Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

Inplace irfft calculations with `mul!` mutate the fourier-space array

Open
  • Is something wrong?
#101 0 comments 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
35/100
Issue type
Bug
Clarity
Clearly specified
Activity status
Stale
Tech stack
julia
Domain
backend

Research direction

Start by running the provided Julia reproduction using plan_irfft and LinearAlgebra.mul! with FFTW. Trace the planned preallocated path and compare it with irfft and plan * uhat; done means mul! produces the correct output without mutating the Fourier-space input array.

Written by the indexing model from the issue text.

Description

I ran into some unexpected behavior with both FFTW.jl and FastTransforms.jl but not CUFFT.jl. I am thinking this repo is the appropriate location to raise this issue, but its possible it is an issue with both FFTW.jl and FastTransforms.jl

Summary

When applying a irfft plan to a Fourier space matrix uhat with mul! and a preallocated output array u, uhat is unexpectedly mutated.

Given a uhat matrix of elements ComplexF64 and shape (3,4), these work as expected:

u = irfft(uhat, 4, 1:2)

plan = plan_irfft(uhat, 4, 1:2)
u = plan * uhat

this generates the correct u, but mutates uhat:

plan = plan_irfft(uhat, 4, 1:2)
u = zeros(4,4)
mul!(u, plan, uhat)

Reproduction

using FFTW
using LinearAlgebra: mul!

# (3,4) array
uhat = ComplexF64.([
    1 2 3 4;
    1 2 3 4;
    1 2 3 4;
])

#
# first, perform an irfft without a plan
#
u = irfft(uhat, 4, 1:2)

println("\n\n########\n########")
println("unplanned irfft:")
display(u)
println("û: unchanged:")
display(uhat)

# generate an irfft plan
plan = plan_irfft(uhat, 4, 1:2)

#
# calculate the irfft with the plan (allocates)
#
u_plan = plan * uhat

println("\n\n########\n########")
println("planned irfft with allocation:")
display(u_plan)
println("uhat: unchanged:")
display(uhat)

#
# calculate the irfft with the plan (reuse allocation)
#

u_plan_preallocation = zeros(4,4)
mul!(u_plan_preallocation, plan, uhat)

println("\n\n########\n########")
println("planned irfft with preallocation:")
display(u_plan)
println("uhat: HAS changed:")
display(uhat)

This outputs:

########
########
unplanned irfft:
4×4 Matrix{Float64}:
 2.5  -0.5   -0.5  -0.5
 0.0   0.25   0.0  -0.25
 0.0   0.0    0.0   0.0
 0.0  -0.25   0.0   0.25
û: unchanged:
3×4 Matrix{ComplexF64}:
 1.0+0.0im  2.0+0.0im  3.0+0.0im  4.0+0.0im
 1.0+0.0im  2.0+0.0im  3.0+0.0im  4.0+0.0im
 1.0+0.0im  2.0+0.0im  3.0+0.0im  4.0+0.0im


########
########
planned irfft with allocation:
4×4 Matrix{Float64}:
 2.5  -0.5   -0.5  -0.5
 0.0   0.25   0.0  -0.25
 0.0   0.0    0.0   0.0
 0.0  -0.25   0.0   0.25
uhat: unchanged:
3×4 Matrix{ComplexF64}:
 1.0+0.0im  2.0+0.0im  3.0+0.0im  4.0+0.0im
 1.0+0.0im  2.0+0.0im  3.0+0.0im  4.0+0.0im
 1.0+0.0im  2.0+0.0im  3.0+0.0im  4.0+0.0im


########
########
planned irfft with preallocation:
4×4 Matrix{Float64}:
 2.5  -0.5   -0.5  -0.5
 0.0   0.25   0.0  -0.25
 0.0   0.0    0.0   0.0
 0.0  -0.25   0.0   0.25
uhat: HAS changed:
3×4 Matrix{ComplexF64}:
 10.0+0.0im  -2.0-2.0im  -2.0+0.0im  -2.0+2.0im
 10.0+0.0im  -2.0-2.0im  -2.0+0.0im  -2.0+2.0im
 10.0+0.0im  -2.0-2.0im  -2.0+0.0im  -2.0+2.0im
Dominant language
Julia
Stars
135
Forks
40
No merged PRs in 30d

Getting set up

This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from JuliaMath/AbstractFFTs.jl

All issues in JuliaMath/AbstractFFTs.jl

Similar issues

More Julia issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.