Skip to content

KA 0.10 #68

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 12 commits into
base: main
Choose a base branch
from
Draft

KA 0.10 #68

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
857669f
KA 0.10
christiangnrd Nov 7, 2025
d8bdc95
[Temp (maybe)] benchmarks improvements
christiangnrd Nov 14, 2025
59fa29e
Extra shape tests
christiangnrd Nov 14, 2025
3321852
Support specifying benchmark output file
christiangnrd Nov 14, 2025
73acfe2
Show standard error and more improvements
christiangnrd Nov 14, 2025
8dce5fa
Implement algorithm selection heuristic
christiangnrd Nov 14, 2025
e8bc09c
Only check `block_size` argument when on GPU
christiangnrd Nov 16, 2025
e12445a
jyf
christiangnrd Nov 16, 2025
dcb01d9
Progress
christiangnrd Nov 16, 2025
a4610f7
Fix test on Metal
christiangnrd Nov 20, 2025
4ba365c
Fix?
christiangnrd Nov 20, 2025
1272f37
TEMP
christiangnrd Nov 20, 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 Project.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,7 @@
name = "AcceleratedKernels"
uuid = "6a4ca0a5-0e36-4168-a932-d9be78d558f1"
authors = ["Andrei-Leonard Nicusan <[email protected]> and contributors"]
version = "0.4.3"
version = "0.5"

[deps]
ArgCheck = "dce04be8-c92d-5529-be00-80e4d2c0e197"
Expand All @@ -19,7 +19,7 @@ AcceleratedKernelsoneAPIExt = "oneAPI"
[compat]
ArgCheck = "2"
GPUArraysCore = "0.2.0"
KernelAbstractions = "0.9.34"
KernelAbstractions = "0.10"
Markdown = "1"
UnsafeAtomics = "0.3.0"
julia = "1.10"
Expand Down
1 change: 1 addition & 0 deletions benchmark/Project.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@
AcceleratedKernels = "6a4ca0a5-0e36-4168-a932-d9be78d558f1"
BenchmarkPlots = "ab8c0f59-4072-4e0d-8f91-a91e1495eb26"
BenchmarkTools = "6e4b80f9-dd63-53aa-95a3-0cdb28fa8baf"
CUDA = "052768ef-5323-5732-b1bb-66c8b64840ba"
FileIO = "5789e2e9-d7fb-5bc7-8068-2c6fae9b9549"
GPUArrays = "0c68f7d7-f131-5f86-a1c3-88cf8149b2d7"
KernelAbstractions = "63c18a36-062a-441e-b654-da1e3ab1ce7c"
Expand Down
20 changes: 11 additions & 9 deletions benchmark/accumulate_nd.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,17 +12,19 @@ for T in [UInt32, Int64, Float32]

local randrange = T == Float32 ? T : T(1):T(100)

_group["base_dims=1"] = @benchmarkable @sb(Base.accumulate(+, v, init=$T(0), dims=1)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
_group["acck_dims=1"] = @benchmarkable @sb(AK.accumulate(+, v, init=$T(0), dims=1)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
for (suff, (n1, n2)) in (("L", (3, 1_000_000)), ("", (512, 1000)))
_group["base_dims=1$suff"] = @benchmarkable @sb(Base.accumulate(+, v, init=$T(0), dims=1)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
_group["acck_dims=1$suff"] = @benchmarkable @sb(AK.accumulate(+, v, init=$T(0), dims=1)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))

_group["base_dims=2"] = @benchmarkable @sb(Base.accumulate(+, v, init=$T(0), dims=2)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
_group["acck_dims=2"] = @benchmarkable @sb(AK.accumulate(+, v, init=$T(0), dims=2)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
_group["base_dims=2$suff"] = @benchmarkable @sb(Base.accumulate(+, v, init=$T(0), dims=2)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
_group["acck_dims=2$suff"] = @benchmarkable @sb(AK.accumulate(+, v, init=$T(0), dims=2)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))

T == Float32 || continue
T == Float32 || continue

_group["base_sincos_dims=1"] = @benchmarkable @sb(Base.accumulate(acc_f, v, init=$T(0), dims=1)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
_group["acck_sincos_dims=1"] = @benchmarkable @sb(AK.accumulate(acc_f, v, init=$T(0), neutral=$T(0), dims=1)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
_group["base_sincos_dims=1$suff"] = @benchmarkable @sb(Base.accumulate(acc_f, v, init=$T(0), dims=1)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
_group["acck_sincos_dims=1$suff"] = @benchmarkable @sb(AK.accumulate(acc_f, v, init=$T(0), neutral=$T(0), dims=1)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))

_group["base_sincos_dims=2"] = @benchmarkable @sb(Base.accumulate(acc_f, v, init=$T(0), dims=2)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
_group["acck_sincos_dims=2"] = @benchmarkable @sb(AK.accumulate(acc_f, v, init=$T(0), neutral=$T(0), dims=2)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
_group["base_sincos_dims=2$suff"] = @benchmarkable @sb(Base.accumulate(acc_f, v, init=$T(0), dims=2)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
_group["acck_sincos_dims=2$suff"] = @benchmarkable @sb(AK.accumulate(acc_f, v, init=$T(0), neutral=$T(0), dims=2)) setup=(v = ArrayType(rand(rng, $randrange, n1, n2)))
end
end
Loading
Loading