From e69f2fdeda520f90d038d3599f1ac3c5137a92bb Mon Sep 17 00:00:00 2001
From: Simone Silvestri <33547697+simone-silvestri@users.noreply.github.com>
Date: Sun, 11 Jun 2023 16:45:38 -0400
Subject: [PATCH 1/9] first commit

---
 Manifest.toml             | 244 ++++++++++++++++++++++++++++++++++++++
 src/KernelAbstractions.jl |   8 +-
 src/macros.jl             |   5 +
 src/nditeration.jl        |  61 +++++++---
 4 files changed, 296 insertions(+), 22 deletions(-)
 create mode 100644 Manifest.toml

diff --git a/Manifest.toml b/Manifest.toml
new file mode 100644
index 000000000..791f408f4
--- /dev/null
+++ b/Manifest.toml
@@ -0,0 +1,244 @@
+# This file is machine-generated - editing it directly is not advised
+
+julia_version = "1.8.0"
+manifest_format = "2.0"
+project_hash = "9033168e08cf56d36e87b64ae995845922bec33c"
+
+[[deps.Adapt]]
+deps = ["LinearAlgebra", "Requires"]
+git-tree-sha1 = "76289dc51920fdc6e0013c872ba9551d54961c24"
+uuid = "79e6a3ab-5dfb-504d-930d-738a2a938a0e"
+version = "3.6.2"
+
+[[deps.ArgTools]]
+uuid = "0dad84c5-d112-42e6-8d28-ef12dabb789f"
+version = "1.1.1"
+
+[[deps.Artifacts]]
+uuid = "56f22d72-fd6d-98f1-02f0-08ddc0907c33"
+
+[[deps.Atomix]]
+deps = ["UnsafeAtomics"]
+git-tree-sha1 = "c06a868224ecba914baa6942988e2f2aade419be"
+uuid = "a9b6321e-bd34-4604-b9c9-b65b8de01458"
+version = "0.1.0"
+
+[[deps.Base64]]
+uuid = "2a0f44e3-6c83-55bd-87e4-b1978d98bd5f"
+
+[[deps.CEnum]]
+git-tree-sha1 = "eb4cb44a499229b3b8426dcfb5dd85333951ff90"
+uuid = "fa961155-64e5-5f13-b03f-caf6b980ea82"
+version = "0.4.2"
+
+[[deps.CompilerSupportLibraries_jll]]
+deps = ["Artifacts", "Libdl"]
+uuid = "e66e0078-7015-5450-92f7-15fbd957f2ae"
+version = "0.5.2+0"
+
+[[deps.Dates]]
+deps = ["Printf"]
+uuid = "ade2ca70-3891-5945-98fb-dc099432e06a"
+
+[[deps.Downloads]]
+deps = ["ArgTools", "FileWatching", "LibCURL", "NetworkOptions"]
+uuid = "f43a241f-c20a-4ad4-852c-f6b1247861c6"
+version = "1.6.0"
+
+[[deps.FileWatching]]
+uuid = "7b1f6079-737a-58dc-b8bc-7a2ca5c1b5ee"
+
+[[deps.InteractiveUtils]]
+deps = ["Markdown"]
+uuid = "b77e0a4c-d291-57a0-90e8-8db25a27a240"
+
+[[deps.JLLWrappers]]
+deps = ["Preferences"]
+git-tree-sha1 = "abc9885a7ca2052a736a600f7fa66209f96506e1"
+uuid = "692b3bcd-3c85-4b1f-b108-f13ce0eb3210"
+version = "1.4.1"
+
+[[deps.LLVM]]
+deps = ["CEnum", "LLVMExtra_jll", "Libdl", "Printf", "Unicode"]
+git-tree-sha1 = "5007c1421563108110bbd57f63d8ad4565808818"
+uuid = "929cbde3-209d-540e-8aea-75f648917ca0"
+version = "5.2.0"
+
+[[deps.LLVMExtra_jll]]
+deps = ["Artifacts", "JLLWrappers", "LazyArtifacts", "Libdl", "TOML"]
+git-tree-sha1 = "1222116d7313cdefecf3d45a2bc1a89c4e7c9217"
+uuid = "dad2f222-ce93-54a1-a47d-0025e8a3acab"
+version = "0.0.22+0"
+
+[[deps.LazyArtifacts]]
+deps = ["Artifacts", "Pkg"]
+uuid = "4af54fe1-eca0-43a8-85a7-787d91b784e3"
+
+[[deps.LibCURL]]
+deps = ["LibCURL_jll", "MozillaCACerts_jll"]
+uuid = "b27032c2-a3e7-50c8-80cd-2d36dbcbfd21"
+version = "0.6.3"
+
+[[deps.LibCURL_jll]]
+deps = ["Artifacts", "LibSSH2_jll", "Libdl", "MbedTLS_jll", "Zlib_jll", "nghttp2_jll"]
+uuid = "deac9b47-8bc7-5906-a0fe-35ac56dc84c0"
+version = "7.84.0+0"
+
+[[deps.LibGit2]]
+deps = ["Base64", "NetworkOptions", "Printf", "SHA"]
+uuid = "76f85450-5226-5b5a-8eaa-529ad045b433"
+
+[[deps.LibSSH2_jll]]
+deps = ["Artifacts", "Libdl", "MbedTLS_jll"]
+uuid = "29816b5a-b9ab-546f-933c-edad1886dfa8"
+version = "1.10.2+0"
+
+[[deps.Libdl]]
+uuid = "8f399da3-3557-5675-b5ff-fb832c97cbdb"
+
+[[deps.LinearAlgebra]]
+deps = ["Libdl", "libblastrampoline_jll"]
+uuid = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
+
+[[deps.Logging]]
+uuid = "56ddb016-857b-54e1-b83d-db4d58db5568"
+
+[[deps.MacroTools]]
+deps = ["Markdown", "Random"]
+git-tree-sha1 = "42324d08725e200c23d4dfb549e0d5d89dede2d2"
+uuid = "1914dd2f-81c6-5fcd-8719-6d5c9610ff09"
+version = "0.5.10"
+
+[[deps.Markdown]]
+deps = ["Base64"]
+uuid = "d6f4376e-aef5-505a-96c1-9c027394607a"
+
+[[deps.MbedTLS_jll]]
+deps = ["Artifacts", "Libdl"]
+uuid = "c8ffd9c3-330d-5841-b78e-0817d7145fa1"
+version = "2.28.0+0"
+
+[[deps.MozillaCACerts_jll]]
+uuid = "14a3606d-f60d-562e-9121-12d972cd8159"
+version = "2022.2.1"
+
+[[deps.NetworkOptions]]
+uuid = "ca575930-c2e3-43a9-ace4-1e988b2c1908"
+version = "1.2.0"
+
+[[deps.OpenBLAS_jll]]
+deps = ["Artifacts", "CompilerSupportLibraries_jll", "Libdl"]
+uuid = "4536629a-c528-5b80-bd46-f80d51c5b363"
+version = "0.3.20+0"
+
+[[deps.Pkg]]
+deps = ["Artifacts", "Dates", "Downloads", "LibGit2", "Libdl", "Logging", "Markdown", "Printf", "REPL", "Random", "SHA", "Serialization", "TOML", "Tar", "UUIDs", "p7zip_jll"]
+uuid = "44cfe95a-1eb2-52ea-b672-e2afdf69b78f"
+version = "1.8.0"
+
+[[deps.PrecompileTools]]
+deps = ["Preferences"]
+git-tree-sha1 = "9673d39decc5feece56ef3940e5dafba15ba0f81"
+uuid = "aea7be01-6a6a-4083-8856-8a6e6704d82a"
+version = "1.1.2"
+
+[[deps.Preferences]]
+deps = ["TOML"]
+git-tree-sha1 = "7eb1686b4f04b82f96ed7a4ea5890a4f0c7a09f1"
+uuid = "21216c6a-2e73-6563-6e65-726566657250"
+version = "1.4.0"
+
+[[deps.Printf]]
+deps = ["Unicode"]
+uuid = "de0858da-6303-5e67-8744-51eddeeeb8d7"
+
+[[deps.REPL]]
+deps = ["InteractiveUtils", "Markdown", "Sockets", "Unicode"]
+uuid = "3fa0cd96-eef1-5676-8a61-b3b8758bbffb"
+
+[[deps.Random]]
+deps = ["SHA", "Serialization"]
+uuid = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
+
+[[deps.Requires]]
+deps = ["UUIDs"]
+git-tree-sha1 = "838a3a4188e2ded87a4f9f184b4b0d78a1e91cb7"
+uuid = "ae029012-a4dd-5104-9daa-d747884805df"
+version = "1.3.0"
+
+[[deps.SHA]]
+uuid = "ea8e919c-243c-51af-8825-aaa63cd721ce"
+version = "0.7.0"
+
+[[deps.Serialization]]
+uuid = "9e88b42a-f829-5b0c-bbe9-9e923198166b"
+
+[[deps.Sockets]]
+uuid = "6462fe0b-24de-5631-8697-dd941f90decc"
+
+[[deps.SparseArrays]]
+deps = ["LinearAlgebra", "Random"]
+uuid = "2f01184e-e22b-5df5-ae63-d93ebab69eaf"
+
+[[deps.StaticArrays]]
+deps = ["LinearAlgebra", "Random", "StaticArraysCore", "Statistics"]
+git-tree-sha1 = "832afbae2a45b4ae7e831f86965469a24d1d8a83"
+uuid = "90137ffa-7385-5640-81b9-e52037218182"
+version = "1.5.26"
+
+[[deps.StaticArraysCore]]
+git-tree-sha1 = "6b7ba252635a5eff6a0b0664a41ee140a1c9e72a"
+uuid = "1e83bf80-4336-4d27-bf5d-d5a4f845583c"
+version = "1.4.0"
+
+[[deps.Statistics]]
+deps = ["LinearAlgebra", "SparseArrays"]
+uuid = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
+
+[[deps.TOML]]
+deps = ["Dates"]
+uuid = "fa267f1f-6049-4f14-aa54-33bafae1ed76"
+version = "1.0.0"
+
+[[deps.Tar]]
+deps = ["ArgTools", "SHA"]
+uuid = "a4e569a6-e804-4fa4-b0f3-eef7a1d5b13e"
+version = "1.10.0"
+
+[[deps.UUIDs]]
+deps = ["Random", "SHA"]
+uuid = "cf7118a7-6976-5b1a-9a39-7adc72f591a4"
+
+[[deps.Unicode]]
+uuid = "4ec0a83e-493e-50e2-b9ac-8f72acf5a8f5"
+
+[[deps.UnsafeAtomics]]
+git-tree-sha1 = "6331ac3440856ea1988316b46045303bef658278"
+uuid = "013be700-e6cd-48c3-b4a1-df204f14c38f"
+version = "0.2.1"
+
+[[deps.UnsafeAtomicsLLVM]]
+deps = ["LLVM", "UnsafeAtomics"]
+git-tree-sha1 = "ea37e6066bf194ab78f4e747f5245261f17a7175"
+uuid = "d80eeb9a-aca5-4d75-85e5-170c8b632249"
+version = "0.1.2"
+
+[[deps.Zlib_jll]]
+deps = ["Libdl"]
+uuid = "83775a58-1f1d-513f-b197-d71354ab007a"
+version = "1.2.12+3"
+
+[[deps.libblastrampoline_jll]]
+deps = ["Artifacts", "Libdl", "OpenBLAS_jll"]
+uuid = "8e850b90-86db-534c-a0d3-1478176c7d93"
+version = "5.1.1+0"
+
+[[deps.nghttp2_jll]]
+deps = ["Artifacts", "Libdl"]
+uuid = "8e850ede-7688-5339-a07c-302acd2aaf8d"
+version = "1.48.0+0"
+
+[[deps.p7zip_jll]]
+deps = ["Artifacts", "Libdl"]
+uuid = "3f19e933-33d8-53b3-aaab-bd5110c3b7a0"
+version = "17.4.0+0"
diff --git a/src/KernelAbstractions.jl b/src/KernelAbstractions.jl
index 756f40f94..9de4f0497 100644
--- a/src/KernelAbstractions.jl
+++ b/src/KernelAbstractions.jl
@@ -489,7 +489,7 @@ end
 
 include("nditeration.jl")
 using .NDIteration
-import .NDIteration: get
+import .NDIteration: get, getrange
 
 ###
 # Kernel closure struct
@@ -535,11 +535,13 @@ function partition(kernel, ndrange, workgroupsize)
         error(errmsg)
     end
 
+    offsets = nothing
+
     if static_ndrange <: StaticSize
         if ndrange !== nothing && ndrange != get(static_ndrange)
             error("Static NDRange ($static_ndrange) and launch NDRange ($ndrange) differ")
         end
-        ndrange = get(static_ndrange)
+        ndrange, offsets = getrange(static_ndrange)
     end
 
     if static_workgroupsize <: StaticSize
@@ -568,7 +570,7 @@ function partition(kernel, ndrange, workgroupsize)
         workgroupsize = CartesianIndices(workgroupsize)
     end
 
-    iterspace = NDRange{length(ndrange), static_blocks, static_workgroupsize}(blocks, workgroupsize)
+    iterspace = NDRange{length(ndrange), offsets, static_blocks, static_workgroupsize}(blocks, workgroupsize)
     return iterspace, dynamic
 end
 
diff --git a/src/macros.jl b/src/macros.jl
index e93bc386d..5078c52de 100644
--- a/src/macros.jl
+++ b/src/macros.jl
@@ -1,5 +1,8 @@
 import MacroTools: splitdef, combinedef, isexpr, postwalk
 
+@inline contiguousrange(range::NTuple{N, Int}, offset::NTuple{N, Int}) where N = 
+            Tuple(1+o:r+o for (r, o) in zip(range, offset))
+
 function find_return(stmt)
     result = false
     postwalk(stmt) do expr
@@ -54,6 +57,8 @@ function __kernel(expr, generate_cpu=true)
             Core.@__doc__ $name(dev) = $name(dev, $DynamicSize(), $DynamicSize())
             $name(dev, size) = $name(dev, $StaticSize(size), $DynamicSize())
             $name(dev, size, range) = $name(dev, $StaticSize(size), $StaticSize(range))
+            $name(dev, size, range::NTuple{N}, offset::NTuple{N}) where N = 
+                    $name(dev, $StaticSize(size), $StaticSize($contiguousrange(range, offset)))
             function $name(dev::Dev, sz::S, range::NDRange) where {Dev, S<:$_Size, NDRange<:$_Size}
                 if $isgpu(dev)
                     return $construct(dev, sz, range, $gpu_name)
diff --git a/src/nditeration.jl b/src/nditeration.jl
index d7598ae20..aaa48595f 100644
--- a/src/nditeration.jl
+++ b/src/nditeration.jl
@@ -13,21 +13,35 @@ abstract type _Size end
 struct DynamicSize <: _Size end
 struct StaticSize{S} <: _Size
     function StaticSize{S}() where S
-        new{S::Tuple{Vararg{Int}}}()
+        new{S::Tuple{Vararg}}()
     end
 end
 
 @pure StaticSize(s::Tuple{Vararg{Int}}) = StaticSize{s}() 
 @pure StaticSize(s::Int...) = StaticSize{s}() 
 @pure StaticSize(s::Type{<:Tuple}) = StaticSize{tuple(s.parameters...)}()
+@pure StaticSize(s::Tuple{Vararg{UnitRange{Int}}}) = StaticSize{s}()
 
 # Some @pure convenience functions for `StaticSize`
 @pure get(::Type{StaticSize{S}}) where {S} = S
 @pure get(::StaticSize{S}) where {S} = S
 @pure Base.getindex(::StaticSize{S}, i::Int) where {S} = i <= length(S) ? S[i] : 1
-@pure Base.ndims(::StaticSize{S}) where {S} = length(S)
-@pure Base.length(::StaticSize{S}) where {S} = prod(S)
+@pure Base.ndims(::StaticSize{S}) where {S}  = length(S)
+@pure Base.length(::StaticSize{S}) where {S} = prod(worksize.(S))
 
+@inline getrange(::StaticSize{S}) where {S} = worksize(S), offsets(S)
+@inline getrange(::Type{StaticSize{S}}) where {S} = worksize(S), offsets(S)
+
+@inline worksize(i::Tuple) = worksize.(i)
+@inline worksize(i::Int) = i
+@inline worksize(i::UnitRange) = length(i)
+@inline worksize(i::StepRange) = length(i)
+
+@inline offsets(i) = offsets.(i)
+@inline offsets(::NTuple{N, Int}) where N = nothing
+@inline offsets(::Int) = nothing
+@inline offsets(i::UnitRange) = i.start - 1
+@inline offsets(i::StepRange) = i.start - 1, i.step
 
 """
     NDRange
@@ -36,7 +50,7 @@ Encodes a blocked iteration space.
 
 # Example
 ```
-ndrange = NDRange{2, DynamicSize, DynamicSize}(CartesianIndices((256, 256)), CartesianIndices((32, 32)))
+ndrange = NDRange{2, nothing, DynamicSize, DynamicSize}(CartesianIndices((256, 256)), CartesianIndices((32, 32)))
 for block in ndrange
     for items in workitems(ndrange)
         I = expand(ndrange, block, items)
@@ -46,23 +60,23 @@ for block in ndrange
 end
 ```
 """
-struct NDRange{N, StaticBlocks, StaticWorkitems, DynamicBlock, DynamicWorkitems}
+struct NDRange{N, Offsets, StaticBlocks, StaticWorkitems, DynamicBlock, DynamicWorkitems}
     blocks::DynamicBlock
     workitems::DynamicWorkitems
 
-    function NDRange{N, B, W}() where {N, B, W}
-        new{N, B, W, Nothing, Nothing}(nothing, nothing)
+    function NDRange{N, O, B, W}() where {N, O, B, W}
+        new{N, O, B, W, Nothing, Nothing}(nothing, nothing)
     end
 
-    function NDRange{N, B, W}(blocks, workitems) where {N, B, W}
-        new{N, B, W, typeof(blocks), typeof(workitems)}(blocks, workitems)
+    function NDRange{N, O, B, W}(blocks, workitems) where {N, O, B, W}
+        new{N, O, B, W, typeof(blocks), typeof(workitems)}(blocks, workitems)
     end
 end
 
-@inline workitems(range::NDRange{N, B, W}) where {N,B,W<:DynamicSize} = range.workitems::CartesianIndices{N}
-@inline workitems(range::NDRange{N, B, W}) where {N,B,W<:StaticSize} = CartesianIndices(get(W))::CartesianIndices{N}
-@inline blocks(range::NDRange{N, B}) where {N,B<:DynamicSize} = range.blocks::CartesianIndices{N}
-@inline blocks(range::NDRange{N, B}) where {N,B<:StaticSize} = CartesianIndices(get(B))::CartesianIndices{N}
+@inline workitems(range::NDRange{N, O, B, W}) where {N,O,B,W<:DynamicSize} = range.workitems::CartesianIndices{N}
+@inline workitems(range::NDRange{N, O, B, W}) where {N,O,B,W<:StaticSize} = CartesianIndices(get(W))::CartesianIndices{N}
+@inline blocks(range::NDRange{N, O, B}) where {N,O,B<:DynamicSize} = range.blocks::CartesianIndices{N}
+@inline blocks(range::NDRange{N, O, B}) where {N,O,B<:StaticSize} = CartesianIndices(get(B))::CartesianIndices{N}
 
 import Base.iterate
 @inline iterate(range::NDRange) = iterate(blocks(range))
@@ -70,12 +84,22 @@ import Base.iterate
 
 Base.length(range::NDRange) = length(blocks(range))
 
-@inline function expand(ndrange::NDRange{N}, groupidx::CartesianIndex{N}, idx::CartesianIndex{N}) where N
+@inline function expand(ndrange::NDRange{N, nothing}, groupidx::CartesianIndex{N}, idx::CartesianIndex{N}) where N
     nI = ntuple(Val(N)) do I
         Base.@_inline_meta
         stride = size(workitems(ndrange), I)
         gidx = groupidx.I[I]
-        (gidx-1)*stride + idx.I[I]
+        (gidx-1)*stride + idx.I[I] 
+    end
+    CartesianIndex(nI)
+end
+
+@inline function expand(ndrange::NDRange{N, Offsets}, groupidx::CartesianIndex{N}, idx::CartesianIndex{N}) where {N, Offsets}
+    nI = ntuple(Val(N)) do I
+        Base.@_inline_meta
+        stride = size(workitems(ndrange), I)
+        gidx = groupidx.I[I]
+        (gidx-1)*stride + idx.I[I] + Offsets[I]
     end
     CartesianIndex(nI)
 end
@@ -116,12 +140,11 @@ needs to perform dynamic bounds-checking.
     end
     let workgroupsize = workgroupsize
         dynamic = Ref(false)
-        blocks = ntuple(Val(length(ndrange))) do I
+        blocks  = ntuple(Val(length(ndrange))) do I
             Base.@_inline_meta
-            dynamic[] |= mod(ndrange[I], workgroupsize[I]) != 0
-            return fld1(ndrange[I], workgroupsize[I])
+            dynamic[] |= mod(worksize(ndrange[I]), workgroupsize[I]) != 0
+            return fld1(worksize(ndrange[I]), workgroupsize[I])
         end
-
         return blocks, workgroupsize, dynamic[] ? DynamicCheck() : NoDynamicCheck()
     end
 end

From d1ed6b7dbfc0f484a7872f367df2e85d511ff252 Mon Sep 17 00:00:00 2001
From: Simone Silvestri <33547697+simone-silvestri@users.noreply.github.com>
Date: Sun, 11 Jun 2023 16:58:31 -0400
Subject: [PATCH 2/9] tests

---
 src/nditeration.jl |  8 ++++++++
 test/test.jl       | 20 ++++++++++++++++++++
 2 files changed, 28 insertions(+)

diff --git a/src/nditeration.jl b/src/nditeration.jl
index aaa48595f..f7a0de9b0 100644
--- a/src/nditeration.jl
+++ b/src/nditeration.jl
@@ -64,6 +64,14 @@ struct NDRange{N, Offsets, StaticBlocks, StaticWorkitems, DynamicBlock, DynamicW
     blocks::DynamicBlock
     workitems::DynamicWorkitems
 
+    function NDRange{N, B, W}() where {N, B, W}
+        new{N, nothing, B, W, Nothing, Nothing}(nothing, nothing)
+    end
+
+    function NDRange{N, B, W}(blocks, workitems) where {N, B, W}
+        new{N, nothing, B, W, typeof(blocks), typeof(workitems)}(blocks, workitems)
+    end
+
     function NDRange{N, O, B, W}() where {N, O, B, W}
         new{N, O, B, W, Nothing, Nothing}(nothing, nothing)
     end
diff --git a/test/test.jl b/test/test.jl
index 88086342f..7a419fd1b 100644
--- a/test/test.jl
+++ b/test/test.jl
@@ -215,6 +215,26 @@ end
     synchronize(Backend())
 end
 
+@kernel function index_global_offset!(a)
+    i, j = @index(Global, NTuple)
+    n, m = size(a)
+    @inbounds a[i, j] = i + n * j
+end
+
+@conditional_testset "Offset iteration space $Backend" skip_test begin
+    a = zeros(7, 9)
+    loop! = index_global_offset!(Backend(), (2, 2), size(a) .- 4, (2, 2))
+    loop!(a)
+    synchronize(Backend())
+
+    b = [i + 7 * j for i in 1:7, j in 1:9]
+
+    @test a[3:5, 3:7] == b[3:5, 3:7]
+    @test a[1:2, :] == zeros(2, 9)
+    @test a[6:7, :] == zeros(2, 9)
+    @test a[:, 1:2] == zeros(7, 2)
+    @test a[:, 8:9] == zeros(7, 2)
+end
 
 @conditional_testset "return statement" skip_tests begin
     try

From 4557353e27b9a46220c347ab5b1a0cfde15c10dc Mon Sep 17 00:00:00 2001
From: Simone Silvestri <33547697+simone-silvestri@users.noreply.github.com>
Date: Sun, 11 Jun 2023 17:00:38 -0400
Subject: [PATCH 3/9] separate offsets prior to blocks()

---
 src/nditeration.jl | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/nditeration.jl b/src/nditeration.jl
index f7a0de9b0..4daa2a3d4 100644
--- a/src/nditeration.jl
+++ b/src/nditeration.jl
@@ -150,8 +150,8 @@ needs to perform dynamic bounds-checking.
         dynamic = Ref(false)
         blocks  = ntuple(Val(length(ndrange))) do I
             Base.@_inline_meta
-            dynamic[] |= mod(worksize(ndrange[I]), workgroupsize[I]) != 0
-            return fld1(worksize(ndrange[I]), workgroupsize[I])
+            dynamic[] |= mod(ndrange[I], workgroupsize[I]) != 0
+            return fld1(ndrange[I], workgroupsize[I])
         end
         return blocks, workgroupsize, dynamic[] ? DynamicCheck() : NoDynamicCheck()
     end

From 8accb00b07fc38727baad393dc94f0d86802c834 Mon Sep 17 00:00:00 2001
From: Simone Silvestri <33547697+simone-silvestri@users.noreply.github.com>
Date: Sun, 11 Jun 2023 17:01:10 -0400
Subject: [PATCH 4/9] fix tests

---
 test/test.jl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/test/test.jl b/test/test.jl
index 7a419fd1b..ba5d8936c 100644
--- a/test/test.jl
+++ b/test/test.jl
@@ -221,7 +221,7 @@ end
     @inbounds a[i, j] = i + n * j
 end
 
-@conditional_testset "Offset iteration space $Backend" skip_test begin
+@conditional_testset "Offset iteration space $Backend" skip_tests begin
     a = zeros(7, 9)
     loop! = index_global_offset!(Backend(), (2, 2), size(a) .- 4, (2, 2))
     loop!(a)

From d7c75929ba19535e97b53912f4d918d49d6d4d08 Mon Sep 17 00:00:00 2001
From: Simone Silvestri <33547697+simone-silvestri@users.noreply.github.com>
Date: Sun, 11 Jun 2023 17:11:48 -0400
Subject: [PATCH 5/9] handle nothing

---
 src/macros.jl | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/src/macros.jl b/src/macros.jl
index 5078c52de..5bde61a30 100644
--- a/src/macros.jl
+++ b/src/macros.jl
@@ -57,8 +57,8 @@ function __kernel(expr, generate_cpu=true)
             Core.@__doc__ $name(dev) = $name(dev, $DynamicSize(), $DynamicSize())
             $name(dev, size) = $name(dev, $StaticSize(size), $DynamicSize())
             $name(dev, size, range) = $name(dev, $StaticSize(size), $StaticSize(range))
-            $name(dev, size, range::NTuple{N}, offset::NTuple{N}) where N = 
-                    $name(dev, $StaticSize(size), $StaticSize($contiguousrange(range, offset)))
+            $name(dev, size, range, ::Nothing) = $name(dev, size, range)
+            $name(dev, size, range, offset) =  $name(dev, $StaticSize(size), $StaticSize($contiguousrange(range, offset)))
             function $name(dev::Dev, sz::S, range::NDRange) where {Dev, S<:$_Size, NDRange<:$_Size}
                 if $isgpu(dev)
                     return $construct(dev, sz, range, $gpu_name)

