Work around REPL invalidating precompilation

[christiangnrd]

Seems to be 1.14-only

[maleadt]

Hmm, I'd rather we investigate this first. The whole premise of JuliaLang/julia#61714 was to avoid needing this.

[github-actions]

Metal Benchmarks

Details

Benchmark suite	Current: c9821eb	Previous: a60a3d1	Ratio
array/accumulate/Float32/1d	814542 ns	810333 ns	1.01
array/accumulate/Float32/dims=1	1017667 ns	961459 ns	1.06
array/accumulate/Float32/dims=1L	10065062 ns	10097312 ns	1.00
array/accumulate/Float32/dims=2	1307854.5 ns	1279770.5 ns	1.02
array/accumulate/Float32/dims=2L	7019208 ns	6770499.5 ns	1.04
array/accumulate/Int64/1d	969916 ns	1002916 ns	0.97
array/accumulate/Int64/dims=1	1188625 ns	1112666 ns	1.07
array/accumulate/Int64/dims=1L	12041437.5 ns	11929875 ns	1.01
array/accumulate/Int64/dims=2	1490458 ns	1497750 ns	1.00
array/accumulate/Int64/dims=2L	9510000 ns	9555875 ns	1.00
array/broadcast	379250 ns	370417 ns	1.02
array/construct	5208 ns	5209 ns	1.00
array/permutedims/2d	569687.5 ns	666292 ns	0.86
array/permutedims/3d	1126312.5 ns	1126625 ns	1.00
array/permutedims/4d	1993479 ns	1995437.5 ns	1.00
array/private/copy	438375 ns	439417 ns	1.00
array/private/copyto!/cpu_to_gpu	379833 ns	378667 ns	1.00
array/private/copyto!/gpu_to_cpu	377000 ns	377625 ns	1.00
array/private/copyto!/gpu_to_gpu	354833 ns	351083 ns	1.01
array/private/iteration/findall/bool	1087500 ns	1110084 ns	0.98
array/private/iteration/findall/int	1254916 ns	1292083 ns	0.97
array/private/iteration/findfirst/bool	1465750 ns	1461354 ns	1.00
array/private/iteration/findfirst/int	1491834 ns	1530124.5 ns	0.97
array/private/iteration/findmin/1d	1605083 ns	1605875 ns	1.00
array/private/iteration/findmin/2d	1276666.5 ns	1344833 ns	0.95
array/private/iteration/logical	1756792 ns	1817354 ns	0.97
array/private/iteration/scalar	2706875 ns	2816708 ns	0.96
array/random/rand/Float32	629875 ns	615083 ns	1.02
array/random/rand/Int64	674000 ns	744334 ns	0.91
array/random/rand!/Float32	584334 ns	574750 ns	1.02
array/random/rand!/Int64	510708 ns	511833 ns	1.00
array/random/randn/Float32	584750 ns	594333 ns	0.98
array/random/randn!/Float32	535875 ns	530333 ns	1.01
array/reductions/mapreduce/Float32/1d	762459 ns	772959 ns	0.99
array/reductions/mapreduce/Float32/dims=1	513042 ns	504208 ns	1.02
array/reductions/mapreduce/Float32/dims=1L	754062.5 ns	744313 ns	1.01
array/reductions/mapreduce/Float32/dims=2	523625 ns	514125 ns	1.02
array/reductions/mapreduce/Float32/dims=2L	1348208 ns	1362041 ns	0.99
array/reductions/mapreduce/Int64/1d	950459 ns	1007292 ns	0.94
array/reductions/mapreduce/Int64/dims=1	800708 ns	809625 ns	0.99
array/reductions/mapreduce/Int64/dims=1L	1381041 ns	1400375 ns	0.99
array/reductions/mapreduce/Int64/dims=2	960791 ns	988791 ns	0.97
array/reductions/mapreduce/Int64/dims=2L	2225083 ns	2229125 ns	1.00
array/reductions/reduce/Float32/1d	754562.5 ns	748000 ns	1.01
array/reductions/reduce/Float32/dims=1	512458 ns	509208 ns	1.01
array/reductions/reduce/Float32/dims=1L	739667 ns	790000 ns	0.94
array/reductions/reduce/Float32/dims=2	521687.5 ns	515916 ns	1.01
array/reductions/reduce/Float32/dims=2L	1346167 ns	1351125 ns	1.00
array/reductions/reduce/Int64/1d	942583 ns	961542 ns	0.98
array/reductions/reduce/Int64/dims=1	797833 ns	799625 ns	1.00
array/reductions/reduce/Int64/dims=1L	1440917 ns	1397666 ns	1.03
array/reductions/reduce/Int64/dims=2	961833 ns	1000396 ns	0.96
array/reductions/reduce/Int64/dims=2L	2208375 ns	2211979.5 ns	1.00
array/shared/copy	242020.5 ns	227666 ns	1.06
array/shared/copyto!/cpu_to_gpu	39583 ns	39458 ns	1.00
array/shared/copyto!/gpu_to_cpu	40083 ns	40250 ns	1.00
array/shared/copyto!/gpu_to_gpu	40542 ns	41187.5 ns	0.98
array/shared/iteration/findall/bool	1094542 ns	1118750 ns	0.98
array/shared/iteration/findall/int	1254584 ns	1284375 ns	0.98
array/shared/iteration/findfirst/bool	1177000 ns	1176958 ns	1.00
array/shared/iteration/findfirst/int	1220584 ns	1304333 ns	0.94
array/shared/iteration/findmin/1d	1332250 ns	1372375 ns	0.97
array/shared/iteration/findmin/2d	1331958 ns	1241416.5 ns	1.07
array/shared/iteration/logical	1611625 ns	1714708 ns	0.94
array/shared/iteration/scalar	4613.142857142857 ns	4678.571428571428 ns	0.99
integration/byval/reference	1158292 ns	1159792 ns	1.00
integration/byval/slices=1	1161500 ns	1163208 ns	1.00
integration/byval/slices=2	2088041 ns	2086166 ns	1.00
integration/byval/slices=3	7886146 ns	13185584 ns	0.60
integration/metaldevrt	489125 ns	501479.5 ns	0.98
kernel/indexing	370834 ns	367375 ns	1.01
kernel/indexing_checked	546208 ns	546958 ns	1.00
kernel/launch	10250 ns	10417 ns	0.98
kernel/rand	567042 ns	556042 ns	1.02
latency/import	1644762584 ns	1640294083 ns	1.00
latency/precompile	35933496937.5 ns	35653121521 ns	1.01
latency/ttfp	1971478333 ns	1969154250 ns	1.00
metal/synchronization/context	610.1833333333333 ns	605.1988950276243 ns	1.01
metal/synchronization/stream	393.9752475247525 ns	397.2772277227723 ns	0.99

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 84.19%. Comparing base (a60a3d1) to head (c9821eb).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #844   +/-   ##
=======================================
  Coverage   84.19%   84.19%           
=======================================
  Files          68       68           
  Lines        4042     4042           
=======================================
  Hits         3403     3403           
  Misses        639      639           

☔ 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.