Skip to content

fix: indexing for gpu_rand #615

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

Open
wants to merge 4 commits into
base: master
Choose a base branch
from
Open

fix: indexing for gpu_rand #615

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
9cb2366
fix: indexing for `gpu_rand`
avik-pal Sep 1, 2025
8daec65
chore: fix runic formatting
avik-pal Sep 1, 2025
39c3f4b
test: see which testset fails
avik-pal Sep 1, 2025
f37e4de
test: try using approx
avik-pal Sep 3, 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
5 changes: 3 additions & 2 deletions src/host/random.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -96,9 +96,10 @@ function Random.randn!(rng::RNG, A::AnyGPUArray{T}) where T <: Number
threads = (length(A) - 1) ÷ 2 + 1
@kernel function randn!(a, randstates)
i = @index(Global, Linear)
threadidx = @index(Local, Linear)
idx = 2*(i - 1) + 1
U1 = gpu_rand(T, i, randstates)
U2 = gpu_rand(T, i, randstates)
U1 = gpu_rand(T, threadidx, randstates)
U2 = gpu_rand(T, threadidx, randstates)
Z0 = sqrt(T(-2.0)*log(U1))*cos(T(2pi)*U2)
Z1 = sqrt(T(-2.0)*log(U1))*sin(T(2pi)*U2)
@inbounds a[idx] = Z0
Expand Down
4 changes: 2 additions & 2 deletions test/testsuite/random.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,7 +44,7 @@
@testset "randn" begin # normally-distributed
# XXX: randn calls sqrt, and Base's sqrt(::Complex) performs
# checked type conversions that throw boxed numbers.
for T in filter(isrealfloattype, eltypes), d in (2, (2,2))
for T in filter(isrealfloattype, eltypes), d in (2, (2, 2), (128, 128))
A = AT{T}(undef, d)
B = copy(A)
randn!(rng, A)
Expand All @@ -56,7 +56,7 @@
randn!(rng, A)
Random.seed!(rng, 1)
randn!(rng, B)
@test Array(A) == Array(B)
@test Array(A) ≈ Array(B)

if rng != cpu_rng
randn!(cpu_rng, A)
Expand Down
Loading