# Serious slowdown with FFTW v1.1.0 and v1.2.0 on Windows

**URL:** <https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726>\
**Category:** Performance\
**Created:** [February 16, 2020, 4:00pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726 "2020-02-16T16:00:24Z")\
**Posts on this page:** 20\
**Page:** 1

<div class="post-metadata">

**Author:** ![turtle](https://avatars.discourse-cdn.com/v4/letter/t/ecb155/32.png) [@turtle](https://discourse.julialang.org/u/turtle)\
**Post date:** [February 16, 2020, 4:00pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/1 "2020-02-16T16:00:24Z")

</div>

I have just upgraded from Julia 0.7.0 → 1.0.5 and 1.3.1, and experienced a serious slowdown.  
It seems to me that the reason is the default FFTW version that is:  
FFTW v0.2.4 for Julia 0.7.0  
FFTW v1.1.0 for Julia 1.0.5  
FFTW v1.2.0 for Julia 1.3.1.

If I revert back to FFTW v0.2.4 on Julia 1.0.5/1.3.1 then the speed is fast again.  
I have created a minimal code that reproduces the \>5x slowdown (see below).  
I have tested it on both 64-bit Windows 7 and Windows 10 with the same results.

I am stuck, so I would greatly appreciate any help!

```julia
using FFTW, LinearAlgebra

function myfft(plan,x,y)
    mul!(y,plan,x)
    return
end

siz = (25,36,45)
x = zeros(ComplexF32,siz)
y = zeros(ComplexF32,siz)
plan = plan_fft(x, flags=FFTW.MEASURE)

println("with function call:")
@time for i=1:5000 myfft(plan,x,y) end
@time for i=1:5000 myfft(plan,x,y) end

println("without function call:")
@time for i=1:5000 mul!(y,plan,x) end
@time for i=1:5000 mul!(y,plan,x) end

```

---

<div class="post-metadata">

**Author:** ![stevengj](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/stevengj/32/71_2.png) [@stevengj](https://discourse.julialang.org/u/stevengj)\
**Post date:** [February 16, 2020, 7:18pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/2 "2020-02-16T19:18:29Z")

</div>

Are you using threads? [slowdown in threaded code from julia 1.2 to julia 1.4-DEV · Issue #121 · JuliaMath/FFTW.jl · GitHub](https://github.com/JuliaMath/FFTW.jl/issues/121)

---

<div class="post-metadata">

**Author:** ![turtle](https://avatars.discourse-cdn.com/v4/letter/t/ecb155/32.png) [@turtle](https://discourse.julialang.org/u/turtle)\
**Post date:** [February 16, 2020, 7:50pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/3 "2020-02-16T19:50:41Z")

</div>

No, I am not using threads or anything fancy.  
The above code is really that minimal, and behaves as reported.

---

<div class="post-metadata">

**Author:** ![turtle](https://avatars.discourse-cdn.com/v4/letter/t/ecb155/32.png) [@turtle](https://discourse.julialang.org/u/turtle)\
**Post date:** [February 17, 2020, 7:05pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/4 "2020-02-17T19:05:24Z")

</div>

Today I had access to an old Ubuntu Linux box.  
At Julia 1.3.1 I have experienced no slowdown switching from FFTW v0.2.4 to v1.2.0.

So it seems to be a Windows-specific problem.  
I have changed the title accordingly.

---

<div class="post-metadata">

**Author:** ![stevengj](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/stevengj/32/71_2.png) [@stevengj](https://discourse.julialang.org/u/stevengj)\
**Post date:** [February 17, 2020, 7:17pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/5 "2020-02-17T19:17:13Z")

</div>

If you look at the `plan`, is it significantly different? I wonder if the Windows build is not using AVX or something?

---

<div class="post-metadata">

**Author:** ![turtle](https://avatars.discourse-cdn.com/v4/letter/t/ecb155/32.png) [@turtle](https://discourse.julialang.org/u/turtle)\
**Post date:** [February 17, 2020, 7:36pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/6 "2020-02-17T19:36:08Z")

</div>

You are absolutely right, the plans differ indeed.  
The two cases:

