-
-
Couldn't load subscription status.
- Fork 243
Description
Description
Iβm hitting a precompilation failure due to a solve method overwrite between SciMLBase and DiffEqBase. This occurs even in a fully isolated session (fresh depot, temp project, no startup file), both when adding DifferentialEquations and when adding OrdinaryDiffEq directly. I had not had the issue before today.
Minimal Reproducible Example π
# Terminal (macOS). This guarantees no leakage from ~/.julia and no startup file.
JULIA_DEPOT_PATH="$(mktemp -d)" JULIA_LOAD_PATH=@ JULIA_PKG_SERVER="" \
julia --startup-file=no -q] activate --temp
] registry add General
] add DifferentialEquations
using DifferentialEquations # <-- Also triggers the same overwrite during precompile
Error & Stacktrace
first few lines:
[ Info: Precompiling DifferentialEquations [...]
WARNING: Method definition solve(SciMLBase.EnsembleProblem{T, T2, T3, T4, T5} where T5 where T4 where T3 where T2 where T, Any...) in module SciMLBase at ~/.julia/packages/SciMLBase/.../basic_ensemble_solve.jl:356
overwritten in module DiffEqBase at ~/.julia/packages/DiffEqBase/.../solve.jl:1224.
ERROR: Method overwriting is not permitted during Module precompilation. Use `__precompile__(false)` to opt-out of precompilation.it does this also while precompile DiffEqBase. then when precompiling DiffEqNoiseProcess,
Info Given DiffEqNoiseProcess was explicitly requested, output will be shown live
β Warning: Module DiffEqBase with build ID ffffffff-ffff-ffff-fc27-e6e18de9fdec is missing from the cache.eDistributionsExt
β This may mean DiffEqBase [2b5f629d-d688-5b77-993f-72d75c75574e] does not support precompilation but is imported by a module that does.
β @ Base loading.jl:2541
? DiffEqBase
? DiffEqBase β DiffEqBaseForwardDiffExt
? DiffEqBase β DiffEqBaseDistributionsExt
? DiffEqBase β DiffEqBaseSparseArraysExt
? DiffEqNoiseProcess
[ Info: Precompiling DiffEqNoiseProcess [77a26b50-5914-5dd7-bc55-306e6241c503]
β Warning: Module DiffEqBase with build ID ffffffff-ffff-ffff-fc27-e6e18de9fdec is missing from the cache.
β This may mean DiffEqBase [2b5f629d-d688-5b77-993f-72d75c75574e] does not support precompilation but is imported by a module that does.
β @ Base loading.jl:2541
β Info: Skipping precompilation due to precompilable error. Importing DiffEqNoiseProcess [77a26b50-5914-5dd7-bc55-306e6241c503].
β exception = Error when precompiling module, potentially caused by a __precompile__(false) declaration in the module.
It goes on to do this warning for many modules. (I'm going to attach the full error in a txt file)
- Output of
using Pkg; Pkg.status()
Status `<name of folder this is stored in >/Project.toml`
[0c46a032] DifferentialEquations v7.16.1
[b964fa9f] LaTeXStrings v1.4.0
[442fdcdd] Measures v0.3.2
[91a5bcdd] Plots v1.40.19
[f27b6e38] Polynomials v4.1.0
[295af30f] Revise v3.9.0
[3a884ed6] UnPack v1.0.2
[37e2e46d] LinearAlgebra v1.11.0
- Output of
versioninfo()
Julia Version 1.11.6
Commit 9615af0f269 (2025-07-09 12:58 UTC)
Build Info:
Official https://julialang.org/ release
Platform Info:
OS: macOS (arm64-apple-darwin24.0.0)
CPU: 12 Γ Apple M3 Pro
WORD_SIZE: 64
LLVM: libLLVM-16.0.6 (ORCJIT, apple-m2)
Threads: 1 default, 0 interactive, 1 GC (on 6 virtual cores)Additional context
the full error output is in the attached txt document