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

MKL: FFTW could not create plan #252

Closed
JakobAsslaender opened this issue Oct 7, 2022 · 2 comments
Closed

MKL: FFTW could not create plan #252

JakobAsslaender opened this issue Oct 7, 2022 · 2 comments

Comments

@JakobAsslaender
Copy link

Hi everyone,
certain ffts with the MKL package throw error messages. The minimum working example is:

using FFTW
FFTW.set_provider!("mkl")

Restart Julia

using FFTW
x = randn(ComplexF32, 10,10,64,64);
fft!(x, 3:ndims(x))

which throws this error message:

ERROR: FFTW could not create plan
Stacktrace:
 [1] error(s::String)
   @ Base ./error.jl:35
 [2] macro expansion
   @ ~/.julia/packages/FFTW/sfy1o/src/fft.jl:600 [inlined]
 [3] (FFTW.cFFTWPlan{ComplexF32, -1, true, 4})(X::Array{ComplexF32, 4}, Y::Array{ComplexF32, 4}, region::UnitRange{Int64}, flags::UInt32, timelimit::Float64)
   @ FFTW ~/.julia/packages/FFTW/sfy1o/src/FFTW.jl:49
 [4] #plan_fft!#8
   @ ~/.julia/packages/FFTW/sfy1o/src/fft.jl:733 [inlined]
 [5] plan_fft!
   @ ~/.julia/packages/FFTW/sfy1o/src/fft.jl:723 [inlined]
 [6] fft!(x::Array{ComplexF32, 4}, region::UnitRange{Int64})
   @ AbstractFFTs ~/.julia/packages/AbstractFFTs/Wg2Yf/src/definitions.jl:63
 [7] top-level scope
   @ REPL[10]:1

I am using Julia 1.8.1 and FFTW v1.5.0.

The error occurs both with 32 and 64 bit floats. It does not occur, if I perform the FFT, e.g., over the dimensions 2:ndims(x).

Thanks for your help!

@stevengj
Copy link
Member

stevengj commented Oct 8, 2022

Duplicate of #242. Use the real FFTW library, and complain to Intel.

@stevengj stevengj closed this as completed Oct 8, 2022
@JakobAsslaender
Copy link
Author

JakobAsslaender commented Oct 8, 2022

Apologies for creating a duplicate. The only reason why I am using MKL is because I get segfaults with FFTW (issue #236)--otherwise I am a huge fan of FFTW, thanks for implementing it!

Would this: master...N5N3:FFTW.jl:SelfUse fix the problem, or are there downsides to this fix?

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

2 participants