search

SciML / SciMLBase.jl

The Base interface of the SciML ecosystem
https://docs.sciml.ai/SciMLBase/stable
MIT License
127 stars 92 forks source link

PythonCall extension hangs precompilation #528

Closed bgroenks96 closed 10 months ago

bgroenks96 commented 10 months ago

The PythonCall extension for this package seems to hang the precompilation process indefinitely (or at least for a very long time).

Could precompilation be temporarily disabled until this issue resolved?

(test) pkg> st PythonCall
Status ...
  [6099a3de] PythonCall v0.9.14

(test) pkg> st SciMLBase
Status ...
  [0bca4576] SciMLBase v2.4.3
julia> versioninfo()
Julia Version 1.9.3
Commit bed2cd540a1 (2023-08-24 14:43 UTC)
Build Info:
  Official https://julialang.org/ release
Platform Info:
  OS: Linux (x86_64-linux-gnu)
  CPU: 96 × Intel(R) Xeon(R) Platinum 8260 CPU @ 2.40GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-14.0.6 (ORCJIT, cascadelake)
  Threads: 96 on 96 virtual cores
Environment:
  JULIA_EDITOR = code
  JULIA_IMAGE_THREADS = 1
ChrisRackauckas commented 10 months ago

Is this a standard Julia installation? I cannot seem to recreate this on any standard OS.

bgroenks96 commented 10 months ago

Yes, it's a standard julia installation, but when I created an isolated environment with just SciMLBase and PythonCall it didn't happen anymore. So maybe it's an interaction with another package or with the CondaPkg.toml environment?

bgroenks96 commented 10 months ago

Ok, the issue is resolved after purging .CondaPkg. My guess is maybe the issue was a lingering lock file or something. The SciMLBase extension was probably just coincidentally the first code to invoke pyimport. Since this is more of a general issue with precompilation and CondaPkg, I will close this for now.