1

Diff since v3.4.0

Merged pull requests:

• fix tests on Julia 1.8+ (#96) (@KristofferC)