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.

Sign up for GitHub

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

Jump to bottom

Remove device_code_agx #512

Merged
merged 2 commits into from
Jan 6, 2025
Merged

Remove device_code_agx #512

merged 2 commits into from
Jan 6, 2025
+13 −206

Conversation

christiangnrd
Copy link
Contributor

No description provided.

@christiangnrd christiangnrd mentioned this pull request Dec 26, 2024
Closed
github-actions[bot]
github-actions bot reviewed Dec 26, 2024
View reviewed changes
Copy link
Contributor

@github-actions github-actions bot left a comment •
edited
Loading

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: 9fdd899 Previous: cf21f9d Ratio
private array/construct 26257 ns 27517.85714285714 ns 0.95
private array/broadcast 465709 ns 457584 ns 1.02
private array/random/randn/Float32 907708 ns 821104.5 ns 1.11
private array/random/randn!/Float32 603125 ns 616500 ns 0.98
private array/random/rand!/Int64 558312.5 ns 548167 ns 1.02
private array/random/rand!/Float32 553958 ns 582125 ns 0.95
private array/random/rand/Int64 932167 ns 764500 ns 1.22
private array/random/rand/Float32 773250 ns 598542 ns 1.29
private array/copyto!/gpu_to_gpu 581208 ns 646125 ns 0.90
private array/copyto!/cpu_to_gpu 668625 ns 810041.5 ns 0.83
private array/copyto!/gpu_to_cpu 696125.5 ns 644937.5 ns 1.08
private array/accumulate/1d 1423083.5 ns 1320000 ns 1.08
private array/accumulate/2d 1478583.5 ns 1378958 ns 1.07
private array/iteration/findall/int 2282291 ns 2020834 ns 1.13
private array/iteration/findall/bool 1975021 ns 1779625 ns 1.11
private array/iteration/findfirst/int 1793437.5 ns 1696916 ns 1.06
private array/iteration/findfirst/bool 1728229 ns 1650188 ns 1.05
private array/iteration/scalar 2477812.5 ns 3549500 ns 0.70
private array/iteration/logical 3496708 ns 3174729.5 ns 1.10
private array/iteration/findmin/1d 1887604 ns 1747792 ns 1.08
private array/iteration/findmin/2d 1409500 ns 1339417 ns 1.05
private array/reductions/reduce/1d 977167 ns 1031312.5 ns 0.95
private array/reductions/reduce/2d 702354.5 ns 656042 ns 1.07
private array/reductions/mapreduce/1d 950917 ns 1027125 ns 0.93
private array/reductions/mapreduce/2d 695875 ns 666791 ns 1.04
private array/permutedims/4d 2671833.5 ns 2554792 ns 1.05
private array/permutedims/2d 1092084 ns 1026291.5 ns 1.06
private array/permutedims/3d 1830604 ns 1591209 ns 1.15
private array/copy 836021 ns 595583 ns 1.40
latency/precompile 5836284979 ns 5856015937.5 ns 1.00
latency/ttfp 3084285646 ns 6607073479.5 ns 0.47
latency/import 1165224083 ns 1171437917 ns 0.99
integration/metaldevrt 748166.5 ns 699417 ns 1.07
integration/byval/slices=1 1673645.5 ns 1594083.5 ns 1.05
integration/byval/slices=3 20180750 ns 11850083 ns 1.70
integration/byval/reference 1666646 ns 1622417 ns 1.03
integration/byval/slices=2 2839187.5 ns 2525791 ns 1.12
kernel/indexing 460792 ns 466979.5 ns 0.99
kernel/indexing_checked 467917 ns 479562.5 ns 0.98
kernel/launch 7959 ns 7958 ns 1.00
metal/synchronization/stream 14958.5 ns 14708 ns 1.02
metal/synchronization/context 15333 ns 14834 ns 1.03
shared array/construct 25895.833333333332 ns 25312.5 ns 1.02
shared array/broadcast 463312.5 ns 466145.5 ns 0.99
shared array/random/randn/Float32 912542 ns 784895.5 ns 1.16
shared array/random/randn!/Float32 594125 ns 648729 ns 0.92
shared array/random/rand!/Int64 555583 ns 543291 ns 1.02
shared array/random/rand!/Float32 560291 ns 579958.5 ns 0.97
shared array/random/rand/Int64 881583 ns 764687.5 ns 1.15
shared array/random/rand/Float32 846458.5 ns 636958 ns 1.33
shared array/copyto!/gpu_to_gpu 81000 ns 87791 ns 0.92
shared array/copyto!/cpu_to_gpu 80542 ns 94417 ns 0.85
shared array/copyto!/gpu_to_cpu 81042 ns 83333 ns 0.97
shared array/accumulate/1d 1428750 ns 1327875 ns 1.08
shared array/accumulate/2d 1493708 ns 1378167 ns 1.08
shared array/iteration/findall/int 2007084 ns 1773979 ns 1.13
shared array/iteration/findall/bool 1745895.5 ns 1575646 ns 1.11
shared array/iteration/findfirst/int 1504271 ns 1391812 ns 1.08
shared array/iteration/findfirst/bool 1425959 ns 1352062.5 ns 1.05
shared array/iteration/scalar 161833 ns 155709 ns 1.04
shared array/iteration/logical 3220687.5 ns 2959083.5 ns 1.09
shared array/iteration/findmin/1d 1575458.5 ns 1459687.5 ns 1.08
shared array/iteration/findmin/2d 1440791 ns 1353292 ns 1.06
shared array/reductions/reduce/1d 695958 ns 727896 ns 0.96
shared array/reductions/reduce/2d 697666.5 ns 660583 ns 1.06
shared array/reductions/mapreduce/1d 724583 ns 730833 ns 0.99
shared array/reductions/mapreduce/2d 699500 ns 662708.5 ns 1.06
shared array/permutedims/4d 2654187.5 ns 2539083.5 ns 1.05
shared array/permutedims/2d 1112791 ns 1019208 ns 1.09
shared array/permutedims/3d 1842813 ns 1600708 ns 1.15
shared array/copy 212375 ns 238729.5 ns 0.89

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

This was linked to issues Jan 6, 2025
@device_code_agx errors when Metal Shader Validation is enabled #463
Closed
Bad interaction between PyCall and Metal #500
Closed
@maleadt maleadt merged commit de8eed5 into main Jan 6, 2025
2 checks passed
@maleadt maleadt deleted the nopython branch January 6, 2025 07:37
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@github-actions github-actions[bot] github-actions[bot] left review comments

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

Bad interaction between PyCall and Metal @device_code_agx errors when Metal Shader Validation is enabled
2 participants
@christiangnrd @maleadt