Loosen compat bounds for CUDA_Runtime_Discovery for cudNN

[kshyatt]

No description provided.

[github-actions]

CUDA.jl Benchmarks

Benchmark suite	Current: 371ada7	Previous: 8ff92f9	Ratio
latency/precompile	42920971255.5 ns	43336576934.5 ns	0.99
latency/ttfp	7042028159 ns	6996576586 ns	1.01
latency/import	3562776071 ns	3575630035 ns	1.00
integration/volumerhs	9614991 ns	9614254.5 ns	1.00
integration/byval/slices=1	147276.5 ns	147031 ns	1.00
integration/byval/slices=3	426131 ns	426033 ns	1.00
integration/byval/reference	145072 ns	145044 ns	1.00
integration/byval/slices=2	286642 ns	286613 ns	1.00
integration/cudadevrt	103573 ns	103601 ns	1.00
kernel/indexing	14285 ns	14203 ns	1.01
kernel/indexing_checked	14987 ns	15046 ns	1.00
kernel/occupancy	670.6518987341772 ns	669.4810126582279 ns	1.00
kernel/launch	2240.4444444444443 ns	2096.1 ns	1.07
kernel/rand	15926 ns	16215 ns	0.98
array/reverse/1d	20119 ns	20044.5 ns	1.00
array/reverse/2d	23971 ns	24931 ns	0.96
array/reverse/1d_inplace	10748 ns	10596 ns	1.01
array/reverse/2d_inplace	13219 ns	12129 ns	1.09
array/copy	21018 ns	21283 ns	0.99
array/iteration/findall/int	159529 ns	158352 ns	1.01
array/iteration/findall/bool	140948.5 ns	140146 ns	1.01
array/iteration/findfirst/int	162156 ns	157269.5 ns	1.03
array/iteration/findfirst/bool	163629 ns	158043 ns	1.04
array/iteration/scalar	73000 ns	74613 ns	0.98
array/iteration/logical	217359 ns	216352 ns	1.00
array/iteration/findmin/1d	47446 ns	46817 ns	1.01
array/iteration/findmin/2d	97072.5 ns	96972.5 ns	1.00
array/reductions/reduce/Int64/1d	43736.5 ns	44160 ns	0.99
array/reductions/reduce/Int64/dims=1	46854.5 ns	51107 ns	0.92
array/reductions/reduce/Int64/dims=2	63051.5 ns	61712 ns	1.02
array/reductions/reduce/Int64/dims=1L	89466 ns	88969.5 ns	1.01
array/reductions/reduce/Int64/dims=2L	88245 ns	88740 ns	0.99
array/reductions/reduce/Float32/1d	35398 ns	35533 ns	1.00
array/reductions/reduce/Float32/dims=1	42076.5 ns	48101 ns	0.87
array/reductions/reduce/Float32/dims=2	59882 ns	60093 ns	1.00
array/reductions/reduce/Float32/dims=1L	52411.5 ns	52663 ns	1.00
array/reductions/reduce/Float32/dims=2L	70265 ns	70814 ns	0.99
array/reductions/mapreduce/Int64/1d	43767 ns	44462 ns	0.98
array/reductions/mapreduce/Int64/dims=1	48401.5 ns	54310.5 ns	0.89
array/reductions/mapreduce/Int64/dims=2	62309.5 ns	62878.5 ns	0.99
array/reductions/mapreduce/Int64/dims=1L	89275 ns	89203 ns	1.00
array/reductions/mapreduce/Int64/dims=2L	87931.5 ns	87190 ns	1.01
array/reductions/mapreduce/Float32/1d	35358 ns	35099 ns	1.01
array/reductions/mapreduce/Float32/dims=1	52256 ns	41869.5 ns	1.25
array/reductions/mapreduce/Float32/dims=2	59986 ns	60074 ns	1.00
array/reductions/mapreduce/Float32/dims=1L	52752 ns	52723 ns	1.00
array/reductions/mapreduce/Float32/dims=2L	70581 ns	70625.5 ns	1.00
array/broadcast	20269 ns	20313 ns	1.00
array/copyto!/gpu_to_gpu	11318 ns	13052 ns	0.87
array/copyto!/cpu_to_gpu	218470 ns	216249 ns	1.01
array/copyto!/gpu_to_cpu	285905 ns	283227 ns	1.01
array/accumulate/Int64/1d	125498 ns	125402 ns	1.00
array/accumulate/Int64/dims=1	84337 ns	84149 ns	1.00
array/accumulate/Int64/dims=2	158645 ns	158389 ns	1.00
array/accumulate/Int64/dims=1L	1711480.5 ns	1712930 ns	1.00
array/accumulate/Int64/dims=2L	966860 ns	968016 ns	1.00
array/accumulate/Float32/1d	109988 ns	109280.5 ns	1.01
array/accumulate/Float32/dims=1	81017 ns	80759 ns	1.00
array/accumulate/Float32/dims=2	148277 ns	148278.5 ns	1.00
array/accumulate/Float32/dims=1L	1619031 ns	1619105.5 ns	1.00
array/accumulate/Float32/dims=2L	699136.5 ns	698810 ns	1.00
array/construct	1278.7 ns	1273.7 ns	1.00
array/random/randn/Float32	44951 ns	48132.5 ns	0.93
array/random/randn!/Float32	25311 ns	25307 ns	1.00
array/random/rand!/Int64	27435 ns	27592 ns	0.99
array/random/rand!/Float32	8772.666666666666 ns	8757 ns	1.00
array/random/rand/Int64	30101 ns	38376 ns	0.78
array/random/rand/Float32	13094 ns	13207 ns	0.99
array/permutedims/4d	60695.5 ns	60719.5 ns	1.00
array/permutedims/2d	54490 ns	54666 ns	1.00
array/permutedims/3d	55599.5 ns	55885 ns	0.99
array/sorting/1d	2758204 ns	2758224 ns	1.00
array/sorting/by	3344878 ns	3344978 ns	1.00
array/sorting/2d	1081602 ns	1081293.5 ns	1.00
cuda/synchronization/stream/auto	1026.6 ns	1048.4 ns	0.98
cuda/synchronization/stream/nonblocking	6826.700000000001 ns	7536.8 ns	0.91
cuda/synchronization/stream/blocking	808.8421052631579 ns	808.7849462365591 ns	1.00
cuda/synchronization/context/auto	1170.5 ns	1201.8 ns	0.97
cuda/synchronization/context/nonblocking	6898.9 ns	8845.7 ns	0.78
cuda/synchronization/context/blocking	908.9245283018868 ns	906.8139534883721 ns	1.00

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

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 89.64%. Comparing base (8ff92f9) to head (371ada7).
⚠️ Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2855      +/-   ##
==========================================
+ Coverage   89.56%   89.64%   +0.08%     
==========================================
  Files         150      150              
  Lines       13250    13250              
==========================================
+ Hits        11867    11878      +11     
+ Misses       1383     1372      -11     

☔ 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.

[RomeoV]

I think this needs a rebase? Or manual merge, seems like auto-merge won't trigger because the branch is out of date...