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.

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

make_zero of CuArray returns the same array instead of deep copy #2186

Closed
utkinis opened this issue Dec 9, 2024 · 3 comments
Closed

make_zero of CuArray returns the same array instead of deep copy #2186

utkinis opened this issue Dec 9, 2024 · 3 comments

Comments

@utkinis
Copy link

utkinis commented Dec 9, 2024

Description

julia> using CUDA

julia> A = CUDA.rand(1)
1-element CuArray{Float32, 1, CUDA.DeviceMemory}:
 0.068770334

julia> make_zero(A) == A
true

julia> B = rand(1)
1-element Vector{Float64}:
 0.12168917809683022

julia> make_zero(B) == B
false

Enzyme version

Enzyme v0.13.21

Julia version

Same bug occurs both on 1.10.7 and 1.11.2

@wsmoses
Copy link
Member

wsmoses commented Dec 15, 2024

Can you reopen this on CUDA.jl? That's where the Enzyme cuda extension lives [and where the code would go]

@maleadt
Copy link

maleadt commented Dec 17, 2024

Can you reopen this on CUDA.jl? That's where the Enzyme cuda extension lives [and where the code would go]

The Enzyme.jl extension in CUDA.jl is only hosted there for practical reasons; none of the CUDA.jl maintainers works on it. So issues regarding it should be filed on Enzyme.jl, IMO (or they risk getting ignored).

@wsmoses
Copy link
Member

wsmoses commented Dec 18, 2024

resolved by CUDA.jl PR

@wsmoses wsmoses closed this as completed Dec 18, 2024
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

No branches or pull requests

3 participants