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.

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

Add a dispatch for LinearAlgebra.norm2 #2302

Draft
wants to merge 1 commit into
base: master
Choose a base branch
from
Draft

Add a dispatch for LinearAlgebra.norm2 #2302

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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: 4 additions & 0 deletions lib/cublas/linalg.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -138,6 +138,10 @@ function LinearAlgebra.norm(x::DenseCuArray{<:Union{Float16, ComplexF16, CublasF
end
end

function LinearAlgebra.norm2(x::SubArray{T,N,P} where {T<:Union{Float16, ComplexF16, CublasFloat}, N, P<:DenseCuArray{<:T}})
return nrm2(x)
end

LinearAlgebra.BLAS.asum(x::StridedCuArray{<:CublasFloat}) = asum(length(x), x)

function LinearAlgebra.axpy!(alpha::Number, x::StridedCuArray{T}, y::StridedCuArray{T}) where T<:Union{Float16, ComplexF16, CublasFloat}
Expand Down
11 changes: 11 additions & 0 deletions test/libraries/cublas.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1767,6 +1767,17 @@ end
@view(p[reshape(1:(out*inn),out,inn)]) * x
end
end

@testset "nrm2 with strided inputs" begin # JuliaGPU/CUDA.jl#2280
cudaTypes = (Float16, ComplexF16, CublasFloat)
for CT in cudaTypes
x = rand(CT, 10, 10, 10)
dx = CuArray(x)
dx_ = @view dx[3:6, 1:5, :]
x_ = @view x[3:6, 1:5, :]
@test norm(dx_, 2) ≈ norm(x_, 2)
end
end
end

############################################################################################
Expand Down