diff --git a/Project.toml b/Project.toml index 16764a9..206a58f 100644 --- a/Project.toml +++ b/Project.toml @@ -1,7 +1,7 @@ name = "StrideArrays" uuid = "d1fa6d79-ef01-42a6-86c9-f7c551f8593b" authors = ["Chris Elrod and contributors"] -version = "0.1.5" +version = "0.1.6" [deps] ArrayInterface = "4fba245c-0d91-5ea0-9b3e-6abc04ee57a9" @@ -24,7 +24,7 @@ SLEEFPirates = "0.6.13" Static = "0.2.4" StrideArraysCore = "0.1.3" ThreadingUtilities = "0.4" -VectorizationBase = "0.19.12" +VectorizationBase = "0.19.32" VectorizedRNG = "0.2.8" julia = "1.5" diff --git a/src/broadcast.jl b/src/broadcast.jl index d367d8f..079f64f 100644 --- a/src/broadcast.jl +++ b/src/broadcast.jl @@ -151,7 +151,7 @@ function add_fs_array!(ls::LoopVectorization.LoopSet, destname::Symbol, bcname:: ref = Symbol[] # aref = LoopVectorization.ArrayReference(bcname, ref) vptrbc = LoopVectorization.vptr(bcname) - LoopVectorization.add_vptr!(ls, bcname, vptrbc, true, false) #TODO: is this necessary? + LoopVectorization.add_vptr!(ls, bcname, vptrbc, true) #TODO: is this necessary? offset = 0 Rnew = Int[] for (i,n) ∈ enumerate(indexes) @@ -183,15 +183,10 @@ function add_fs_array!(ls::LoopVectorization.LoopSet, destname::Symbol, bcname:: sp = sort_indices!(mref, Rnew, C) if sp === nothing LoopVectorization.pushprepreamble!(ls, Expr(:(=), bctemp, bcname)) - # LoopVectorization.add_vptr!(ls, bcname, vptrbc, true, false) else ssp = Expr(:tuple); append!(ssp.args, sp) ssp = Expr(:call, Expr(:curly, :StaticInt, ssp)) LoopVectorization.pushprepreamble!(ls, Expr(:(=), bctemp, Expr(:call, :permutedims, bcname, ssp))) - # LoopVectorization.add_vptr!(ls, bctemp, vptrbc, true, false) - # vptemp = gensym(vptrbc) - # LoopVectorization.add_vptr!(ls, bcname, vptemp, true, false) - # LoopVectorization.pushprepreamble!(Expr(:(=), vptrbc, Expr(:call, :PermutedDimsArray, vptemp, ssp))) end loadop = LoopVectorization.add_simple_load!(ls, destname, mref, mref.ref.indices, elementbytes) LoopVectorization.doaddref!(ls, loadop)