Add tests for fast_div and fast rcp intrinsics

[vchuravy]

Follow-up to #3077

[github-actions]

CUDA.jl Benchmarks

Details

Benchmark suite	Current: ab56cd0	Previous: 5065018	Ratio
array/accumulate/Float32/1d	101208 ns	101779 ns	0.99
array/accumulate/Float32/dims=1	76528 ns	76943 ns	0.99
array/accumulate/Float32/dims=1L	1585194 ns	1585563 ns	1.00
array/accumulate/Float32/dims=2	143253 ns	144109 ns	0.99
array/accumulate/Float32/dims=2L	657669 ns	658119 ns	1.00
array/accumulate/Int64/1d	118532 ns	118854 ns	1.00
array/accumulate/Int64/dims=1	80227 ns	80389 ns	1.00
array/accumulate/Int64/dims=1L	1695245 ns	1694708 ns	1.00
array/accumulate/Int64/dims=2	156577 ns	156166 ns	1.00
array/accumulate/Int64/dims=2L	962022.5 ns	961594 ns	1.00
array/broadcast	20810 ns	20512 ns	1.01
array/construct	1313.85 ns	1320.1 ns	1.00
array/copy	18968 ns	18914 ns	1.00
array/copyto!/cpu_to_gpu	217082 ns	214662 ns	1.01
array/copyto!/gpu_to_cpu	283361 ns	285022 ns	0.99
array/copyto!/gpu_to_gpu	11529 ns	11357 ns	1.02
array/iteration/findall/bool	131938 ns	132106 ns	1.00
array/iteration/findall/int	148879 ns	148958 ns	1.00
array/iteration/findfirst/bool	81910 ns	82321 ns	1.00
array/iteration/findfirst/int	84045.5 ns	84247.5 ns	1.00
array/iteration/findmin/1d	84855 ns	85453 ns	0.99
array/iteration/findmin/2d	117375 ns	117211 ns	1.00
array/iteration/logical	200038 ns	200370 ns	1.00
array/iteration/scalar	68049 ns	69501 ns	0.98
array/permutedims/2d	52836 ns	52561 ns	1.01
array/permutedims/3d	52824.5 ns	53080 ns	1.00
array/permutedims/4d	51488 ns	51880.5 ns	0.99
array/random/rand/Float32	13093 ns	12980 ns	1.01
array/random/rand/Int64	31000 ns	29766 ns	1.04
array/random/rand!/Float32	8495.666666666666 ns	8371 ns	1.01
array/random/rand!/Int64	34060 ns	33996 ns	1.00
array/random/randn/Float32	38197 ns	38221 ns	1.00
array/random/randn!/Float32	31407 ns	31486 ns	1.00
array/reductions/mapreduce/Float32/1d	35181 ns	35499.5 ns	0.99
array/reductions/mapreduce/Float32/dims=1	43967.5 ns	39750 ns	1.11
array/reductions/mapreduce/Float32/dims=1L	51836 ns	51923 ns	1.00
array/reductions/mapreduce/Float32/dims=2	56969.5 ns	56902 ns	1.00
array/reductions/mapreduce/Float32/dims=2L	69383.5 ns	70119.5 ns	0.99
array/reductions/mapreduce/Int64/1d	43171 ns	43301 ns	1.00
array/reductions/mapreduce/Int64/dims=1	42905.5 ns	53296 ns	0.81
array/reductions/mapreduce/Int64/dims=1L	87788 ns	87758 ns	1.00
array/reductions/mapreduce/Int64/dims=2	59737 ns	60300 ns	0.99
array/reductions/mapreduce/Int64/dims=2L	85291 ns	85288 ns	1.00
array/reductions/reduce/Float32/1d	35370 ns	35511 ns	1.00
array/reductions/reduce/Float32/dims=1	40119 ns	49667 ns	0.81
array/reductions/reduce/Float32/dims=1L	51989 ns	52100 ns	1.00
array/reductions/reduce/Float32/dims=2	57339 ns	56986 ns	1.01
array/reductions/reduce/Float32/dims=2L	70274 ns	70384 ns	1.00
array/reductions/reduce/Int64/1d	43053 ns	43476 ns	0.99
array/reductions/reduce/Int64/dims=1	50490 ns	42765.5 ns	1.18
array/reductions/reduce/Int64/dims=1L	87932 ns	87783 ns	1.00
array/reductions/reduce/Int64/dims=2	60003 ns	59896 ns	1.00
array/reductions/reduce/Int64/dims=2L	85024.5 ns	85038 ns	1.00
array/reverse/1d	18101 ns	18408 ns	0.98
array/reverse/1dL	68759 ns	69079 ns	1.00
array/reverse/1dL_inplace	65860 ns	65864 ns	1.00
array/reverse/1d_inplace	10185.333333333334 ns	10281.333333333334 ns	0.99
array/reverse/2d	20573 ns	20739 ns	0.99
array/reverse/2dL	72719 ns	72716 ns	1.00
array/reverse/2dL_inplace	65820 ns	65932 ns	1.00
array/reverse/2d_inplace	10668 ns	10105.5 ns	1.06
array/sorting/1d	2735643 ns	2734769 ns	1.00
array/sorting/2d	1069201 ns	1076074 ns	0.99
array/sorting/by	3304362 ns	3327052 ns	0.99
cuda/synchronization/context/auto	1178.9 ns	1184.7 ns	1.00
cuda/synchronization/context/blocking	928.8235294117648 ns	940.969696969697 ns	0.99
cuda/synchronization/context/nonblocking	7010.2 ns	8469.4 ns	0.83
cuda/synchronization/stream/auto	986.5555555555555 ns	1021.1 ns	0.97
cuda/synchronization/stream/blocking	780.0833333333334 ns	832.7631578947369 ns	0.94
cuda/synchronization/stream/nonblocking	7285.5 ns	7582.5 ns	0.96
integration/byval/reference	143932 ns	144056 ns	1.00
integration/byval/slices=1	145654 ns	145820 ns	1.00
integration/byval/slices=2	284592 ns	284683 ns	1.00
integration/byval/slices=3	423021 ns	423445 ns	1.00
integration/cudadevrt	102549 ns	102525.5 ns	1.00
integration/volumerhs	23440557.5 ns	23421325.5 ns	1.00
kernel/indexing	13083 ns	13267 ns	0.99
kernel/indexing_checked	13981 ns	14039 ns	1.00
kernel/launch	2119.2 ns	2141.4444444444443 ns	0.99
kernel/occupancy	665.6100628930818 ns	695.4267515923567 ns	0.96
kernel/rand	14345 ns	15508 ns	0.93
latency/import	3801168926.5 ns	3810464780.5 ns	1.00
latency/precompile	4587956035 ns	4591174243 ns	1.00
latency/ttfp	4382456999 ns	4390561943.5 ns	1.00

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

[maleadt]

CI failure related.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 90.42%. Comparing base (01a0795) to head (ab56cd0).
⚠️ Report is 3 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #3096      +/-   ##
==========================================
- Coverage   90.43%   90.42%   -0.01%     
==========================================
  Files         141      141              
  Lines       12025    12025              
==========================================
- Hits        10875    10874       -1     
- Misses       1150     1151       +1     

☔ 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.