Exclude Int128 from testing on linalg/core

[luraess]

Possibly workaround #1066 according to #1066 (comment)

[github-actions]

AMDGPU.jl Benchmarks

Details

Benchmark suite	Current: 651d624	Previous: 4ff61e3	Ratio
amdgpu/synchronization/context/device	557.5 ns	567.5 ns	0.98
amdgpu/synchronization/stream/blocking	230 ns	232.5 ns	0.99
amdgpu/synchronization/stream/nonblocking	317.5 ns	312.5 ns	1.02
applications/bitonic_sort	1652671.75 ns	1650734.75 ns	1.00
applications/convolution	107349.25 ns	106826.75 ns	1.00
applications/floyd_warshall	11808251.75 ns	11787759.25 ns	1.00
applications/histogram	770688.75 ns	789913 ns	0.98
applications/prefix_sum	251121.25 ns	251799.25 ns	1.00
array/accumulate/Float32/1d	75416 ns	72711.25 ns	1.04
array/accumulate/Float32/dims=1	275724 ns	284589.75 ns	0.97
array/accumulate/Float32/dims=1L	80803.5 ns	81133.75 ns	1.00
array/accumulate/Float32/dims=2	79246.25 ns	70688.75 ns	1.12
array/accumulate/Float32/dims=2L	2833622 ns	2754571.75 ns	1.03
array/accumulate/Int64/1d	77836.25 ns	77051.25 ns	1.01
array/accumulate/Int64/dims=1	242773.75 ns	242136.5 ns	1.00
array/accumulate/Int64/dims=1L	83648.75 ns	83854 ns	1.00
array/accumulate/Int64/dims=2	85453.75 ns	70858.75 ns	1.21
array/accumulate/Int64/dims=2L	2893928 ns	2892754.5 ns	1.00
array/broadcast	72526 ns	72816.5 ns	1.00
array/construct	2175 ns	2200 ns	0.99
array/copy	37223 ns	36975.5 ns	1.01
array/copyto!/cpu_to_gpu	108201.5 ns	110439.5 ns	0.98
array/copyto!/gpu_to_cpu	112266.5 ns	110576.75 ns	1.02
array/copyto!/gpu_to_gpu	50865.75 ns	58938.5 ns	0.86
array/iteration/findall/bool	134257 ns	133184.25 ns	1.01
array/iteration/findall/int	142842 ns	146924.25 ns	0.97
array/iteration/findfirst/bool	184637.75 ns	183770 ns	1.00
array/iteration/findfirst/int	145224.75 ns	145197 ns	1.00
array/iteration/findmin/1d	110374.25 ns	109714 ns	1.01
array/iteration/findmin/2d	107514.25 ns	106499 ns	1.01
array/iteration/logical	241818.75 ns	240093.25 ns	1.01
array/iteration/scalar	290442 ns	292796.75 ns	0.99
array/permutedims/2d	59036 ns	70961.25 ns	0.83
array/permutedims/3d	70543.5 ns	70623.75 ns	1.00
array/permutedims/4d	73363.75 ns	73141.25 ns	1.00
array/random/rand/Float32	44675.75 ns	44920.75 ns	0.99
array/random/rand/Int64	53660.75 ns	53576 ns	1.00
array/random/rand!/Float32	64278.5 ns	64641 ns	0.99
array/random/rand!/Int64	69798.5 ns	71331.25 ns	0.98
array/random/randn/Float32	79013.75 ns	78961.25 ns	1.00
array/random/randn!/Float32	80593.75 ns	72186.25 ns	1.12
array/reductions/mapreduce/Float32/1d	93211.5 ns	94511.5 ns	0.99
array/reductions/mapreduce/Float32/dims=1	83478.75 ns	75061.25 ns	1.11
array/reductions/mapreduce/Float32/dims=1L	833745 ns	832329.5 ns	1.00
array/reductions/mapreduce/Float32/dims=2	83126 ns	83001.5 ns	1.00
array/reductions/mapreduce/Float32/dims=2L	143984.75 ns	143060 ns	1.01
array/reductions/mapreduce/Int64/1d	94008.75 ns	93981.75 ns	1.00
array/reductions/mapreduce/Int64/dims=1	82366.25 ns	82693.75 ns	1.00
array/reductions/mapreduce/Int64/dims=1L	831837.5 ns	833094.75 ns	1.00
array/reductions/mapreduce/Int64/dims=2	77356 ns	83931.5 ns	0.92
array/reductions/mapreduce/Int64/dims=2L	144074.75 ns	143907.5 ns	1.00
array/reductions/reduce/Float32/1d	94079 ns	94144.25 ns	1.00
array/reductions/reduce/Float32/dims=1	83418.75 ns	75216.5 ns	1.11
array/reductions/reduce/Float32/dims=1L	832732.5 ns	832537.25 ns	1.00
array/reductions/reduce/Float32/dims=2	84366.25 ns	83219 ns	1.01
array/reductions/reduce/Float32/dims=2L	143614.5 ns	143200 ns	1.00
array/reductions/reduce/Int64/1d	91921.25 ns	91536.25 ns	1.00
array/reductions/reduce/Int64/dims=1	82046.25 ns	83593.5 ns	0.98
array/reductions/reduce/Int64/dims=1L	831680 ns	829128.75 ns	1.00
array/reductions/reduce/Int64/dims=2	77326.25 ns	83766 ns	0.92
array/reductions/reduce/Int64/dims=2L	143104.75 ns	143587 ns	1.00
array/reverse/1d	45100.75 ns	45153 ns	1.00
array/reverse/1dL	72941.25 ns	75516 ns	0.97
array/reverse/1dL_inplace	63831 ns	79963.75 ns	0.80
array/reverse/1d_inplace	58435.75 ns	60771 ns	0.96
array/reverse/2d	49878.25 ns	43250.75 ns	1.15
array/reverse/2dL	83038.75 ns	83061 ns	1.00
array/reverse/2dL_inplace	91466.25 ns	89986.25 ns	1.02
array/reverse/2d_inplace	63198.5 ns	55665.75 ns	1.14
array/sorting/1d	335124.75 ns	333063 ns	1.01
gemm/tiled	1908395.75 ns	1899696.5 ns	1.00
gemm/tiled_unbounded	1928013.5 ns	1922439.5 ns	1.00
integration/byval/reference	38761 ns	39001 ns	0.99
integration/byval/slices=1	40381 ns	39781 ns	1.02
integration/byval/slices=2	127752 ns	147233 ns	0.87
integration/byval/slices=3	236534 ns	245404 ns	0.96
integration/volumerhs	4893182 ns	4962111 ns	0.99
kernel/indexing	56991 ns	57025.75 ns	1.00
kernel/indexing_checked	58271 ns	56630.75 ns	1.03
kernel/launch	1380 ns	1365.25 ns	1.01
kernel/rand	98739 ns	88876 ns	1.11
latency/import	1747637837 ns	1752022784 ns	1.00
latency/precompile	39817987654 ns	39860711432 ns	1.00
latency/ttfp	2331545188 ns	2327930658 ns	1.00
stencil/diffusion3d	1623525 ns	1624179.5 ns	1.00
stencil/diffusion3d_checked	1651010.25 ns	1655412.5 ns	1.00

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

[simeonschaub]

Why this check of the LLVM version? It's broken everywhere atm, so I don't see a reason to restrict this to only LLVM >= 19

[luraess]

I thought the Yggdrasil PR fixed it for < v"19". But can remove the check