Try enabling fwd broadcating Enzyme test

[kshyatt]

No description provided.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 0.00%. Comparing base (65f6ddd) to head (ef28de6).

Additional details and impacted files

@@            Coverage Diff             @@
##             main   #3214       +/-   ##
==========================================
- Coverage   92.64%   0.00%   -92.65%     
==========================================
  Files         194     145       -49     
  Lines       16988   11983     -5005     
==========================================
- Hits        15738       0    -15738     
- Misses       1250   11983    +10733     

☔ View full report in Codecov by Harness.
📢 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.

[github-actions]

CUDA.jl Benchmarks

Details

Benchmark suite	Current: 1006860	Previous: 069cdff	Ratio
array/accumulate/Float32/1d	97155 ns	98074 ns	0.99
array/accumulate/Float32/dims=1	71139 ns	74919 ns	0.95
array/accumulate/Float32/dims=1L	1598600 ns	1600002 ns	1.00
array/accumulate/Float32/dims=2	136323 ns	140367 ns	0.97
array/accumulate/Float32/dims=2L	659812 ns	659745 ns	1.00
array/accumulate/Int64/1d	117189 ns	117818 ns	0.99
array/accumulate/Int64/dims=1	75434 ns	79251 ns	0.95
array/accumulate/Int64/dims=1L	1715039 ns	1717606 ns	1.00
array/accumulate/Int64/dims=2	148320 ns	153507 ns	0.97
array/accumulate/Int64/dims=2L	986302 ns	986800 ns	1.00
array/broadcast	17442 ns	18060 ns	0.97
array/broadcast launch	8142.333333333333 ns
array/construct	900.8840579710145 ns	869.1052631578947 ns	1.04
array/copy	16120 ns	15967 ns	1.01
array/copyto!/cpu_to_gpu	208371 ns	208599 ns	1.00
array/copyto!/gpu_to_cpu	241018 ns	241280 ns	1.00
array/copyto!/gpu_to_gpu	8782.333333333334 ns	9209.333333333334 ns	0.95
array/iteration/findall/bool	130642 ns	132021 ns	0.99
array/iteration/findall/int	144511 ns	145286 ns	0.99
array/iteration/findfirst/bool	67218 ns	67274 ns	1.00
array/iteration/findfirst/int	68875 ns	68529 ns	1.01
array/iteration/findmin/1d	61659 ns	64309 ns	0.96
array/iteration/findmin/2d	98627 ns	99940 ns	0.99
array/iteration/logical	181793 ns	185815 ns	0.98
array/iteration/scalar	61876 ns	63136 ns	0.98
array/permutedims/2d	47540 ns	48431 ns	0.98
array/permutedims/3d	48910 ns	50026 ns	0.98
array/permutedims/4d	49196 ns	49849 ns	0.99
array/random/rand/Float32	11660 ns	11669 ns	1.00
array/random/rand/Int64	22239 ns	22883 ns	0.97
array/random/rand!/Float32	7653.75 ns	7731.75 ns	0.99
array/random/rand!/Int64	19867 ns	20210 ns	0.98
array/random/randn/Float32	32444 ns	32574 ns	1.00
array/random/randn!/Float32	23053 ns	23189 ns	0.99
array/reductions/mapreduce/Float32/1d	31100 ns	31371 ns	0.99
array/reductions/mapreduce/Float32/dims=1	36772 ns	37491 ns	0.98
array/reductions/mapreduce/Float32/dims=1L	49816 ns	49977 ns	1.00
array/reductions/mapreduce/Float32/dims=2	54135 ns	54508 ns	0.99
array/reductions/mapreduce/Float32/dims=2L	65938 ns	66329 ns	0.99
array/reductions/mapreduce/Int64/1d	37892 ns	37643 ns	1.01
array/reductions/mapreduce/Int64/dims=1	40039 ns	39961 ns	1.00
array/reductions/mapreduce/Int64/dims=1L	87861 ns	87917 ns	1.00
array/reductions/mapreduce/Int64/dims=2	56739 ns	57047 ns	0.99
array/reductions/mapreduce/Int64/dims=2L	81825 ns	82751 ns	0.99
array/reductions/reduce/Float32/1d	31155 ns	31482 ns	0.99
array/reductions/reduce/Float32/dims=1	36991 ns	37630 ns	0.98
array/reductions/reduce/Float32/dims=1L	49513 ns	50073 ns	0.99
array/reductions/reduce/Float32/dims=2	54316 ns	54518 ns	1.00
array/reductions/reduce/Float32/dims=2L	67515 ns	67963 ns	0.99
array/reductions/reduce/Int64/1d	38471 ns	38054 ns	1.01
array/reductions/reduce/Int64/dims=1	39921 ns	39983 ns	1.00
array/reductions/reduce/Int64/dims=1L	87622 ns	87988 ns	1.00
array/reductions/reduce/Int64/dims=2	56699 ns	56860 ns	1.00
array/reductions/reduce/Int64/dims=2L	82090 ns	82191 ns	1.00
array/reverse/1d	16416 ns	16098 ns	1.02
array/reverse/1dL	69054 ns	68844 ns	1.00
array/reverse/1dL_inplace	67058 ns	66918 ns	1.00
array/reverse/1d_inplace	8172.333333333333 ns	9757 ns	0.84
array/reverse/2d	19106 ns	19108 ns	1.00
array/reverse/2dL	72632 ns	72561 ns	1.00
array/reverse/2dL_inplace	66890 ns	66785 ns	1.00
array/reverse/2d_inplace	9557 ns	9397 ns	1.02
array/sorting/1d	2638930 ns	2656939 ns	0.99
array/sorting/2d	1027515 ns	1038211 ns	0.99
array/sorting/by	3181446 ns	3192363 ns	1.00
cuda/synchronization/context/auto	1014.5 ns	1015.3 ns	1.00
cuda/synchronization/context/blocking	787.9514563106796 ns	784.6960784313726 ns	1.00
cuda/synchronization/context/nonblocking	5791.2 ns	5655.166666666667 ns	1.02
cuda/synchronization/stream/auto	874.5555555555555 ns	857.3174603174604 ns	1.02
cuda/synchronization/stream/blocking	661.2784810126582 ns	654.7730061349694 ns	1.01
cuda/synchronization/stream/nonblocking	5682.666666666667 ns	5491 ns	1.03
integration/byval/reference	147357 ns	147245 ns	1.00
integration/byval/slices=1	151394 ns	149267 ns	1.01
integration/byval/slices=2	294544 ns	292191 ns	1.01
integration/byval/slices=3	437519 ns	434868 ns	1.01
integration/cudadevrt	104490 ns	104357 ns	1.00
integration/volumerhs	9144772 ns	9315717 ns	0.98
kernel/indexing	12473 ns	12183 ns	1.02
kernel/indexing_checked	13171 ns	13078 ns	1.01
kernel/launch	2015.111111111111 ns	2032.3333333333333 ns	0.99
kernel/occupancy	653.3475609756098 ns	667.8466257668712 ns	0.98
kernel/rand	13690 ns	14903 ns	0.92
latency/import	4128007713 ns	4116528672 ns	1.00
latency/precompile	4814647233 ns	4814427609 ns	1.00
latency/ttfp	5205590998 ns	5198634459 ns	1.00

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

[wsmoses]

you'll need to backport this to an earlier cuda to test per the gpucompiler incompat

[kshyatt]

I think I'll just wait until you merge Valentin's PR :D