Skip to content

[FEATURE] Shrink getprime runs for OSTest #3620

Open

Description

@linguini1

Is your feature request related to a problem? Please describe.

OSTest is a huge pain in the ass to run on small targets with the getprime test enabled because it takes so long to complete that step.

Describe the solution you'd like

I don't think we need 10 runs of getprime in the test; maybe just 3 is fine?

Describe alternatives you've considered

Make the number of getprime runs configurable so it can be fewer on small targets.

Verification

  • I have verified before submitting the report.

Activity

  1. cederom commented on Jul 8, 2026

    @cederom
    Contributor

    Hmm, I guess this is to detect computational / memory / types related issues?

    Why 3 is better than 10? Is it faster or smaller?

  2. linguini1 commented on Jul 8, 2026

    @linguini1
    ContributorAuthor

    Hmm, I guess this is to detect computational / memory / types related issues?

    Why 3 is better than 10? Is it faster or smaller?

    Mostly just because it's faster. Running OSTest on the raspberrypi-pico gets stuck on getprime for about 5 minutes because of the number of runs. That seems excessive to me, I'm not sure we'd really catch any new memory errors that way? I could be wrong, but I feel like a smaller number might be enough for really weak targets?

    This isn't really a glaring issue but it slow down my testing of RC2 today haha!

  3. cederom commented on Jul 8, 2026

    @cederom
    Contributor

    Are you sure this is not some other rpi-pico issues like hanging ostest and/or usbnsh that I am also observing? Does it always take 5 minutes for you on these primes? :-)

  4. linguini1 commented on Jul 8, 2026

    @linguini1
    ContributorAuthor

    Are you sure this is not some other rpi-pico issues like hanging ostest and/or usbnsh that I am also observing? Does it always take 5 minutes for you on these primes? :-)

    This was over UART shell. It does take a lot of time yes, I should maybe time it precisely. I know it isn't hanging because after the really long duration, the rest of the test continues and succeeds.

  5. cederom commented on Jul 8, 2026

    @cederom
    Contributor

    Allright, for me rpi-pico* seems unreliable but I used it only over usbnsh :-)

  6. cederom commented on Oct 11, 2026

    @cederom
    Contributor

    Negative :-( Unfortunately #3809 that changed iterations count broke CI and probably some platforms. #3832 reverts this. Iterations needs to be decreased only for small platforms.. or some additional fix in the code is required in addition to only defines change.

  7. added a commit that references this issue on Oct 11, 2026
    27ecdd7
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions