Skip to content

change div to cld to avoid DivideError in FFT #648

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
jpsamaroo merged 1 commit into from
Aug 8, 2025
Merged

change div to cld to avoid DivideError in FFT #648

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
28 changes: 14 additions & 14 deletions ext/AbstractFFTsExt.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -98,19 +98,19 @@ end
function _fft!(output::DMatrix{T}, input::DMatrix{T}, dims=(1, 2)) where T
N = size(input, 1)
np = length(Dagger.compatible_processors())
A = zeros(Blocks(N, div(N, np)), T, size(input))
A = zeros(Blocks(N, cld(N, np)), T, size(input))
copyto!(A, input)
B = zeros(Blocks(div(N, np), N), T, size(input))
B = zeros(Blocks(cld(N, np), N), T, size(input))
__fft!(A, B, dims)
copyto!(output, B)
return output
end
function _ifft!(output::DMatrix{T}, input::DMatrix{T}, dims=(1, 2)) where T
N = size(input, 1)
np = length(Dagger.compatible_processors())
A = zeros(Blocks(N, div(N, np)), T, size(input))
A = zeros(Blocks(N, cld(N, np)), T, size(input))
copyto!(A, input)
B = zeros(Blocks(div(N, np), N), T, size(input))
B = zeros(Blocks(cld(N, np), N), T, size(input))
__ifft!(A, B, dims)
copyto!(output, B)
return output
Expand All @@ -121,16 +121,16 @@ function _fft!(output::DArray{T,3}, input::DArray{T,3}, dims=(1, 2, 3); decomp::
N = size(input, 1)
np = length(Dagger.compatible_processors())
if decomp isa Pencil
A = zeros(Blocks(N, div(N, np), div(N, np)), T, size(input))
B = zeros(Blocks(div(N, np), N, div(N, np)), T, size(input))
C = zeros(Blocks(div(N, np), div(N, np), N), T, size(input))
A = zeros(Blocks(N, cld(N, np), cld(N, np)), T, size(input))
B = zeros(Blocks(cld(N, np), N, cld(N, np)), T, size(input))
C = zeros(Blocks(cld(N, np), cld(N, np), N), T, size(input))
copyto!(A, input)
__fft!(decomp, A, B, C, dims)
copyto!(output, C)
return output
elseif decomp isa Slab
A = zeros(Blocks(N, N, div(N, np)), T, size(input))
B = zeros(Blocks(div(N, np), div(N, np), N), T, size(input))
A = zeros(Blocks(N, N, cld(N, np)), T, size(input))
B = zeros(Blocks(cld(N, np), cld(N, np), N), T, size(input))
copyto!(A, input)
__fft!(decomp, A, B, dims)
copyto!(output, B)
Expand All @@ -143,16 +143,16 @@ function _ifft!(output::DArray{T,3}, input::DArray{T,3}, dims=(1, 2, 3); decomp:
N = size(input, 1)
np = length(Dagger.compatible_processors())
if decomp isa Pencil
A = zeros(Blocks(div(N, np), div(N, np), N), T, size(input))
B = zeros(Blocks(div(N, np), N, div(N, np)), T, size(input))
C = zeros(Blocks(N, div(N, np), div(N, np)), T, size(input))
A = zeros(Blocks(cld(N, np), cld(N, np), N), T, size(input))
B = zeros(Blocks(cld(N, np), N, cld(N, np)), T, size(input))
C = zeros(Blocks(N, cld(N, np), cld(N, np)), T, size(input))
copyto!(A, input)
__ifft!(decomp, A, B, C, dims)
copyto!(output, C)
return output
elseif decomp isa Slab
A = zeros(Blocks(div(N, np), div(N, np), N), T, size(input))
B = zeros(Blocks(N, N, div(N, np)), T, size(input))
A = zeros(Blocks(cld(N, np), cld(N, np), N), T, size(input))
B = zeros(Blocks(N, N, cld(N, np)), T, size(input))
copyto!(A, input)
__ifft!(decomp, A, B, dims)
copyto!(output, B)
Expand Down