Segfault using PythonCall on Apple M1 with Julia 1.11.0-rc4 #563
Labels
bug
Something isn't working
Comments
|
I have a very similar issue with the final julia 1.11.0. The segfault is very non-deterministic. It doesn't always happen and when it happens it doesn't always happen at the same point. Some backtraces look like this: Others like this: Running the same code with julia 1.10.4 doesn't have this issue. |
|
I also get seg faults, on linux, only on multithread, in applications which worked with julia 1.10 |
|
Fwiw, here's a somewhat self-contained example that reliably crashes for me within a handful of runs (but it's non-deterministic), see this repository. Its only dependencies are import QuadGK
using PythonCall
const camb = pyimport("camb")
const Pm = let
params = camb.set_params(H0=70)
params.set_matter_power(redshifts=0:.1:1.5, kmax=9.5)
res = camb.get_results(params)
power = res.get_matter_power_interpolator(nonlinear=false, k_hunit=false, hubble_units=false, extrap_kmax=1e6).P
(z, k) -> let
pyval = power(z, k)
pyconvert(Float64, pyval)
end
end
calculate_stuff = (R, z) -> let
θ = R/900.
integrand = lθ -> let
kl = lθ/((1+z) * R)
sqrt(lθ)*cos(lθ)*Pm(z, kl) / (θ*θ)
end
QuadGK.quadgk(integrand, 0, 50)[1]
end
zvalues = [0.05, 0.1, 0.15, 0.2, 0.25, 0.3, 0.35, 0.4, 0.45, 0.5, 0.55, 0.6, 0.65, 0.7]
Rvalues = collect(.1:.2:7.0)
@info @. calculate_stuff(Rvalues, zvalues') |
|
I get a segmentation error on julia v1.11.2 with another example. This is my system: My example involves just loading HuggingFaceDatasets.jl (which uses PythonCall) from another package. It works fine if just used in a script instead. The error happens at precompilation time. julia> activate .
julia> using GNNDatasets
Precompiling GNNDatasets...
Info Given GNNDatasets was explicitly requested, output will be shown live
[68039] signal 11 (2): Segmentation fault: 11
in expression starting at none:0
_ZN3Aws4Http19CurlHandleContainerD2Ev at /Users/carlo/.julia/dev/GraphNeuralNetworks/GNNDatasets/.CondaPkg/env/lib/libaws-cpp-sdk-core.dylib (unknown line)s
Allocations: 3387461 (Pool: 3387279; Big: 182); GC: 9
✗ GNNDatasets
0 dependencies successfully precompiled in 7 seconds. 171 already precompiled.
ERROR: The following 1 direct dependency failed to precompile:
GNNDatasets
Failed to precompile GNNDatasets [2afd1aaa-255b-4d45-920d-62446192254f] to "/Users/carlo/.julia/compiled/v1.11/GNNDatasets/jl_R2rNZG".
[68039] signal 11 (2): Segmentation fault: 11
in expression starting at none:0
_ZN3Aws4Http19CurlHandleContainerD2Ev at /Users/carlo/.julia/dev/GraphNeuralNetworks/GNNDatasets/.CondaPkg/env/lib/libaws-cpp-sdk-core.dylib (unknown line)
Allocations: 3387461 (Pool: 3387279; Big: 182); GC: I can provide a full reproducer if needed. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Affects: PythonCall
Describe the bug
I got a segfault using PythonCall on GitHub CI for Apple M1.
https://github.com/JuliaIO/ZipArchives.jl/actions/runs/11219970406/job/31187122812?pr=82
This seems to be an intermittent issue.
Your system
Additional context
Stack trace:
The text was updated successfully, but these errors were encountered: