Skip to content

Support CUDA arrays #84

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

Merged
YingboMa merged 5 commits into from
May 2, 2022
Merged

Support CUDA arrays #84

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
250f10c
make CuYao work
GiggleLiu Apr 21, 2022
3460b72
fix note
GiggleLiu Apr 21, 2022
a2cbfa1
use restructure
GiggleLiu Apr 27, 2022
3b3f67d
use restructure
GiggleLiu Apr 27, 2022
bb75b74
update, revert changes
GiggleLiu Apr 27, 2022
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
1 change: 1 addition & 0 deletions .gitignore
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,3 +3,4 @@
*.jl.mem
deps/deps.jl
Manifest.toml
.vscode
Copy link
Copy Markdown
Member

@YingboMa YingboMa Apr 27, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

BTW, you can set up global gitignore.

Copy link
Copy Markdown
Contributor Author

@GiggleLiu GiggleLiu Apr 27, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the tip, should I revert this file?

Copy link
Copy Markdown
Member

@YingboMa YingboMa Apr 27, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nah, it's fine.

7 changes: 4 additions & 3 deletions src/arnoldi.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -101,9 +101,10 @@ function arnoldi(A, b; m=min(30, size(A, 1)), ishermitian=LinearAlgebra.ishermit
n = length(b)
U = ishermitian ? real(T) : T

V = similar(A, T, (n, m + 1))
H = similar(A, U, (m+1, m))
fill!(H, zero(U))
# V stores the Krylov vectors
V = similar(b, T, (n, m + 1))
# H is a small dense array in tridiagonal form
H = zeros(U, (m+1, m))
Copy link
Copy Markdown
Member

@YingboMa YingboMa Apr 27, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

H will alway be Matrix if you don't use similar. Is that intended?

Copy link
Copy Markdown
Contributor Author

@GiggleLiu GiggleLiu Apr 27, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, it is intended. This matrix is usually a very small one that can be easily diagonalized on CPU.
Also, there are a lot of element-wise operations to it.


Ks = KrylovSubspace{T, U, real(T), typeof(V), typeof(H)}(m, m, false, zero(real(T)), V, H)

Expand Down
4 changes: 3 additions & 1 deletion src/krylov_phiv.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -123,9 +123,11 @@ function expv!(w::AbstractVector{Complex{Tw}}, t::Complex{Tt}, Ks::KrylovSubspac
expH = exponential!(t * cache,expmethod)
expHe = @view(expH[:, 1])
end
lmul!(beta, mul!(w, @view(V[:, 1:m]), expHe)) # exp(A) ≈ norm(b) * V * exp(H)e
# `ArrayInterface.restructure` will convert the `expHe` to the target matrix type that can interact with `V`.
lmul!(beta, mul!(w, @view(V[:, 1:m]), compatible_multiplicative_operand(V, expHe))) # exp(A) ≈ norm(b) * V * exp(H)e
end

compatible_multiplicative_operand(::AbstractArray, source::AbstractArray) = source

############################
# Cache for phiv
Expand Down