Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Investigate use of PkgLock.jl to avoid race conditions on module precompilation with MPI #214

Closed
sloede opened this issue Oct 11, 2020 · 4 comments · Fixed by #438
Closed
Labels
parallelization Related to MPI, threading, tasks etc.

Comments

@sloede
Copy link
Member

sloede commented Oct 11, 2020

You could also use PkgLock.jl to precompile all packages without race conditions and use them afterwards in parallel.

Xref: https://discourse.julialang.org/t/precompilation-error-using-hpc/17094

Originally posted by @ranocha in #167 (comment)

We should check if and how it makes sense to include it in our testing environment, and maybe document it for use on clusters etc.

@sloede sloede added parallelization Related to MPI, threading, tasks etc. performance We are greedy and removed performance We are greedy labels Oct 11, 2020
@ranocha
Copy link
Member

ranocha commented Nov 2, 2020

Maybe we can also just sit this out: JuliaLang/julia#36416

@ranocha
Copy link
Member

ranocha commented Nov 8, 2020

I've tested it on the current Julia nightly: We won't need the flag --compiled-modules=no on Julia v1.6 anymore 🎉

@sloede
Copy link
Member Author

sloede commented Nov 8, 2020

Excellent news! Do you know if there's a tentative release date for 1.6?

@ranocha
Copy link
Member

ranocha commented Nov 8, 2020

The soft feature freeze is today: https://discourse.julialang.org/t/1-6-soft-feature-freeze-nov-8th/49379. Then, we will have the usual release process (finishing works, alpha, beta, iterate RCs until convergence) I guess. Usually, there isn't a tentative schedule for this last part of the process.

@ranocha ranocha mentioned this issue Feb 7, 2021
5 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
parallelization Related to MPI, threading, tasks etc.
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants