Skip to content

Test GPUArrays reverse #648

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

Draft
wants to merge 3 commits into
base: main
Choose a base branch
from
Draft

Test GPUArrays reverse #648

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
6b7e97c
Whitespace
christiangnrd Aug 1, 2025
25f4c0b
Add `reverse` benchmarks
christiangnrd Oct 9, 2025
e3ce3ae
[TEMP]
christiangnrd Oct 9, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions lib/mtl/capture.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -59,7 +59,7 @@ function MTLCaptureDescriptor()
end

# TODO: Add capture state
function MTLCaptureDescriptor(obj::Union{MTLDevice,MTLCommandQueue, MTLCaptureScope},
function MTLCaptureDescriptor(obj::Union{MTLDevice,MTLCommandQueue,MTLCaptureScope},
destination::MTLCaptureDestination;
folder::String=nothing)
desc = MTLCaptureDescriptor()
Expand Down Expand Up @@ -110,7 +110,7 @@ end

Start GPU frame capture using the default capture object and specifying capture descriptor parameters directly.
"""
function startCapture(obj::Union{MTLDevice,MTLCommandQueue, MTLCaptureScope},
function startCapture(obj::Union{MTLDevice,MTLCommandQueue,MTLCaptureScope},
destination::MTLCaptureDestination=MTLCaptureDestinationGPUTraceDocument;
folder::String=nothing)
if destination == MTLCaptureDestinationGPUTraceDocument && folder === nothing
Expand Down
1 change: 1 addition & 0 deletions perf/Project.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,5 +3,6 @@ BenchmarkTools = "6e4b80f9-dd63-53aa-95a3-0cdb28fa8baf"
HTTP = "cd3eb016-35fb-5094-929b-558a96fad6f3"
JSON = "682c06a0-de6a-54ab-a142-c8b1cf79cde6"
Metal = "dde4c033-4e86-420c-a63e-0dd931031962"
Pkg = "44cfe95a-1eb2-52ea-b672-e2afdf69b78f"
StableRNGs = "860ef19b-820b-49d6-a774-d7a799459cd3"
StaticArrays = "90137ffa-7385-5640-81b9-e52037218182"
17 changes: 11 additions & 6 deletions perf/array.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,18 +53,23 @@ cpu_mat = rand(rng, Float32, m, n)
gpu_mat = MtlMatrix{Float32}(cpu_mat)
gpu_mat_long = MtlMatrix{Float32}(rand(rng, Float32, m_long, n_long))
gpu_vec = reshape(gpu_mat, length(gpu_mat))
gpu_vec_long = reshape(gpu_mat_long, length(gpu_mat_long))
gpu_arr_3d = reshape(gpu_mat, (m, 40, 25))
gpu_arr_4d = reshape(gpu_mat, (m, 10, 10, 10))
gpu_mat_ints = MtlMatrix{Int64}(rand(rng, -10:10, m, n))
gpu_mat_long_ints = MtlMatrix{Int64}(rand(rng, -10:10, m_long, n_long))
gpu_vec_ints = reshape(gpu_mat_ints, length(gpu_mat_ints))

# let group = addgroup!(group, "reverse")
# group["1d"] = @benchmarkable Metal.@sync reverse($gpu_vec)
# group["2d"] = @benchmarkable Metal.@sync reverse($gpu_mat; dims=1)
# group["1d_inplace"] = @benchmarkable Metal.@sync reverse!($gpu_vec)
# group["2d_inplace"] = @benchmarkable Metal.@sync reverse!($gpu_mat; dims=1)
# end
let group = addgroup!(group, "reverse")
group["1d"] = @benchmarkable Metal.@sync reverse($gpu_vec)
group["1dL"] = @benchmarkable Metal.@sync reverse($gpu_vec_long)
group["2d"] = @benchmarkable Metal.@sync reverse($gpu_mat; dims=1)
group["2dL"] = @benchmarkable Metal.@sync reverse($gpu_mat_long; dims=1)
group["1d_inplace"] = @benchmarkable Metal.@sync reverse!($gpu_vec)
group["1dL_inplace"] = @benchmarkable Metal.@sync reverse!($gpu_vec_long)
group["2d_inplace"] = @benchmarkable Metal.@sync reverse!($gpu_mat; dims=1)
group["2dL_inplace"] = @benchmarkable Metal.@sync reverse!($gpu_mat_long; dims=2)
end

# 'evals=1' added to prevent hang when running benchmarks of CI
# TODO: Investigate cause and properly fix.
Expand Down
3 changes: 3 additions & 0 deletions perf/runbenchmarks.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,8 @@
# benchmark suite execution and codespeed submission

using Pkg
Pkg.add(url="https://github.com/christiangnrd/GPUArrays.jl", rev="reverse")

using Metal

using BenchmarkTools
Expand Down
1 change: 1 addition & 0 deletions test/Project.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,7 @@ KernelAbstractions = "63c18a36-062a-441e-b654-da1e3ab1ce7c"
LLVM = "929cbde3-209d-540e-8aea-75f648917ca0"
LinearAlgebra = "37e2e46d-f89d-539d-b4ee-838fcccc9c8e"
ObjectiveC = "e86c9b32-1129-44ac-8ea0-90d5bb39ded9"
Pkg = "44cfe95a-1eb2-52ea-b672-e2afdf69b78f"
Printf = "de0858da-6303-5e67-8744-51eddeeeb8d7"
REPL = "3fa0cd96-eef1-5676-8a61-b3b8758bbffb"
Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
Expand Down
3 changes: 3 additions & 0 deletions test/runtests.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,3 +1,6 @@
using Pkg
Pkg.add(url="https://github.com/christiangnrd/GPUArrays.jl", rev="reverse")

using Distributed
using Dates
using Metal
Expand Down