-
-
Notifications
You must be signed in to change notification settings - Fork 197
Fix hanging system image creation if multiple threads are set #1042
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
Fix hanging system image creation if multiple threads are set #1042
Conversation
Make sure, that the final system image is built single-threaded and override any values set by "-t", "--threads" in sysimage_build_args` or provided via JULIA_NUM_THREADS. This is needed until the underlying bug is fixed (see JuliaLang#963 and especially JuliaLang#990 containing a `git bisect` to the commit introducing the problem) Fixes JuliaLang#963 and JuliaLang#990
I have a wish: |
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## master #1042 +/- ##
=======================================
Coverage 86.16% 86.16%
=======================================
Files 3 3
Lines 795 795
=======================================
Hits 685 685
Misses 110 110 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Thanks. I can confirm it fixes the problem, see also my post in Discourse |
I also tested this with my multithreaded code and it worked |
Co-authored-by: Dilum Aluthge <[email protected]>
I'm working on releasing 2.2.1, which will include this. |
Make sure, that the final system image is built single-threaded and override any values set by "-t", "--threads" in sysimage_build_args
or provided via JULIA_NUM_THREADS. This is needed until the underlying bug is fixed (see #963 and especially #990 containing a
git bisect` to the commit introducing the problem)References #963 and #990 (but does not fix the underlying bug)
Supersedes #1041 (closes #1041).
Thanks a lot to @KristofferC for the close hit with #1041 limiting down the search area enormously and also to everyone reporting or contributing to the discussions around #963 and #990, especially to @dpinol, @juanromerohb, @matthewgcooper and last, but certainly not least for a lot of patience with
git bisect
, to @paulmelis.