-
Notifications
You must be signed in to change notification settings - Fork 16
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
[ocaml5-issue] Out_channel Lin test takes very long on macOS #321
Comments
I had previously observed a problem shrinking |
Also spotted on macOS 5.0.0
This has a jump of nearly 9000s on testing 29 inputs (from 61 to 90) |
I tried giving ocaml/ocaml#11903 a spin to see whether it would solve this issue.
Notice the jump between input 348 and 349, taking 2000 seconds... |
There is definitely something wrong in there. Trying to look into a specific case of this Out_channel test entering seemingly a livelock in the debug runtime, I finally ran into:
(on Linux, OCaml 5.0.0; the same seed seems to get stuck (one way or another) regularly also on trunk) |
Just observed a similar behaviour on Windows bytecode
The next to last shrinking step (from 1.0002 to 1.0003) spends 3972s on 1 successful shrink step (there may be several unsuccessful ones though) - that's 66 minutes - over an hour! |
Closing by replacing the Lin test with an STM one in #431 |
As part of #318 I spotted an
Out_channel
test taking very long on macOS withtrunk
to the point of causing a timeout:https://github.com/ocaml-multicore/multicoretests/actions/runs/4540532958/jobs/8001562877?pr=318
That's 2h40m - to test 585 random triples (modulo some missing output flushing) without any shrinking going on.
Of those, 80 inputs took ~100m (~6000s) from input 487 to 567. This looks like a runtime issue, doesn't it?
The text was updated successfully, but these errors were encountered: