unsafe_wrap for symbols

[vchuravy]

Spiritual follow-up on #2624

Noticed while working on trixi-framework/Trixi.jl#2212

[github-actions]

Your PR requires formatting changes to meet the project's style guidelines.
Please consider running Runic (git runic master) to apply these changes.

Click here to view the suggested changes.

diff --git a/src/array.jl b/src/array.jl
index d66396bfc..d843209ad 100644
--- a/src/array.jl
+++ b/src/array.jl
@@ -50,19 +50,19 @@ end
 # As well as "mutable singleton" types like `Symbol` that use pointer-identity
 
 function valid_type(@nospecialize(T))
-  if Base.allocatedinline(T)
-    if hasfieldcount(T)
-      return all(valid_type, fieldtypes(T))
+    if Base.allocatedinline(T)
+        if hasfieldcount(T)
+            return all(valid_type, fieldtypes(T))
+        end
+        return true
+    elseif Base.ismutabletype(T)
+        return Base.datatype_fieldcount(T) == 0
     end
-    return true
-  elseif Base.ismutabletype(T)
-    return Base.datatype_fieldcount(T) == 0
-  end
-  return false
+    return false
 end
 
-@inline function check_eltype(name, T)                      
-  if !valid_type(T) 
+@inline function check_eltype(name, T)
+    return if !valid_type(T)
     explanation = explain_eltype(T)
     error("""
       $name only supports element types that are allocated inline.
@@ -247,7 +247,7 @@ end
 function Base.unsafe_wrap(::Type{CuArray{T,N,M}},
                           ptr::CuPtr{T}, dims::NTuple{N,Int};
                           own::Bool=false, ctx::CuContext=context()) where {T,N,M}
-  check_eltype("unsafe_wrap(CuArray, ...)", T)
+    check_eltype("unsafe_wrap(CuArray, ...)", T)
   sz = prod(dims) * aligned_sizeof(T)
 
   # create a memory object
diff --git a/test/base/array.jl b/test/base/array.jl
index 42f4668af..83335046b 100644
--- a/test/base/array.jl
+++ b/test/base/array.jl
@@ -176,14 +176,14 @@ end
 
     # symbols and tuples thereof
     let a = CuArray([:a])
-      b = unsafe_wrap(CuArray, pointer(a), 1)
-      @test typeof(b) <: CuArray{Symbol,1}
-      @test size(b) == (1,)
+        b = unsafe_wrap(CuArray, pointer(a), 1)
+        @test typeof(b) <: CuArray{Symbol, 1}
+        @test size(b) == (1,)
     end
-    let a = CuArray([(:a,:b)])
-      b = unsafe_wrap(CuArray, pointer(a), 1)
-      @test typeof(b) <: CuArray{Tuple{Symbol,Symbol},1}
-      @test size(b) == (1,)
+    let a = CuArray([(:a, :b)])
+        b = unsafe_wrap(CuArray, pointer(a), 1)
+        @test typeof(b) <: CuArray{Tuple{Symbol, Symbol}, 1}
+        @test size(b) == (1,)
     end
 end
 

[maleadt]

CI failures related.

[vchuravy]

Fun!

This fails:

CuArray([:a])

but this doesn't

CuArray([(:a,)])

The latter being the one I tested locally

[maleadt]

Having s/sizeof/argsize all over the wrappers feels very wrong. That was supposed to be an isolated thing for use during argument conversion.

[vchuravy]

Yeah, that's why I reverted the commit that added it to memory.jl

The crux is that we need a version of sizeof that returns sizeof(Ptr{Cvoid}) instead of sizeof(Symbol) == nothing.

I am doubly amused that :

julia> sizeof(Tuple{Symbol})
8

Of course works...

I am half tempted to shadow sizeof inside CUDA.jl with an arguably correct implementation.
Or we add something like morally_correct_sizeof to GPUToolkit.jl.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 89.61%. Comparing base (3d44e32) to head (dac3df0).
Report is 1 commits behind head on master.

