Switch to 1.12 for benchmarks

[christiangnrd]

Just changing this so it shows up on the benchmark graphs and makes it easier to figure out why a the sudden jump/drop in some benchmarks

[github-actions]

CUDA.jl Benchmarks

Details

Benchmark suite	Current: 3f78272	Previous: e2eab84	Ratio
latency/precompile	45051704463.5 ns	55351775865 ns	0.81
latency/ttfp	14298177097 ns	7675959782 ns	1.86
latency/import	4533035006 ns	4025008043 ns	1.13
integration/volumerhs	9439094.5 ns	9625265.5 ns	0.98
integration/byval/slices=1	145633.5 ns	147037 ns	0.99
integration/byval/slices=3	422923 ns	426089 ns	0.99
integration/byval/reference	143856 ns	144983 ns	0.99
integration/byval/slices=2	284175 ns	286492 ns	0.99
integration/cudadevrt	102423 ns	103531 ns	0.99
kernel/indexing	13579 ns	14161 ns	0.96
kernel/indexing_checked	14005 ns	15039 ns	0.93
kernel/occupancy	692.6013513513514 ns	792.6237623762377 ns	0.87
kernel/launch	2197.6666666666665 ns	2304.777777777778 ns	0.95
kernel/rand	14320 ns	18605.5 ns	0.77
array/reverse/1d	18739 ns	19692 ns	0.95
array/reverse/2dL_inplace	66142 ns	66886 ns	0.99
array/reverse/1dL	68931 ns	69816 ns	0.99
array/reverse/2d	20818 ns	21785 ns	0.96
array/reverse/1d_inplace	8842.333333333334 ns	9783 ns	0.90
array/reverse/2d_inplace	12691 ns	13411 ns	0.95
array/reverse/2dL	72797.5 ns	73680 ns	0.99
array/reverse/1dL_inplace	66162 ns	66877 ns	0.99
array/copy	18055 ns	20494 ns	0.88
array/iteration/findall/int	145603.5 ns	159432 ns	0.91
array/iteration/findall/bool	130483 ns	141132 ns	0.92
array/iteration/findfirst/int	84143 ns	161404 ns	0.52
array/iteration/findfirst/bool	81456 ns	162399 ns	0.50
array/iteration/scalar	68124 ns	73008 ns	0.93
array/iteration/logical	195818.5 ns	220542 ns	0.89
array/iteration/findmin/1d	84052.5 ns	94388 ns	0.89
array/iteration/findmin/2d	116890 ns	121456 ns	0.96
array/reductions/reduce/Int64/1d	39246 ns	43817 ns	0.90
array/reductions/reduce/Int64/dims=1	51371 ns	44722 ns	1.15
array/reductions/reduce/Int64/dims=2	58829 ns	61524.5 ns	0.96
array/reductions/reduce/Int64/dims=1L	86990 ns	88932 ns	0.98
array/reductions/reduce/Int64/dims=2L	84622 ns	88232 ns	0.96
array/reductions/reduce/Float32/1d	34377 ns	37401.5 ns	0.92
array/reductions/reduce/Float32/dims=1	40438 ns	51945.5 ns	0.78
array/reductions/reduce/Float32/dims=2	56463 ns	59868 ns	0.94
array/reductions/reduce/Float32/dims=1L	51417 ns	52569 ns	0.98
array/reductions/reduce/Float32/dims=2L	70207 ns	72250 ns	0.97
array/reductions/mapreduce/Int64/1d	39306 ns	43757 ns	0.90
array/reductions/mapreduce/Int64/dims=1	43505 ns	51077 ns	0.85
array/reductions/mapreduce/Int64/dims=2	58972 ns	61716 ns	0.96
array/reductions/mapreduce/Int64/dims=1L	86971 ns	89026 ns	0.98
array/reductions/mapreduce/Int64/dims=2L	84890 ns	88266 ns	0.96
array/reductions/mapreduce/Float32/1d	33962 ns	37130 ns	0.91
array/reductions/mapreduce/Float32/dims=1	39705.5 ns	42011 ns	0.95
array/reductions/mapreduce/Float32/dims=2	56085 ns	60008 ns	0.93
array/reductions/mapreduce/Float32/dims=1L	51159 ns	52656 ns	0.97
array/reductions/mapreduce/Float32/dims=2L	69891 ns	72333.5 ns	0.97
array/broadcast	20412 ns	20154 ns	1.01
array/copyto!/gpu_to_gpu	10660.666666666666 ns	12828 ns	0.83
array/copyto!/cpu_to_gpu	213289 ns	216999 ns	0.98
array/copyto!/gpu_to_cpu	282857 ns	282452 ns	1.00
array/accumulate/Int64/1d	117824.5 ns	125219 ns	0.94
array/accumulate/Int64/dims=1	79135 ns	88138 ns	0.90
array/accumulate/Int64/dims=2	155937.5 ns	162165 ns	0.96
array/accumulate/Int64/dims=1L	1705343 ns	1714037 ns	0.99
array/accumulate/Int64/dims=2L	960402.5 ns	971078.5 ns	0.99
array/accumulate/Float32/1d	100373 ns	109882.5 ns	0.91
array/accumulate/Float32/dims=1	76042 ns	84424 ns	0.90
array/accumulate/Float32/dims=2	143997 ns	151744.5 ns	0.95
array/accumulate/Float32/dims=1L	1590640 ns	1622762 ns	0.98
array/accumulate/Float32/dims=2L	658982 ns	702590.5 ns	0.94
array/construct	1313.4 ns	1267.85 ns	1.04
array/random/randn/Float32	42758.5 ns	48207 ns	0.89
array/random/randn!/Float32	29806 ns	25000 ns	1.19
array/random/rand!/Int64	34591 ns	27295 ns	1.27
array/random/rand!/Float32	8171.333333333333 ns	8737.333333333334 ns	0.94
array/random/rand/Int64	30153 ns	30000 ns	1.01
array/random/rand/Float32	12165 ns	13155 ns	0.92
array/permutedims/4d	50924 ns	55023 ns	0.93
array/permutedims/2d	52358 ns	53878 ns	0.97
array/permutedims/3d	52707 ns	54959 ns	0.96
array/sorting/1d	2735396 ns	2758315 ns	0.99
array/sorting/by	3304406 ns	3344753.5 ns	0.99
array/sorting/2d	1067036 ns	1081270 ns	0.99
cuda/synchronization/stream/auto	1022.7 ns	1017.0833333333334 ns	1.01
cuda/synchronization/stream/nonblocking	7781.700000000001 ns	7295.6 ns	1.07
cuda/synchronization/stream/blocking	815.3333333333334 ns	798.6601941747573 ns	1.02
cuda/synchronization/context/auto	1192.6 ns	1158.8 ns	1.03
cuda/synchronization/context/nonblocking	6878.8 ns	7658.9 ns	0.90
cuda/synchronization/context/blocking	880.2037037037037 ns	902.5217391304348 ns	0.98

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