Skip to content

[do not merge] Test GPUArrays sort #2830

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

Open
wants to merge 3 commits into
base: master
Choose a base branch
from

Conversation

christiangnrd
Copy link
Member

No description provided.

Temporarily disable old interface to Base
[only julia]
[only benchmarks]
Copy link
Contributor

github-actions bot commented Aug 3, 2025

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

Click here to view the suggested changes.
diff --git a/perf/array.jl b/perf/array.jl
index 2a75a1b4a..d3e7fade2 100644
--- a/perf/array.jl
+++ b/perf/array.jl
@@ -140,7 +140,7 @@ end
 let group = addgroup!(group, "sorting")
     let group = addgroup!(group, "Float32")
         group["1d"] = @async_benchmarkable sort($gpu_vec)
-        group["by=sin"] = @async_benchmarkable sort($gpu_vec; by=sin)
+        group["by=sin"] = @async_benchmarkable sort($gpu_vec; by = sin)
         # group["dims=1"] = @async_benchmarkable sort($gpu_mat; dims=1)
         # group["dims=2"] = @async_benchmarkable sort($gpu_mat; dims=2)
         # group["dims=1L"] = @async_benchmarkable sort($gpu_mat_long; dims=1)
diff --git a/perf/runbenchmarks.jl b/perf/runbenchmarks.jl
index 98935bc46..f73d7630d 100644
--- a/perf/runbenchmarks.jl
+++ b/perf/runbenchmarks.jl
@@ -1,6 +1,6 @@
 # benchmark suite execution and codespeed submission
 using Pkg
-Pkg.add(url="https://github.com/christiangnrd/GPUArrays.jl", rev="sort")
+Pkg.add(url = "https://github.com/christiangnrd/GPUArrays.jl", rev = "sort")
 
 using CUDA
 
diff --git a/test/runtests.jl b/test/runtests.jl
index 4e41a17fc..f2b352149 100644
--- a/test/runtests.jl
+++ b/test/runtests.jl
@@ -1,5 +1,5 @@
 using Pkg
-Pkg.add(url="https://github.com/christiangnrd/GPUArrays.jl", rev="sort")
+Pkg.add(url = "https://github.com/christiangnrd/GPUArrays.jl", rev = "sort")
 
 using Distributed
 using Dates

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.

CUDA.jl Benchmarks

Benchmark suite Current: 5c1468f Previous: 205c238 Ratio
latency/precompile 42964341534 ns 42934926801 ns 1.00
latency/ttfp 7077207639 ns 7008552789 ns 1.01
latency/import 3573106910 ns 3569139582 ns 1.00
integration/volumerhs 9610147.5 ns 9606581 ns 1.00
integration/byval/slices=1 147346 ns 147311 ns 1.00
integration/byval/slices=3 426505 ns 426127 ns 1.00
integration/byval/reference 145113 ns 145282 ns 1.00
integration/byval/slices=2 287082 ns 286537 ns 1.00
integration/cudadevrt 103675 ns 103674 ns 1.00
kernel/indexing 14458 ns 14638.5 ns 0.99
kernel/indexing_checked 15270 ns 15045 ns 1.01
kernel/occupancy 674.2229299363057 ns 669.9465408805031 ns 1.01
kernel/launch 2240.222222222222 ns 2202.4444444444443 ns 1.02
kernel/rand 18560 ns 17466 ns 1.06
array/reverse/1d 20358.5 ns 20143 ns 1.01
array/reverse/2d 25090.5 ns 24692 ns 1.02
array/reverse/1d_inplace 11080 ns 11332 ns 0.98
array/reverse/2d_inplace 12921 ns 13662 ns 0.95
array/copy 21261 ns 21281 ns 1.00
array/iteration/findall/int 158451 ns 159966.5 ns 0.99
array/iteration/findall/bool 140550 ns 141602 ns 0.99
array/iteration/findfirst/int 162981 ns 163419 ns 1.00
array/iteration/findfirst/bool 165431.5 ns 165377 ns 1.00
array/iteration/scalar 71921.5 ns 76152 ns 0.94
array/iteration/logical 218979 ns 219912.5 ns 1.00
array/iteration/findmin/1d 47637.5 ns 47580 ns 1.00
array/iteration/findmin/2d 97345 ns 97060 ns 1.00
array/reductions/reduce/Int64/1d 44822 ns 43742.5 ns 1.02
array/reductions/reduce/Int64/dims=1 48131 ns 47519.5 ns 1.01
array/reductions/reduce/Int64/dims=2 61527 ns 62503 ns 0.98
array/reductions/reduce/Int64/dims=1L 89146 ns 89134 ns 1.00
array/reductions/reduce/Int64/dims=2L 87602 ns 87634.5 ns 1.00
array/reductions/reduce/Float32/1d 35149 ns 35637 ns 0.99
array/reductions/reduce/Float32/dims=1 41809 ns 51967.5 ns 0.80
array/reductions/reduce/Float32/dims=2 59863 ns 59824 ns 1.00
array/reductions/reduce/Float32/dims=1L 52759 ns 52680 ns 1.00
array/reductions/reduce/Float32/dims=2L 70573 ns 70568 ns 1.00
array/reductions/mapreduce/Int64/1d 43580 ns 43514 ns 1.00
array/reductions/mapreduce/Int64/dims=1 47410 ns 46605.5 ns 1.02
array/reductions/mapreduce/Int64/dims=2 61876 ns 62143.5 ns 1.00
array/reductions/mapreduce/Int64/dims=1L 89301 ns 89174 ns 1.00
array/reductions/mapreduce/Int64/dims=2L 87660 ns 87305.5 ns 1.00
array/reductions/mapreduce/Float32/1d 35567 ns 35464 ns 1.00
array/reductions/mapreduce/Float32/dims=1 41818 ns 42505.5 ns 0.98
array/reductions/mapreduce/Float32/dims=2 60063 ns 60252 ns 1.00
array/reductions/mapreduce/Float32/dims=1L 53013 ns 52803 ns 1.00
array/reductions/mapreduce/Float32/dims=2L 71181.5 ns 70795 ns 1.01
array/broadcast 20470 ns 20737 ns 0.99
array/copyto!/gpu_to_gpu 11225 ns 13192 ns 0.85
array/copyto!/cpu_to_gpu 218362 ns 217123 ns 1.01
array/copyto!/gpu_to_cpu 285943 ns 287100 ns 1.00
array/accumulate/Int64/1d 125581 ns 126109 ns 1.00
array/accumulate/Int64/dims=1 84313 ns 84201 ns 1.00
array/accumulate/Int64/dims=2 159644 ns 158968 ns 1.00
array/accumulate/Int64/dims=1L 1721644 ns 1710638 ns 1.01
array/accumulate/Int64/dims=2L 968733.5 ns 967410.5 ns 1.00
array/accumulate/Float32/1d 110060 ns 109994 ns 1.00
array/accumulate/Float32/dims=1 81469 ns 81343 ns 1.00
array/accumulate/Float32/dims=2 148307.5 ns 148659 ns 1.00
array/accumulate/Float32/dims=1L 1620335 ns 1619411 ns 1.00
array/accumulate/Float32/dims=2L 699169 ns 699433 ns 1.00
array/construct 1348.2 ns 1288.5 ns 1.05
array/random/randn/Float32 44934.5 ns 45344 ns 0.99
array/random/randn!/Float32 25169 ns 25330 ns 0.99
array/random/rand!/Int64 27718 ns 27554 ns 1.01
array/random/rand!/Float32 9023.666666666666 ns 8908.333333333334 ns 1.01
array/random/rand/Int64 30459 ns 30218 ns 1.01
array/random/rand/Float32 13314 ns 13361 ns 1.00
array/permutedims/4d 60398 ns 60397 ns 1.00
array/permutedims/2d 54386.5 ns 54394 ns 1.00
array/permutedims/3d 55326 ns 55362 ns 1.00
array/sorting/Int64/1d 786808 ns
array/sorting/Float32/1d 680277 ns
array/sorting/Float32/by=sin 1503703 ns
cuda/synchronization/stream/auto 1037.7 ns 1066.6 ns 0.97
cuda/synchronization/stream/nonblocking 7362.1 ns 7691.3 ns 0.96
cuda/synchronization/stream/blocking 816.2444444444444 ns 844.0121951219512 ns 0.97
cuda/synchronization/context/auto 1210.4 ns 1211.4 ns 1.00
cuda/synchronization/context/nonblocking 8473.7 ns 6881.1 ns 1.23
cuda/synchronization/context/blocking 953 ns 924.7692307692307 ns 1.03

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

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