Additional details and impacted files

@@             Coverage Diff             @@
##           master    #2753       +/-   ##
===========================================
+ Coverage   77.34%   89.61%   +12.27%     
===========================================
  Files         153      153               
  Lines       13108    13202       +94     
===========================================
+ Hits        10138    11831     +1693     
+ Misses       2970     1371     -1599     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

CUDA.jl Benchmarks

Details

Benchmark suite	Current: dc105d3	Previous: 7a83380	Ratio
latency/precompile	42852388609.5 ns	42926176632.5 ns	1.00
latency/ttfp	7124007951 ns	7106699468 ns	1.00
latency/import	3376439835 ns	3389948934 ns	1.00
integration/volumerhs	9622831.5 ns	9608052 ns	1.00
integration/byval/slices=1	147617 ns	147162 ns	1.00
integration/byval/slices=3	426742 ns	425761 ns	1.00
integration/byval/reference	145513 ns	145313 ns	1.00
integration/byval/slices=2	287399 ns	286425 ns	1.00
integration/cudadevrt	104082 ns	103604 ns	1.00
kernel/indexing	14845 ns	14311 ns	1.04
kernel/indexing_checked	15595 ns	15150.5 ns	1.03
kernel/occupancy	712.2765957446809 ns	707.8943661971831 ns	1.01
kernel/launch	2430.4444444444443 ns	2395.4444444444443 ns	1.01
kernel/rand	16234 ns	18678 ns	0.87
array/reverse/1d	20378 ns	19617 ns	1.04
array/reverse/2d	25377.5 ns	24183.5 ns	1.05
array/reverse/1d_inplace	11470 ns	10894 ns	1.05
array/reverse/2d_inplace	13470 ns	12860 ns	1.05
array/copy	21313 ns	20975 ns	1.02
array/iteration/findall/int	159797.5 ns	157315 ns	1.02
array/iteration/findall/bool	140863 ns	138316 ns	1.02
array/iteration/findfirst/int	163877 ns	154273.5 ns	1.06
array/iteration/findfirst/bool	165584 ns	154976.5 ns	1.07
array/iteration/scalar	73623 ns	73010 ns	1.01
array/iteration/logical	220743.5 ns	215446 ns	1.02
array/iteration/findmin/1d	48764 ns	41421 ns	1.18
array/iteration/findmin/2d	99894 ns	94396 ns	1.06
array/reductions/reduce/1d	47927 ns	43605.5 ns	1.10
array/reductions/reduce/2d	52530 ns	45135 ns	1.16
array/reductions/mapreduce/1d	46890 ns	40394 ns	1.16
array/reductions/mapreduce/2d	43615.5 ns	51842 ns	0.84
array/broadcast	21622 ns	21077 ns	1.03
array/copyto!/gpu_to_gpu	11405 ns	11034 ns	1.03
array/copyto!/cpu_to_gpu	218981 ns	216527 ns	1.01
array/copyto!/gpu_to_cpu	285643 ns	283217 ns	1.01
array/accumulate/1d	110396 ns	108949 ns	1.01
array/accumulate/2d	81232 ns	80328 ns	1.01
array/construct	1264.2 ns	1243.7 ns	1.02
array/random/randn/Float32	44970.5 ns	47595 ns	0.94
array/random/randn!/Float32	25502 ns	25281 ns	1.01
array/random/rand!/Int64	27522 ns	27337 ns	1.01
array/random/rand!/Float32	8964.333333333334 ns	8944.333333333334 ns	1.00
array/random/rand/Int64	30484 ns	33758 ns	0.90
array/random/rand/Float32	13412 ns	13235 ns	1.01
array/permutedims/4d	61881 ns	61033.5 ns	1.01
array/permutedims/2d	55855 ns	54927 ns	1.02
array/permutedims/3d	56883 ns	55899 ns	1.02
array/sorting/1d	2778520 ns	2777029.5 ns	1.00
array/sorting/by	3369694 ns	3367967 ns	1.00
array/sorting/2d	1087235 ns	1085746 ns	1.00
cuda/synchronization/stream/auto	1042.8 ns	1030 ns	1.01
cuda/synchronization/stream/nonblocking	7186 ns	8027.8 ns	0.90
cuda/synchronization/stream/blocking	810.5052631578948 ns	803.8265306122449 ns	1.01
cuda/synchronization/context/auto	1185.4 ns	1173.1 ns	1.01
cuda/synchronization/context/nonblocking	7433 ns	7706.2 ns	0.96
cuda/synchronization/context/blocking	906.625 ns	906.5853658536586 ns	1.00

This comment was automatically generated by workflow using github-action-benchmark.