1. Windows 7 64-bit | Julia 0.7.0 | FFTW v0.2.4 | ~1.7 seconds

```julia
FFTW forward plan for 25×36×45 array of Complex{Float32}
(dft-rank>=2/1                                          
  (dft-vrank>=1-x45/1                                   
    (dft-rank>=2/1                                      
      (dft-direct-25-x36 "n1fv_25_avx2")                
      (dft-buffered-36-x5/25-2                          
        (dft-vrank>=1-x5/1                              
          (dft-ct-dit/9                                 
            (dftw-direct-9/64 "t1fv_9_avx2")            
            (dft-direct-4-x9 "n1fv_4_sse2")))           
        (dft-r2hc-1                                     
          (rdft-rank0-iter-co/2-x5-x36))                
        (dft-nop))))                                    
  (dft-buffered-45-x225/900-1                           
    (dft-vrank>=1-x225/1                                
      (dft-ct-dit/3                                     
        (dftw-direct-3/8 "t1fuv_3_avx2_128")            
        (dft-direct-15-x3 "n1fv_15_avx2_128")))         
    (dft-r2hc-1                                         
      (rdft-rank0-iter-co/2-x225-x45))                  
    (dft-nop)))                                         

```

1. Windows 7 64-bit | Julia 1.3.1 | FFTW v1.2.0 | ~9.5 seconds

```julia
FFTW forward plan for 25×36×45 array of Complex{Float32}
(dft-rank>=2/1                                          
  (dft-vrank>=1-x45/1                                   
    (dft-rank>=2/1                                      
      (dft-direct-25-x36 "n1_25")                       
      (dft-buffered-36-x5/25-2                          
        (dft-vrank>=1-x5/1                              
          (dft-ct-dit/3                                 
            (dftw-direct-3/4 "t1_3")                    
            (dft-direct-12-x3 "n1_12")))                
        (dft-r2hc-1                                     
          (rdft-rank0-iter-co/2-x5-x36))                
        (dft-nop))))                                    
  (dft-buffered-45-x225/900-1                           
    (dft-vrank>=1-x225/1                                
      (dft-ct-dit/3                                     
        (dftw-direct-3/4 "t1_3")                        
        (dft-direct-15-x3 "n1_15")))                    
    (dft-r2hc-1                                         
      (rdft-rank0-iter-co/2-x225-x45))                  
    (dft-nop)))

```

---

<div class="post-metadata">

**Author:** ![stevengj](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/stevengj/32/71_2.png) [@stevengj](https://discourse.julialang.org/u/stevengj)\
**Post date:** [February 18, 2020, 9:59pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/7 "2020-02-18T21:59:19Z")

</div>

> [@turtle](#):
>
> `(dft-direct-12-x3 "n1_12")`

This is not using AVX (or SSE), so it is not surprising that it is a lot slower. You can get the configuration options by looking at the `fftw_version` C global:

```julia
julia> using FFTW

julia> unsafe_string(cglobal((:fftw_version,FFTW.libfftw3), UInt8))
"fftw-3.3.9-sse2-avx2-avx2_128"

```

(which indicates that, on my machine, FFTW was configured with SSE2 and AVX2.)

---

<div class="post-metadata">

**Author:** ![turtle](https://avatars.discourse-cdn.com/v4/letter/t/ecb155/32.png) [@turtle](https://discourse.julialang.org/u/turtle)\
**Post date:** [February 18, 2020, 10:49pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/8 "2020-02-18T22:49:30Z")

</div>

Yes, I realize the difference.  
However, my “slow” setup of Julia 1.3.1 + FFTW v1.2.0 also returns the info:

```julia
julia> using FFTW

julia> unsafe_string(cglobal((:fftw_version,FFTW.libfftw3), UInt8))
"fftw-3.3.9-sse2-avx2-avx2_128"

```

So I am still badly stuck.  
It is also strange that nobody else complained about the precompiled binaries on Windows.  
So I beg all Windows users here to test this minimal code for the appropriate AVX plan:

```julia
siz = (25,36,45)
x = zeros(ComplexF32,siz)
plan = plan_fft(x, flags=FFTW.MEASURE)

println(plan)

```

---

<div class="post-metadata">

**Author:** ![jonas-kr](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/jonas-kr/32/1810_2.png) [@jonas-kr](https://discourse.julialang.org/u/jonas-kr)\
**Post date:** [February 18, 2020, 11:00pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/9 "2020-02-18T23:00:07Z")

</div>

I would like to test the MWE myself, however, printing the plan’s contents, either with display or with print, does not provide anything besides

```julia
julia> println(plan)
FFTW forward plan for 25×36×45 array of Complex{Float32}

julia> display(plan)
FFTW forward plan for 25×36×45 array of Complex{Float32}

```

---

<div class="post-metadata">

**Author:** ![stevengj](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/stevengj/32/71_2.png) [@stevengj](https://discourse.julialang.org/u/stevengj)\
**Post date:** [February 18, 2020, 11:18pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/10 "2020-02-18T23:18:03Z")

</div>

Sounds like you are using MKL, not FFTW.

