How to increase stack size for Julia in Windows?

I mirror here my question at StackOverflow: How to increase stack size for Julia in Windows? - Stack Overflow

I wrote a recursive function (basically a flood fill), it works fine on smaller datasets, but for slightly larger input it throws StackOverflowError .

How to increase the stack size for Julia under Windows 10? Ideally the solution should be applicable to JupyterLab.

It’s a singe use program, no point in optimizing/rewriting it, I just need to peak at the result and forget about the code.

As a test case, I provide the following MWE. This is just a simple algorithm that recursively visits each cell of n by n array:

n = 120

visited = fill(false, (n,n))

function visit_single_neighbour(i,j,Δi,Δj)
  if 1 ≤ i + Δi ≤ n && 1 ≤ j + Δj ≤ n
    if !visited[i+Δi, j+Δj]
      visited[i+Δi, j+Δj] = true
      visit_four_neighbours(i+Δi, j+Δj)

function visit_four_neighbours(i,j)

@time visit_four_neighbours(1,1)

For n = 120 the output is 0.003341 seconds , but for n = 121 it throws StackOverflowError .

On a Linux machine with ulimit -s unlimited the code runs no problem for n = 2000 and takes about 2.4 seconds.

Welcome to Julia.

Ignoring that, assuming that your real problem is similar, I say it would be easy to use a loop instead of recursion and it would be much faster too. Doing some more thinking you can get rid of


This doesn’t exist. I bet you already run it several times.

AFAIK there is nothing like ulimit for windows. The stack size is set during build/compile of the .exe. There is editbin, which comes with Visual Studio. You can try it on the julia.exe but I doubt that this is working seamless as you expect it: /STACK | Microsoft Docs

Proving myself wrong:

c:\Users\Oli\AppData\Local\Programs\Julia-1.7.2\bin>"c:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Tools\MSVC\14.29.30133\bin\Hostx86\x64\editbin.exe" /STACK:2000000000 julia.exe

did it:

julia> n = 2000

julia> visited = fill(false, (n,n));

julia> @time visit_four_neighbours(1,1)
  4.018976 seconds (53.59 M allocations: 817.735 MiB, 62.21% gc time)

Well, I hate it.
If you do it in a well designed loop you will get increased performance by huge orders of magnitudes!!!

1 Like

Someone can get the bounty there. I don’t like my answer, so I refuse to answer at SO :wink: