KernelInterface

[christiangnrd]

Do not merge until KernelInterface has been reviewed and interface fully decided

[github-actions]

AMDGPU.jl Benchmarks

Details

Benchmark suite	Current: d3d9a02	Previous: 4ff61e3	Ratio
amdgpu/synchronization/context/device	555 ns	567.5 ns	0.98
amdgpu/synchronization/stream/blocking	225 ns	232.5 ns	0.97
amdgpu/synchronization/stream/nonblocking	315 ns	312.5 ns	1.01
applications/bitonic_sort	1646825 ns	1650734.75 ns	1.00
applications/convolution	126214.25 ns	106826.75 ns	1.18
applications/floyd_warshall	11747728.25 ns	11787759.25 ns	1.00
applications/histogram	785728.25 ns	789913 ns	0.99
applications/prefix_sum	254411 ns	251799.25 ns	1.01
array/accumulate/Float32/1d	83368.75 ns	72711.25 ns	1.15
array/accumulate/Float32/dims=1	276336.25 ns	284589.75 ns	0.97
array/accumulate/Float32/dims=1L	102209 ns	81133.75 ns	1.26
array/accumulate/Float32/dims=2	97881.25 ns	70688.75 ns	1.38
array/accumulate/Float32/dims=2L	2656819.25 ns	2754571.75 ns	0.96
array/accumulate/Int64/1d	95263.75 ns	77051.25 ns	1.24
array/accumulate/Int64/dims=1	244978.5 ns	242136.5 ns	1.01
array/accumulate/Int64/dims=1L	109904 ns	83854 ns	1.31
array/accumulate/Int64/dims=2	100923.75 ns	70858.75 ns	1.42
array/accumulate/Int64/dims=2L	2839874.25 ns	2892754.5 ns	0.98
array/broadcast	74686 ns	72816.5 ns	1.03
array/construct	2285 ns	2200 ns	1.04
array/copy	37668 ns	36975.5 ns	1.02
array/copyto!/cpu_to_gpu	111526.5 ns	110439.5 ns	1.01
array/copyto!/gpu_to_cpu	111591.5 ns	110576.75 ns	1.01
array/copyto!/gpu_to_gpu	59555.75 ns	58938.5 ns	1.01
array/iteration/findall/bool	151147 ns	133184.25 ns	1.13
array/iteration/findall/int	154939.75 ns	146924.25 ns	1.05
array/iteration/findfirst/bool	147924.75 ns	183770 ns	0.80
array/iteration/findfirst/int	164992.5 ns	145197 ns	1.14
array/iteration/findmin/1d	136229.5 ns	109714 ns	1.24
array/iteration/findmin/2d	123044.25 ns	106499 ns	1.16
array/iteration/logical	246893.5 ns	240093.25 ns	1.03
array/iteration/scalar	293911.5 ns	292796.75 ns	1.00
array/permutedims/2d	70928.25 ns	70961.25 ns	1.00
array/permutedims/3d	72578.5 ns	70623.75 ns	1.03
array/permutedims/4d	78673.5 ns	73141.25 ns	1.08
array/random/rand/Float32	45515.75 ns	44920.75 ns	1.01
array/random/rand/Int64	54813.25 ns	53576 ns	1.02
array/random/rand!/Float32	64958.5 ns	64641 ns	1.00
array/random/rand!/Int64	73371 ns	71331.25 ns	1.03
array/random/randn/Float32	67978.5 ns	78961.25 ns	0.86
array/random/randn!/Float32	81006 ns	72186.25 ns	1.12
array/reductions/mapreduce/Float32/1d	114239 ns	94511.5 ns	1.21
array/reductions/mapreduce/Float32/dims=1	93968.75 ns	75061.25 ns	1.25
array/reductions/mapreduce/Float32/dims=1L	828463.75 ns	832329.5 ns	1.00
array/reductions/mapreduce/Float32/dims=2	99216.25 ns	83001.5 ns	1.20
array/reductions/mapreduce/Float32/dims=2L	144102.25 ns	143060 ns	1.01
array/reductions/mapreduce/Int64/1d	113944 ns	93981.75 ns	1.21
array/reductions/mapreduce/Int64/dims=1	94489 ns	82693.75 ns	1.14
array/reductions/mapreduce/Int64/dims=1L	834026.5 ns	833094.75 ns	1.00
array/reductions/mapreduce/Int64/dims=2	97691.25 ns	83931.5 ns	1.16
array/reductions/mapreduce/Int64/dims=2L	144877 ns	143907.5 ns	1.01
array/reductions/reduce/Float32/1d	114099 ns	94144.25 ns	1.21
array/reductions/reduce/Float32/dims=1	94106.25 ns	75216.5 ns	1.25
array/reductions/reduce/Float32/dims=1L	833751.5 ns	832537.25 ns	1.00
array/reductions/reduce/Float32/dims=2	98699 ns	83219 ns	1.19
array/reductions/reduce/Float32/dims=2L	144209.5 ns	143200 ns	1.01
array/reductions/reduce/Int64/1d	113709 ns	91536.25 ns	1.24
array/reductions/reduce/Int64/dims=1	94406.25 ns	83593.5 ns	1.13
array/reductions/reduce/Int64/dims=1L	832139.25 ns	829128.75 ns	1.00
array/reductions/reduce/Int64/dims=2	97771.5 ns	83766 ns	1.17
array/reductions/reduce/Int64/dims=2L	144629.5 ns	143587 ns	1.01
array/reverse/1d	45635.5 ns	45153 ns	1.01
array/reverse/1dL	61623.25 ns	75516 ns	0.82
array/reverse/1dL_inplace	80468.5 ns	79963.75 ns	1.01
array/reverse/1d_inplace	61478.25 ns	60771 ns	1.01
array/reverse/2d	50730.75 ns	43250.75 ns	1.17
array/reverse/2dL	101061.25 ns	83061 ns	1.22
array/reverse/2dL_inplace	92268.75 ns	89986.25 ns	1.03
array/reverse/2d_inplace	60468.5 ns	55665.75 ns	1.09
array/sorting/1d	334629.75 ns	333063 ns	1.00
gemm/tiled	1908441.25 ns	1899696.5 ns	1.00
gemm/tiled_unbounded	1901676 ns	1922439.5 ns	0.99
integration/byval/reference	39101 ns	39001 ns	1.00
integration/byval/slices=1	40760 ns	39781 ns	1.02
integration/byval/slices=2	154242 ns	147233 ns	1.05
integration/byval/slices=3	245253 ns	245404 ns	1.00
integration/volumerhs	4910217 ns	4962111 ns	0.99
kernel/indexing	57723.5 ns	57025.75 ns	1.01
kernel/indexing_checked	58838.5 ns	56630.75 ns	1.04
kernel/launch	1462.5 ns	1365.25 ns	1.07
kernel/rand	113086.75 ns	88876 ns	1.27
latency/import	1689017058 ns	1752022784 ns	0.96
latency/precompile	45476574397 ns	39860711432 ns	1.14
latency/ttfp	2319363732 ns	2327930658 ns	1.00
stencil/diffusion3d	1626248.25 ns	1624179.5 ns	1.00
stencil/diffusion3d_checked	1657958.75 ns	1655412.5 ns	1.00

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