From c5d969cdbb40afe71212c04c609764af4450379b Mon Sep 17 00:00:00 2001
From: Simone Silvestri <33547697+simone-silvestri@users.noreply.github.com>
Date: Sun, 11 Jun 2023 18:27:12 -0400
Subject: [PATCH 6/9] fix test

---
 test/test.jl | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/test/test.jl b/test/test.jl
index ba5d8936c..ffbfb7ce4 100644
--- a/test/test.jl
+++ b/test/test.jl
@@ -223,8 +223,7 @@ end
 
 @conditional_testset "Offset iteration space $Backend" skip_tests begin
     a = zeros(7, 9)
-    loop! = index_global_offset!(Backend(), (2, 2), size(a) .- 4, (2, 2))
-    loop!(a)
+    index_global_offset!(Backend(), (2, 2), size(a) .- 4, (2, 2))(a)
     synchronize(Backend())
 
     b = [i + 7 * j for i in 1:7, j in 1:9]

From 4692df9cc2b503af890cda67e5403b460f5a9933 Mon Sep 17 00:00:00 2001
From: Simone Silvestri <33547697+simone-silvestri@users.noreply.github.com>
Date: Sun, 11 Jun 2023 18:27:28 -0400
Subject: [PATCH 7/9] remove manifest

---
 Manifest.toml | 244 --------------------------------------------------
 1 file changed, 244 deletions(-)
 delete mode 100644 Manifest.toml

