v0.6.13

LoopVectorization v0.6.13

Diff since v0.6.12

Closed issues:

• ERROR: MethodError: no method matching promote_vtype(::Type{VectorizationBase._MM{8}}, ::Type{Int32}) (#54)
• ERROR: TypeError: non-boolean (UInt16) used in boolean context (#56)

v0.6.12

LoopVectorization v0.6.12

Diff since v0.6.11

v0.6.11

LoopVectorization v0.6.11

Diff since v0.6.10

Closed issues:

• ERROR: InexactError: trunc(Int64, Inf) (#40)
• MethodError from @avx tanh.(x) (#47)
• Wrong result with Int16 and offset index in for loop (#52)
• ERROR: MethodError: no method matching Int16(::VectorizationBase.SVec{8,Int32}) (#53)

v0.6.10

LoopVectorization v0.6.10

Diff since v0.6.9

Closed issues:

• matrix assignment with offset other than 1 (#48)
• wrong result with index offset (#50)

Merged pull requests:

• Install TagBot as a GitHub Action (#49) (@JuliaTagBot)

v0.6.9

v0.6.9 (2020-02-07)

Diff since v0.6.8

v0.6.8

v0.6.8 (2020-02-06)

Diff since v0.6.7

Closed issues:

• ERROR: UndefVarError: ##op#5321 not defined (#45)
• ERROR: MethodError: no method matching <<(::VectorizationBase._MM{8}, ::Int32) (#44)

v0.6.7

v0.6.7 (2020-02-05)

Diff since v0.6.6

Closed issues:

• Is it possible to vectorize filter? (#42)
• for loop with a negative step (#21)

v0.6.6

v0.6.6 (2020-02-03)

Diff since v0.6.5

v0.6.5

v0.6.5 (2020-02-03)

Diff since v0.6.4

v0.6.4

v0.6.4 (2020-02-03)

Diff since v0.6.3