---

<div class="post-metadata">

**Author:** ![Elrod](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/elrod/32/22461_2.png) [@Elrod](https://discourse.julialang.org/u/Elrod)\
**Post date:** [February 18, 2020, 11:30pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/11 "2020-02-18T23:30:53Z")

</div>

> [@turtle](#):
>
> siz = (25,36,45) x = zeros(ComplexF32,siz) y = zeros(ComplexF32,siz) plan = plan\_fft(x, flags=FFTW.MEASURE)

My work computer runs Windows, so I can confirm

```julia
julia> plan = plan_fft(x, flags=FFTW.MEASURE)
FFTW forward plan for 25×36×45 array of Complex{Float32}
(dft-rank>=2/1
  (dft-vrank>=1-x45/1
    (dft-rank>=2/1
      (dft-direct-25-x36 "n1_25")
      (dft-buffered-36-x25/25-2
        (dft-vrank>=1-x25/1
          (dft-ct-dit/3
            (dftw-direct-3/4 "t1_3")
            (dft-direct-12-x3 "n1_12")))
        (dft-r2hc-1
          (rdft-rank0-iter-ci/2-x25-x36))
        (dft-nop))))
  (dft-buffered-45-x225/900-1
    (dft-vrank>=1-x225/1
      (dft-ct-dit/3
        (dftw-direct-3/4 "t1_3")
        (dft-direct-15-x3 "n1_15")))
    (dft-r2hc-1
      (rdft-rank0-iter-ci/2-x225-x45))
    (dft-nop)))

julia> @benchmark mul!($y, $plan, $x)
BenchmarkTools.Trial:
  memory estimate: 0 bytes
  allocs estimate: 0
  --------------
  minimum time: 1.909 ms (0.00% GC)
  median time: 1.957 ms (0.00% GC)
  mean time: 2.270 ms (0.00% GC)
  maximum time: 54.895 ms (0.00% GC)
  --------------
  samples: 2193
  evals/sample: 1

```

It works within WSL:

```julia
julia> plan = plan_fft(x, flags=FFTW.MEASURE) FFTW forward plan for 25×36×45 array of Complex{Float32} (dft-rank>=2/1 (dft-vrank>=1-x45/1 (dft-rank>=2/1 (dft-direct-25-x36 "n1fv_25_avx2") (dft-buffered-36-x25/25-2 (dft-vrank>=1-x25/1 (dft-ct-dit/9 (dftw-direct-9/64 "t1fv_9_avx2") (dft-direct-4-x9 "n1fv_4_sse2"))) (dft-r2hc-1 (rdft-rank0-iter-ci/2-x25-x36)) (dft-nop)))) (dft-buffered-45-x225/900-1 (dft-vrank>=1-x225/1 (dft-ct-dit/3 (dftw-direct-3/8 "t1fuv_3_avx2_128") (dft-direct-15-x3 "n1fv_15_avx2_128"))) (dft-r2hc-1 (rdft-rank0-iter-co/2-x225-x45)) (dft-nop))) julia> @benchmark mul!($y, $plan, $x) BenchmarkTools.Trial: memory estimate: 0 bytes allocs estimate: 0 -------------- minimum time: 278.800 μs (0.00% GC) median time: 285.100 μs (0.00% GC) mean time: 295.158 μs (0.00% GC) maximum time: 835.200 μs (0.00% GC) -------------- samples: 10000 evals/sample: 1               

```

Unfortunately, pasting from WSL does not work. But the plan shows avx, and it benchmarks at 278.8 microseconds vs the 1.9 milliseconds (about 7x faster in WSL than Cygwin).

---

<div class="post-metadata">

**Author:** ![jebej](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/jebej/32/1784_2.png) [@jebej](https://discourse.julialang.org/u/jebej)\
**Post date:** [February 18, 2020, 11:34pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/12 "2020-02-18T23:34:24Z")

</div>

I can confirm that while the version string reports FFTW being configured with AVX and SSE2, the plan does not include those instructions, on FFTW.jl v1.1.0 & Julia 1.1.1.

The last fast version is v0.3. As soon as the PARTR thread work was merged in with v1.0.0 things get slower (~6x).

---

<div class="post-metadata">

**Author:** ![mbaz](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/mbaz/32/17295_2.png) [@mbaz](https://discourse.julialang.org/u/mbaz)\
**Post date:** [February 18, 2020, 11:47pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/13 "2020-02-18T23:47:13Z")

</div>

I have Julia 1.1.0 on Windows 10. I get

```
julia> unsafe_string(cglobal((:fftw_version,FFTW.libfftw3), UInt8))
"fftw-3.3.9-sse2-avx2-avx2_128"

```

but the plan is identical to @turtle’s slow one.

---

<div class="post-metadata">

**Author:** ![turtle](https://avatars.discourse-cdn.com/v4/letter/t/ecb155/32.png) [@turtle](https://discourse.julialang.org/u/turtle)\
**Post date:** [February 19, 2020, 12:43pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/14 "2020-02-19T12:43:47Z")

</div>

Dear @stevengj, @Elrod, @jebej, @mbaz and @jonas-kr,  
Many thanks for all your precious time, feedback and help!

Now it is clear that this is a general problem of the compiled Windows binary,  
and at present the user has no other choice than reverting to a previous version.  
So with all due respect I am asking the core developers to fix this problem somehow.  
Thank you very much in advance.

---

<div class="post-metadata">

**Author:** ![stevengj](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/stevengj/32/71_2.png) [@stevengj](https://discourse.julialang.org/u/stevengj)\
**Post date:** [February 19, 2020, 2:14pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/15 "2020-02-19T14:14:58Z")

</div>

What is the result of

```nohighlight
ccall((:fftw_have_simd_sse2, FFTW.libfftw3), Cint, ())

```

---

<div class="post-metadata">

**Author:** ![turtle](https://avatars.discourse-cdn.com/v4/letter/t/ecb155/32.png) [@turtle](https://discourse.julialang.org/u/turtle)\
**Post date:** [February 19, 2020, 2:20pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/16 "2020-02-19T14:20:58Z")

</div>

```julia
julia> using FFTW

julia> ccall((:fftw_have_simd_sse2, FFTW.libfftw3), Cint, ())
ERROR: ccall: could not find function fftw_have_simd_sse2 in library libfftw3-3.dll
Stacktrace:
 [1] top-level scope at .\REPL[2]:100:

```

But this is also the same for both Julia 1.3.1 + FFTW v1.2.0 and Julia 0.7.0 + FFTW v0.2.4

---

<div class="post-metadata">

**Author:** ![stevengj](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/stevengj/32/71_2.png) [@stevengj](https://discourse.julialang.org/u/stevengj)\
**Post date:** [February 19, 2020, 2:25pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/17 "2020-02-19T14:25:05Z")

</div>

> [@turtle](#):
>
> But this is also the same for both Julia 1.3.1 + FFTW v1.2.0 and Julia 0.7.0 + FFTW v0.2.4

Oh, grr, that’s because it’s not explicitly marked as a DLL export on Windows, so even if it is there you can’t call it. (It might still be working fine internally to FFTW, but there is no way to check that from Julia.)

---

<div class="post-metadata">

**Author:** ![stevengj](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/stevengj/32/71_2.png) [@stevengj](https://discourse.julialang.org/u/stevengj)\
**Post date:** [February 19, 2020, 5:03pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/18 "2020-02-19T17:03:06Z")

</div>

For further discussion, see: [FFTW windows build is not using SIMD · Issue #534 · JuliaPackaging/Yggdrasil · GitHub](https://github.com/JuliaPackaging/Yggdrasil/issues/534)

---

<div class="post-metadata">

**Author:** ![giordano](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/giordano/32/2166_2.png) [@giordano](https://discourse.julialang.org/u/giordano)\
**Post date:** [February 20, 2020, 8:43pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/19 "2020-02-20T20:43:58Z")

</div>

Is there anyone willing to test the binary at [https://github.com/JuliaPackaging/Yggdrasil/pull/536#issuecomment-588548546](https://github.com/JuliaPackaging/Yggdrasil/pull/536#issuecomment-588548546)?

---

<div class="post-metadata">

**Author:** ![turtle](https://avatars.discourse-cdn.com/v4/letter/t/ecb155/32.png) [@turtle](https://discourse.julialang.org/u/turtle)\
**Post date:** [February 20, 2020, 9:11pm UTC](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726/20 "2020-02-20T21:11:16Z")

</div>

Good news:

I have downloaded the suggested file, unzipped and replaced all directories in  
`C:\Users\MYNAME\.julia\artifacts\2193a89e52669d43b28c47b83e738b73d6ed7a50`  
i.e. `bin, include, lib, logs` and `share`. (actually share was not there originally)

And magic happened:  
`Julia 1.3.1 + FFTW v1.2.0` got fast again with the plan containing SSE2 + AVX instructions!

[Next page](https://discourse.julialang.org/t/serious-slowdown-with-fftw-v1-1-0-and-v1-2-0-on-windows/34726.md?page=2)