diff --git a/Manifest.toml b/Manifest.toml
deleted file mode 100644
index 791f408f4..000000000
--- a/Manifest.toml
+++ /dev/null
@@ -1,244 +0,0 @@
-# This file is machine-generated - editing it directly is not advised
-
-julia_version = "1.8.0"
-manifest_format = "2.0"
-project_hash = "9033168e08cf56d36e87b64ae995845922bec33c"
-
-[[deps.Adapt]]
-deps = ["LinearAlgebra", "Requires"]
-git-tree-sha1 = "76289dc51920fdc6e0013c872ba9551d54961c24"
-uuid = "79e6a3ab-5dfb-504d-930d-738a2a938a0e"
-version = "3.6.2"
-
-[[deps.ArgTools]]
-uuid = "0dad84c5-d112-42e6-8d28-ef12dabb789f"
-version = "1.1.1"
-
-[[deps.Artifacts]]
-uuid = "56f22d72-fd6d-98f1-02f0-08ddc0907c33"
-
-[[deps.Atomix]]
-deps = ["UnsafeAtomics"]
-git-tree-sha1 = "c06a868224ecba914baa6942988e2f2aade419be"
-uuid = "a9b6321e-bd34-4604-b9c9-b65b8de01458"
-version = "0.1.0"
-
-[[deps.Base64]]
-uuid = "2a0f44e3-6c83-55bd-87e4-b1978d98bd5f"
-
-[[deps.CEnum]]
-git-tree-sha1 = "eb4cb44a499229b3b8426dcfb5dd85333951ff90"
-uuid = "fa961155-64e5-5f13-b03f-caf6b980ea82"
-version = "0.4.2"
-
-[[deps.CompilerSupportLibraries_jll]]
-deps = ["Artifacts", "Libdl"]
-uuid = "e66e0078-7015-5450-92f7-15fbd957f2ae"
-version = "0.5.2+0"
-
-[[deps.Dates]]
-deps = ["Printf"]
-uuid = "ade2ca70-3891-5945-98fb-dc099432e06a"
-
-[[deps.Downloads]]
-deps = ["ArgTools", "FileWatching", "LibCURL", "NetworkOptions"]
-uuid = "f43a241f-c20a-4ad4-852c-f6b1247861c6"
-version = "1.6.0"
-
-[[deps.FileWatching]]
-uuid = "7b1f6079-737a-58dc-b8bc-7a2ca5c1b5ee"
-
-[[deps.InteractiveUtils]]
-deps = ["Markdown"]
-uuid = "b77e0a4c-d291-57a0-90e8-8db25a27a240"
-
-[[deps.JLLWrappers]]
-deps = ["Preferences"]
-git-tree-sha1 = "abc9885a7ca2052a736a600f7fa66209f96506e1"
-uuid = "692b3bcd-3c85-4b1f-b108-f13ce0eb3210"
-version = "1.4.1"
-
-[[deps.LLVM]]
-deps = ["CEnum", "LLVMExtra_jll", "Libdl", "Printf", "Unicode"]
-git-tree-sha1 = "5007c1421563108110bbd57f63d8ad4565808818"
-uuid = "929cbde3-209d-540e-8aea-75f648917ca0"
-version = "5.2.0"
-
-[[deps.LLVMExtra_jll]]
-deps = ["Artifacts", "JLLWrappers", "LazyArtifacts", "Libdl", "TOML"]
-git-tree-sha1 = "1222116d7313cdefecf3d45a2bc1a89c4e7c9217"
-uuid = "dad2f222-ce93-54a1-a47d-0025e8a3acab"
-version = "0.0.22+0"
-
-[[deps.LazyArtifacts]]
-deps = ["Artifacts", "Pkg"]
-uuid = "4af54fe1-eca0-43a8-85a7-787d91b784e3"
-
-[[deps.LibCURL]]
-deps = ["LibCURL_jll", "MozillaCACerts_jll"]
-uuid = "b27032c2-a3e7-50c8-80cd-2d36dbcbfd21"
-version = "0.6.3"
-
-[[deps.LibCURL_jll]]
-deps = ["Artifacts", "LibSSH2_jll", "Libdl", "MbedTLS_jll", "Zlib_jll", "nghttp2_jll"]
-uuid = "deac9b47-8bc7-5906-a0fe-35ac56dc84c0"
-version = "7.84.0+0"
-
-[[deps.LibGit2]]
-deps = ["Base64", "NetworkOptions", "Printf", "SHA"]
-uuid = "76f85450-5226-5b5a-8eaa-529ad045b433"
-
-[[deps.LibSSH2_jll]]
-deps = ["Artifacts", "Libdl", "MbedTLS_jll"]
-uuid = "29816b5a-b9ab-546f-933c-edad1886dfa8"
-version = "1.10.2+0"
-
-[[deps.Libdl]]
-uuid = "8f399da3-3557-5675-b5ff-fb832c97cbdb"
-
-[[deps.LinearAlgebra]]
-deps = ["Libdl", "libblastrampoline_jll"]
-uuid = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
-
-[[deps.Logging]]
-uuid = "56ddb016-857b-54e1-b83d-db4d58db5568"
-
-[[deps.MacroTools]]
-deps = ["Markdown", "Random"]
-git-tree-sha1 = "42324d08725e200c23d4dfb549e0d5d89dede2d2"
-uuid = "1914dd2f-81c6-5fcd-8719-6d5c9610ff09"
-version = "0.5.10"
-
-[[deps.Markdown]]
-deps = ["Base64"]
-uuid = "d6f4376e-aef5-505a-96c1-9c027394607a"
-
-[[deps.MbedTLS_jll]]
-deps = ["Artifacts", "Libdl"]
-uuid = "c8ffd9c3-330d-5841-b78e-0817d7145fa1"
-version = "2.28.0+0"
-
-[[deps.MozillaCACerts_jll]]
-uuid = "14a3606d-f60d-562e-9121-12d972cd8159"
-version = "2022.2.1"
-
-[[deps.NetworkOptions]]
-uuid = "ca575930-c2e3-43a9-ace4-1e988b2c1908"
-version = "1.2.0"
-
-[[deps.OpenBLAS_jll]]
-deps = ["Artifacts", "CompilerSupportLibraries_jll", "Libdl"]
-uuid = "4536629a-c528-5b80-bd46-f80d51c5b363"
-version = "0.3.20+0"
-
-[[deps.Pkg]]
-deps = ["Artifacts", "Dates", "Downloads", "LibGit2", "Libdl", "Logging", "Markdown", "Printf", "REPL", "Random", "SHA", "Serialization", "TOML", "Tar", "UUIDs", "p7zip_jll"]
-uuid = "44cfe95a-1eb2-52ea-b672-e2afdf69b78f"
-version = "1.8.0"
-
-[[deps.PrecompileTools]]
-deps = ["Preferences"]
-git-tree-sha1 = "9673d39decc5feece56ef3940e5dafba15ba0f81"
-uuid = "aea7be01-6a6a-4083-8856-8a6e6704d82a"
-version = "1.1.2"
-
-[[deps.Preferences]]
-deps = ["TOML"]
-git-tree-sha1 = "7eb1686b4f04b82f96ed7a4ea5890a4f0c7a09f1"
-uuid = "21216c6a-2e73-6563-6e65-726566657250"
-version = "1.4.0"
-
-[[deps.Printf]]
-deps = ["Unicode"]
-uuid = "de0858da-6303-5e67-8744-51eddeeeb8d7"
-
-[[deps.REPL]]
-deps = ["InteractiveUtils", "Markdown", "Sockets", "Unicode"]
-uuid = "3fa0cd96-eef1-5676-8a61-b3b8758bbffb"
-
-[[deps.Random]]
-deps = ["SHA", "Serialization"]
-uuid = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
-
-[[deps.Requires]]
-deps = ["UUIDs"]
-git-tree-sha1 = "838a3a4188e2ded87a4f9f184b4b0d78a1e91cb7"
-uuid = "ae029012-a4dd-5104-9daa-d747884805df"
-version = "1.3.0"
-
-[[deps.SHA]]
-uuid = "ea8e919c-243c-51af-8825-aaa63cd721ce"
-version = "0.7.0"
-
-[[deps.Serialization]]
-uuid = "9e88b42a-f829-5b0c-bbe9-9e923198166b"
-
-[[deps.Sockets]]
-uuid = "6462fe0b-24de-5631-8697-dd941f90decc"
-
-[[deps.SparseArrays]]
-deps = ["LinearAlgebra", "Random"]
-uuid = "2f01184e-e22b-5df5-ae63-d93ebab69eaf"
-
-[[deps.StaticArrays]]
-deps = ["LinearAlgebra", "Random", "StaticArraysCore", "Statistics"]
-git-tree-sha1 = "832afbae2a45b4ae7e831f86965469a24d1d8a83"
-uuid = "90137ffa-7385-5640-81b9-e52037218182"
-version = "1.5.26"
-
-[[deps.StaticArraysCore]]
-git-tree-sha1 = "6b7ba252635a5eff6a0b0664a41ee140a1c9e72a"
-uuid = "1e83bf80-4336-4d27-bf5d-d5a4f845583c"
-version = "1.4.0"
-
-[[deps.Statistics]]
-deps = ["LinearAlgebra", "SparseArrays"]
-uuid = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
-
-[[deps.TOML]]
-deps = ["Dates"]
-uuid = "fa267f1f-6049-4f14-aa54-33bafae1ed76"
-version = "1.0.0"
-
-[[deps.Tar]]
-deps = ["ArgTools", "SHA"]
-uuid = "a4e569a6-e804-4fa4-b0f3-eef7a1d5b13e"
-version = "1.10.0"
-
-[[deps.UUIDs]]
-deps = ["Random", "SHA"]
-uuid = "cf7118a7-6976-5b1a-9a39-7adc72f591a4"
-
-[[deps.Unicode]]
-uuid = "4ec0a83e-493e-50e2-b9ac-8f72acf5a8f5"
-
-[[deps.UnsafeAtomics]]
-git-tree-sha1 = "6331ac3440856ea1988316b46045303bef658278"
-uuid = "013be700-e6cd-48c3-b4a1-df204f14c38f"
-version = "0.2.1"
-
-[[deps.UnsafeAtomicsLLVM]]
-deps = ["LLVM", "UnsafeAtomics"]
-git-tree-sha1 = "ea37e6066bf194ab78f4e747f5245261f17a7175"
-uuid = "d80eeb9a-aca5-4d75-85e5-170c8b632249"
-version = "0.1.2"
-
-[[deps.Zlib_jll]]
-deps = ["Libdl"]
-uuid = "83775a58-1f1d-513f-b197-d71354ab007a"
-version = "1.2.12+3"
-
-[[deps.libblastrampoline_jll]]
-deps = ["Artifacts", "Libdl", "OpenBLAS_jll"]
-uuid = "8e850b90-86db-534c-a0d3-1478176c7d93"
-version = "5.1.1+0"
-
-[[deps.nghttp2_jll]]
-deps = ["Artifacts", "Libdl"]
-uuid = "8e850ede-7688-5339-a07c-302acd2aaf8d"
-version = "1.48.0+0"
-
-[[deps.p7zip_jll]]
-deps = ["Artifacts", "Libdl"]
-uuid = "3f19e933-33d8-53b3-aaab-bd5110c3b7a0"
-version = "17.4.0+0"

From 13e6de9dddeac9a8435a494427feff6c02ba9a5d Mon Sep 17 00:00:00 2001
From: Simone Silvestri <33547697+simone-silvestri@users.noreply.github.com>
Date: Mon, 12 Jun 2023 09:46:39 -0400
Subject: [PATCH 8/9] remove `StepRange`

---
 src/nditeration.jl | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/src/nditeration.jl b/src/nditeration.jl
index 4daa2a3d4..d6d4730f3 100644
--- a/src/nditeration.jl
+++ b/src/nditeration.jl
@@ -35,13 +35,11 @@ end
 @inline worksize(i::Tuple) = worksize.(i)
 @inline worksize(i::Int) = i
 @inline worksize(i::UnitRange) = length(i)
-@inline worksize(i::StepRange) = length(i)
 
 @inline offsets(i) = offsets.(i)
 @inline offsets(::NTuple{N, Int}) where N = nothing
 @inline offsets(::Int) = nothing
 @inline offsets(i::UnitRange) = i.start - 1
-@inline offsets(i::StepRange) = i.start - 1, i.step
 
 """
     NDRange
@@ -148,7 +146,7 @@ needs to perform dynamic bounds-checking.
     end
     let workgroupsize = workgroupsize
         dynamic = Ref(false)
-        blocks  = ntuple(Val(length(ndrange))) do I
+        blocks = ntuple(Val(length(ndrange))) do I
             Base.@_inline_meta
             dynamic[] |= mod(ndrange[I], workgroupsize[I]) != 0
             return fld1(ndrange[I], workgroupsize[I])

From 08c0c02bc10523a30a22bdf942a5ab3627fff6de Mon Sep 17 00:00:00 2001
From: Simone Silvestri <33547697+simone-silvestri@users.noreply.github.com>
Date: Mon, 12 Jun 2023 11:32:02 -0400
Subject: [PATCH 9/9] fixed test

---
 test/test.jl | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/test/test.jl b/test/test.jl
index ffbfb7ce4..aaeb6710d 100644
--- a/test/test.jl
+++ b/test/test.jl
@@ -222,17 +222,20 @@ end
 end
 
 @conditional_testset "Offset iteration space $Backend" skip_tests begin
-    a = zeros(7, 9)
+    a = KernelAbstractions.zeros(Backend(), 7, 9)
     index_global_offset!(Backend(), (2, 2), size(a) .- 4, (2, 2))(a)
     synchronize(Backend())
 
-    b = [i + 7 * j for i in 1:7, j in 1:9]
+    b = KernelAbstractions.zeros(CPU(), 7, 9)
+    b .= a
 
-    @test a[3:5, 3:7] == b[3:5, 3:7]
-    @test a[1:2, :] == zeros(2, 9)
-    @test a[6:7, :] == zeros(2, 9)
-    @test a[:, 1:2] == zeros(7, 2)
-    @test a[:, 8:9] == zeros(7, 2)
+    c = [i + 7 * j for i in 1:7, j in 1:9]
+
+    @test b[3:5, 3:7] == c[3:5, 3:7]
+    @test b[1:2, :] == zeros(2, 9)
+    @test b[6:7, :] == zeros(2, 9)
+    @test b[:, 1:2] == zeros(7, 2)
+    @test b[:, 8:9] == zeros(7, 2)
 end
 
 @conditional_testset "return statement" skip_tests begin