Use an appropriate amount of threads in unified memory example

[christiangnrd]

Split out 602e4f4 to merge earlier since it was also encountered in another PR: https://buildkite.com/julialang/metal-dot-jl/builds/1881#0196445f-e2c3-4961-9eb1-6819b6b2adab

I don't actually know if this will fix the issue, but this is something that should be fixed regardless

[github-actions]

Your PR requires formatting changes to meet the project's style guidelines.
Please consider running Runic (git runic main) to apply these changes.

Click here to view the suggested changes.

diff --git a/examples/unified_memory.jl b/examples/unified_memory.jl
index 3093db36..98866464 100644
--- a/examples/unified_memory.jl
+++ b/examples/unified_memory.jl
@@ -37,7 +37,7 @@ arr_cpu .= pi
 Metal.@allowscalar @test arr_mtl[1] == Float32(pi)
 
 # Now launch a kernel altering the Metal array
-Metal.@sync @metal threads=16 groups=16 simple_kernel(arr_mtl)
+Metal.@sync @metal threads = 16 groups = 16 simple_kernel(arr_mtl)
 
 # These changes are reflected in the wrapped CPU array
 synchronize()

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 80.73%. Comparing base (fab6fc2) to head (31f4b6d).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #583   +/-   ##
=======================================
  Coverage   80.73%   80.73%           
=======================================
  Files          61       61           
  Lines        2657     2657           
=======================================
  Hits         2145     2145           
  Misses        512      512           

☔ 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.

[github-actions]

Metal Benchmarks

Benchmark suite	Current: 31f4b6d	Previous: fab6fc2	Ratio
private array/construct	24694.333333333332 ns	24708.333333333332 ns	1.00
private array/broadcast	462541 ns	468458 ns	0.99
private array/random/randn/Float32	757583 ns	793417 ns	0.95
private array/random/randn!/Float32	626145.5 ns	636333 ns	0.98
private array/random/rand!/Int64	569375 ns	573208 ns	0.99
private array/random/rand!/Float32	599458 ns	603208 ns	0.99
private array/random/rand/Int64	734041.5 ns	775270.5 ns	0.95
private array/random/rand/Float32	639520.5 ns	626292 ns	1.02
private array/copyto!/gpu_to_gpu	658667 ns	642833 ns	1.02
private array/copyto!/cpu_to_gpu	694791 ns	826459 ns	0.84
private array/copyto!/gpu_to_cpu	815916 ns	662500 ns	1.23
private array/accumulate/1d	1323875 ns	1350959 ns	0.98
private array/accumulate/2d	1396708.5 ns	1396625 ns	1.00
private array/iteration/findall/int	2093187.5 ns	2086458 ns	1.00
private array/iteration/findall/bool	1843000 ns	1836208 ns	1.00
private array/iteration/findfirst/int	1712125 ns	1709417 ns	1.00
private array/iteration/findfirst/bool	1665187.5 ns	1664708 ns	1.00
private array/iteration/scalar	3884000 ns	3895500 ns	1.00
private array/iteration/logical	3196813 ns	3194188 ns	1.00
private array/iteration/findmin/1d	1765750 ns	1768916.5 ns	1.00
private array/iteration/findmin/2d	1348291 ns	1362729 ns	0.99
private array/reductions/reduce/1d	1038750 ns	1040771 ns	1.00
private array/reductions/reduce/2d	665145.5 ns	668666 ns	0.99
private array/reductions/mapreduce/1d	1044000 ns	1053500 ns	0.99
private array/reductions/mapreduce/2d	659792 ns	671333 ns	0.98
private array/permutedims/4d	2497666 ns	2545250 ns	0.98
private array/permutedims/2d	1020083.5 ns	1010417 ns	1.01
private array/permutedims/3d	1597812.5 ns	1577583 ns	1.01
private array/copy	613416 ns	597583 ns	1.03
latency/precompile	9740084292 ns	9625622625 ns	1.01
latency/ttfp	3751656542 ns	3681887084 ns	1.02
latency/import	1266783021 ns	1253500083 ns	1.01
integration/metaldevrt	715291 ns	712750 ns	1.00
integration/byval/slices=1	1600709 ns	1515459 ns	1.06
integration/byval/slices=3	11774584 ns	9046333 ns	1.30
integration/byval/reference	1624104 ns	1617542 ns	1.00
integration/byval/slices=2	2604292 ns	2591083 ns	1.01
kernel/indexing	485250 ns	479250 ns	1.01
kernel/indexing_checked	479833 ns	476500 ns	1.01
kernel/launch	8000 ns	9743.166666666668 ns	0.82
metal/synchronization/stream	14459 ns	14875 ns	0.97
metal/synchronization/context	14875 ns	15083 ns	0.99
shared array/construct	24736.166666666668 ns	25194.5 ns	0.98
shared array/broadcast	460209 ns	462792 ns	0.99
shared array/random/randn/Float32	823416 ns	782000 ns	1.05
shared array/random/randn!/Float32	642291 ns	641834 ns	1.00
shared array/random/rand!/Int64	582208 ns	579584 ns	1.00
shared array/random/rand!/Float32	603937.5 ns	594875 ns	1.02
shared array/random/rand/Int64	782354.5 ns	763250 ns	1.03
shared array/random/rand/Float32	610750 ns	614083 ns	0.99
shared array/copyto!/gpu_to_gpu	84125 ns	83083 ns	1.01
shared array/copyto!/cpu_to_gpu	85916 ns	83459 ns	1.03
shared array/copyto!/gpu_to_cpu	82458 ns	82458 ns	1
shared array/accumulate/1d	1347375 ns	1364875 ns	0.99
shared array/accumulate/2d	1400042 ns	1400125 ns	1.00
shared array/iteration/findall/int	1788041.5 ns	1836791.5 ns	0.97
shared array/iteration/findall/bool	1579000 ns	1594000 ns	0.99
shared array/iteration/findfirst/int	1411083 ns	1402333 ns	1.01
shared array/iteration/findfirst/bool	1524542 ns	1378916.5 ns	1.11
shared array/iteration/scalar	156084 ns	160334 ns	0.97
shared array/iteration/logical	3021458 ns	3001833 ns	1.01
shared array/iteration/findmin/1d	1472562.5 ns	1475937.5 ns	1.00
shared array/iteration/findmin/2d	1371833 ns	1372979 ns	1.00
shared array/reductions/reduce/1d	741291.5 ns	719291 ns	1.03
shared array/reductions/reduce/2d	668917 ns	673937.5 ns	0.99
shared array/reductions/mapreduce/1d	746729.5 ns	752104 ns	0.99
shared array/reductions/mapreduce/2d	671375 ns	669166.5 ns	1.00
shared array/permutedims/4d	2531334 ns	2524375 ns	1.00
shared array/permutedims/2d	1039875 ns	1021916.5 ns	1.02
shared array/permutedims/3d	1604916 ns	1572708 ns	1.02
shared array/copy	246791 ns	245041 ns	1.01

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