From 17fd485a3bb82021077c815451bb331bf8ff8f42 Mon Sep 17 00:00:00 2001
From: chethega <chethega@gmail.com>
Date: Sun, 21 Oct 2018 00:16:04 +0200
Subject: [PATCH 1/3] speed up logical indexing by bitarray

---
 base/bitarray.jl         |  1 +
 base/multidimensional.jl | 26 +++++++++++++++-----------
 2 files changed, 16 insertions(+), 11 deletions(-)

diff --git a/base/bitarray.jl b/base/bitarray.jl
index 526f027c6aa9b..120ce0b5c93fa 100644
--- a/base/bitarray.jl
+++ b/base/bitarray.jl
@@ -84,6 +84,7 @@ IndexStyle(::Type{<:BitArray}) = IndexLinear()
 const _msk64 = ~UInt64(0)
 @inline _div64(l) = l >> 6
 @inline _mod64(l) = l & 63
+@inline _BLSR(x)= x & (x-1) #zeros the last set bit. Has native instruction on many archs. needed in multidimensional.jl
 @inline _msk_end(l::Integer) = _msk64 >>> _mod64(-l)
 @inline _msk_end(B::BitArray) = _msk_end(length(B))
 num_bit_chunks(n::Int) = _div64(n+63)
diff --git a/base/multidimensional.jl b/base/multidimensional.jl
index a24745b952e19..15b676075b05d 100644
--- a/base/multidimensional.jl
+++ b/base/multidimensional.jl
@@ -523,19 +523,23 @@ end
         L.mask[idx] && return (idx, s)
     end
 end
-# When wrapping a BitArray, lean heavily upon its internals -- this is a common
-# case. Just use the Int index and count as its state.
-@inline function iterate(L::LogicalIndex{Int,<:BitArray}, s=(0,1))
-    s[2] > length(L) && return nothing
-    i, n = s
+# When wrapping a BitArray, lean heavily upon its internals.
+@inline function iterate(L::Base.LogicalIndex{Int,<:BitArray})
+    L.sum == 0 && return nothing
     Bc = L.mask.chunks
-    while true
-        if Bc[_div64(i)+1] & (UInt64(1)<<_mod64(i)) != 0
-            i += 1
-            return (i, (i, n+1))
-        end
-        i += 1
+    return iterate(L::Base.LogicalIndex{Int,<:BitArray}, (1, @inbounds Bc[1]))
+end
+@propagate_inbounds function iterate(L::Base.LogicalIndex{Int,<:BitArray}, s)
+    Bc = L.mask.chunks
+    i1, c = s
+    while c==0
+        i1 == length(Bc) && return nothing
+        i1 += 1
+        c = Bc[i1]
     end
+    tz = trailing_zeros(c) + 1
+    c = _BLSR(c)
+    return ((i1-1)<<6 + tz, (i1, c))
 end
 
 @inline checkbounds(::Type{Bool}, A::AbstractArray, I::LogicalIndex{<:Any,<:AbstractArray{Bool,1}}) =

From ada8bc8e9b22469351e0afe5bf9e043a9a9e4381 Mon Sep 17 00:00:00 2001
From: chethega <chethega@gmail.com>
Date: Mon, 22 Oct 2018 19:16:19 +0200
Subject: [PATCH 2/3] changed inbounds / OOB behavior to match array iterators

switched spelling of _blsr
---
 base/bitarray.jl         | 2 +-
 base/multidimensional.jl | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/base/bitarray.jl b/base/bitarray.jl
index 120ce0b5c93fa..aea54aa3385bf 100644
--- a/base/bitarray.jl
+++ b/base/bitarray.jl
@@ -84,7 +84,7 @@ IndexStyle(::Type{<:BitArray}) = IndexLinear()
 const _msk64 = ~UInt64(0)
 @inline _div64(l) = l >> 6
 @inline _mod64(l) = l & 63
-@inline _BLSR(x)= x & (x-1) #zeros the last set bit. Has native instruction on many archs. needed in multidimensional.jl
+@inline _blsr(x)= x & (x-1) #zeros the last set bit. Has native instruction on many archs. needed in multidimensional.jl
 @inline _msk_end(l::Integer) = _msk64 >>> _mod64(-l)
 @inline _msk_end(B::BitArray) = _msk_end(length(B))
 num_bit_chunks(n::Int) = _div64(n+63)
diff --git a/base/multidimensional.jl b/base/multidimensional.jl
index 15b676075b05d..ecfd0cb461628 100644
--- a/base/multidimensional.jl
+++ b/base/multidimensional.jl
@@ -529,16 +529,16 @@ end
     Bc = L.mask.chunks
     return iterate(L::Base.LogicalIndex{Int,<:BitArray}, (1, @inbounds Bc[1]))
 end
-@propagate_inbounds function iterate(L::Base.LogicalIndex{Int,<:BitArray}, s)
+@inline function iterate(L::Base.LogicalIndex{Int,<:BitArray}, s)
     Bc = L.mask.chunks
     i1, c = s
     while c==0
-        i1 == length(Bc) && return nothing
+        i1 % UInt >= length(Bc) % UInt && return nothing
         i1 += 1
-        c = Bc[i1]
+        @inbounds c = Bc[i1]
     end
     tz = trailing_zeros(c) + 1
-    c = _BLSR(c)
+    c = _blsr(c)
     return ((i1-1)<<6 + tz, (i1, c))
 end
 

From 1328295bb4808f334f7674a2685534b7ac053171 Mon Sep 17 00:00:00 2001
From: Matt Bauman <mbauman@gmail.com>
Date: Wed, 24 Oct 2018 02:18:41 +0200
Subject: [PATCH 3/3] Update base/multidimensional.jl

Co-Authored-By: chethega <chethega@gmail.com>
---
 base/multidimensional.jl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/base/multidimensional.jl b/base/multidimensional.jl
index ecfd0cb461628..d79715299fffd 100644
--- a/base/multidimensional.jl
+++ b/base/multidimensional.jl
@@ -527,7 +527,7 @@ end
 @inline function iterate(L::Base.LogicalIndex{Int,<:BitArray})
     L.sum == 0 && return nothing
     Bc = L.mask.chunks
-    return iterate(L::Base.LogicalIndex{Int,<:BitArray}, (1, @inbounds Bc[1]))
+    return iterate(L, (1, @inbounds Bc[1]))
 end
 @inline function iterate(L::Base.LogicalIndex{Int,<:BitArray}, s)
     Bc = L.mask.chunks