search

Clemapfel / jluna

Julia Wrapper for C++ with Focus on Safety, Elegance, and Ease of Use
https://clemens-cords.com/jluna
MIT License
252 stars 13 forks source link

Thread adoption #45

Closed Taaitaaiger closed 1 year ago

Taaitaaiger commented 1 year ago

Hi!

I've got to admit I'm not a user of JLuna, but rather someone who's working on a similar library. Considering you've put a lot of effort in mutithreading support, I think you might also be interested in thread adoption. This is a feature that will be available in Julia 1.9 that can be used to create a pool of threads that can call into Julia.

Feel free to ask any questions you have, or close this issue if you're not interested :)

Clemapfel commented 1 year ago

Thank you, it is nice to find a fellow independent-julia-wrapper-library creator.

I was aware of this feature the moment the PR for it was opened on the Julia GitHub, just a few days ago (https://github.com/Clemapfel/jluna/issues/42) I was informed how the crash is related to thread local storage, which is interesting. I might update the docs with that context.

I'll be deprecating the jluna multi-threading modules once 1.9 becomes stable, however, that will be many months in the future, I'm sure.

Thank you for thinking of informing me, I feel like it's possible to feel the undercurrent of frustration in how I wrote the multi-threading docs, mostly because I spend 3 days trying to reverse engineer the crash thinking it was a bug. I'm actually still kinda upset about that but what's done is done.

I'll be closing this issue as a duplicate of https://github.com/Clemapfel/jluna/issues/26 and https://github.com/Clemapfel/jluna/issues/42

Taaitaaiger commented 1 year ago

Ah, I see I should have checked the closed issues before I opened this, sorry for the noise.

The first RC for 1.9 seems to be around the corner, given that a pr was opened for it, so let's hope it's a few weeks rather than several months 😉

Op ma 20 feb. 2023 03:11 schreef Clem Cords @.***>:

Thank you, I was aware of this feature the moment the PR for it was opened on the Julia GitHub, just a few days ago (#42 https://github.com/Clemapfel/jluna/issues/42) I was informed how the crash is related to thread local storage, which is interesting. I might update the docs with that context.

I'll be deprecating the jluna multi-threading modules once 1.9 becomes stable, however, that will be many months in the future, I'm sure.

Thank you for thinking of informing me, I feel like it's possible to feel the undercurrent of frustration in how I wrote the multi-threading docs, mostly because I spend 3 days trying to reverse engineer the crash thinking it was a bug. I'm actually still kinda upset about that but what's done is done.

I'll be closing this issue as a duplicate of #26 https://github.com/Clemapfel/jluna/issues/26 and #42 https://github.com/Clemapfel/jluna/issues/42

— Reply to this email directly, view it on GitHub https://github.com/Clemapfel/jluna/issues/45#issuecomment-1436200438, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOQBHHYAMT2CAR3VUS7BL5DWYLHF7ANCNFSM6AAAAAAVBDMYZE . You are receiving this because you authored the thread.Message ID: @.***>