Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix copy tests #493

Merged
merged 5 commits into from
Dec 10, 2024
Merged

Fix copy tests #493

merged 5 commits into from
Dec 10, 2024

Conversation

christiangnrd
Copy link
Contributor

@christiangnrd christiangnrd commented Dec 10, 2024

MPSDataType enum definition had incorrect Complex numbers definitions. I also forgot that adjoint and transpose are different for those types when I originally wrote the test.

Only the first and last commit have functional changes

@christiangnrd

This comment was marked as outdated.

@christiangnrd christiangnrd marked this pull request as ready for review December 10, 2024 18:02
Copy link
Contributor

@github-actions github-actions bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Metal Benchmarks

Benchmark suite Current: 7114e17 Previous: d14426c Ratio
private array/construct 26351.285714285714 ns 27132 ns 0.97
private array/broadcast 465250 ns 462417 ns 1.01
private array/random/randn/Float32 826333.5 ns 793708.5 ns 1.04
private array/random/randn!/Float32 664875 ns 658417 ns 1.01
private array/random/rand!/Int64 572542 ns 556167 ns 1.03
private array/random/rand!/Float32 611458 ns 593875 ns 1.03
private array/random/rand/Int64 789750 ns 766209 ns 1.03
private array/random/rand/Float32 646709 ns 610667 ns 1.06
private array/copyto!/gpu_to_gpu 649417 ns 565084 ns 1.15
private array/copyto!/cpu_to_gpu 710750 ns 824875 ns 0.86
private array/copyto!/gpu_to_cpu 838062.5 ns 825938 ns 1.01
private array/accumulate/1d 1348041 ns 1344562 ns 1.00
private array/accumulate/2d 1392541.5 ns 1389833 ns 1.00
private array/iteration/findall/int 2108875 ns 2093583 ns 1.01
private array/iteration/findall/bool 1852604.5 ns 1830874.5 ns 1.01
private array/iteration/findfirst/int 1704958 ns 1689375 ns 1.01
private array/iteration/findfirst/bool 1667916.5 ns 1661708 ns 1.00
private array/iteration/scalar 3601396 ns 3921666 ns 0.92
private array/iteration/logical 3195999.5 ns 3195958.5 ns 1.00
private array/iteration/findmin/1d 1762000 ns 1765125 ns 1.00
private array/iteration/findmin/2d 1362042 ns 1349500 ns 1.01
private array/reductions/reduce/1d 1054708.5 ns 1035375 ns 1.02
private array/reductions/reduce/2d 651041.5 ns 660583 ns 0.99
private array/reductions/mapreduce/1d 1047562.5 ns 1044687 ns 1.00
private array/reductions/mapreduce/2d 667584 ns 660917 ns 1.01
private array/permutedims/4d 2711521 ns 2716417 ns 1.00
private array/permutedims/2d 1013292 ns 1015292 ns 1.00
private array/permutedims/3d 1585313 ns 1576791 ns 1.01
private array/copy 589437 ns 590167 ns 1.00
latency/precompile 5132796708 ns 5133928042 ns 1.00
latency/ttfp 6515887437.5 ns 6740222625 ns 0.97
latency/import 1146348708.5 ns 1147095896.5 ns 1.00
integration/metaldevrt 724792 ns 711042 ns 1.02
integration/byval/slices=1 1570083 ns 1631916 ns 0.96
integration/byval/slices=3 10912104.5 ns 8785708 ns 1.24
integration/byval/reference 1594083 ns 1557875 ns 1.02
integration/byval/slices=2 2592250.5 ns 2693167 ns 0.96
kernel/indexing 456541.5 ns 490416 ns 0.93
kernel/indexing_checked 455500 ns 478416 ns 0.95
kernel/launch 8334 ns 8208 ns 1.02
metal/synchronization/stream 15042 ns 14875 ns 1.01
metal/synchronization/context 15125 ns 14792 ns 1.02
shared array/construct 26927.166666666668 ns 26375 ns 1.02
shared array/broadcast 455084 ns 464917 ns 0.98
shared array/random/randn/Float32 812709 ns 789083 ns 1.03
shared array/random/randn!/Float32 641042 ns 622041.5 ns 1.03
shared array/random/rand!/Int64 563709 ns 561208 ns 1.00
shared array/random/rand!/Float32 603000 ns 570167 ns 1.06
shared array/random/rand/Int64 771000 ns 766459 ns 1.01
shared array/random/rand/Float32 611208 ns 607750 ns 1.01
shared array/copyto!/gpu_to_gpu 88208 ns 87792 ns 1.00
shared array/copyto!/cpu_to_gpu 89792 ns 86000 ns 1.04
shared array/copyto!/gpu_to_cpu 84000 ns 78083 ns 1.08
shared array/accumulate/1d 1345875 ns 1355875 ns 0.99
shared array/accumulate/2d 1400313 ns 1395104.5 ns 1.00
shared array/iteration/findall/int 1821188 ns 1817917 ns 1.00
shared array/iteration/findall/bool 1584667 ns 1572666 ns 1.01
shared array/iteration/findfirst/int 1390770.5 ns 1385542 ns 1.00
shared array/iteration/findfirst/bool 1366291.5 ns 1361167 ns 1.00
shared array/iteration/scalar 160875 ns 158250 ns 1.02
shared array/iteration/logical 2973042 ns 2966688 ns 1.00
shared array/iteration/findmin/1d 1462167 ns 1444000 ns 1.01
shared array/iteration/findmin/2d 1369667 ns 1362146 ns 1.01
shared array/reductions/reduce/1d 732792 ns 727709 ns 1.01
shared array/reductions/reduce/2d 667792 ns 664750 ns 1.00
shared array/reductions/mapreduce/1d 746416 ns 742479.5 ns 1.01
shared array/reductions/mapreduce/2d 674354 ns 657917 ns 1.02
shared array/permutedims/4d 2725020.5 ns 2713250 ns 1.00
shared array/permutedims/2d 1015708.5 ns 1013333 ns 1.00
shared array/permutedims/3d 1587292 ns 1627375 ns 0.98
shared array/copy 242250 ns 252375 ns 0.96

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

@christiangnrd christiangnrd merged commit c2207d2 into main Dec 10, 2024
2 checks passed
@christiangnrd christiangnrd deleted the fixcopytests branch December 10, 2024 20:15
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant