Skip to content

item-bodies/item-types timings regressed for issue-32062-equality-relations #33889

Closed
Closed
item-bodies/item-types timings regressed for issue-32062-equality-relations#33889
Assignees
pnkfelix
Labels
A-type-systemArea: Type systemI-compiletimeIssue: Problems and improvements with respect to compile times.P-highHigh priority

Description

@nagisa
nagisa
opened on May 26, 2016

See this and this.

Activity

  1. nagisa
    added
    A-type-systemArea: Type system
    I-slowIssue: Problems and improvements with respect to performance of generated code.
    on May 26, 2016

  2. nagisa commented on May 26, 2016

    @nagisa
    nagisa
    on May 26, 2016
    MemberAuthor

    cc @eddyb @nikomatsakis

  3. jonas-schievink commented on May 26, 2016

    @jonas-schievink
    jonas-schievink
    on May 26, 2016
    Contributor

    cc @Marwes

  4. Marwes commented on May 27, 2016

    @Marwes
    Marwes
    on May 27, 2016
    Contributor

    While testing #33816 on my projects I figured I'd run through this example on that version of rustc as well and it seems it doesn't seem to give this slowdown and given that #33816 seems to be rebased on top of master after this slowdown occurred it might have been fixed by that PR (if only by chance?). I may take a look at this sometime this weekend if I get the time though as it still seems unfortunate. (It would be really nice if perf.rust-lang.org could display the commit hash of each run it does. That would make it much easier to pin down which version introduced a regression).

  5. nikomatsakis commented on Jun 2, 2016

    @nikomatsakis
    nikomatsakis
    on Jun 2, 2016
    Contributor

    Probably the same problem is causing a regr in jld-day15-parser

  6. nikomatsakis commented on Jun 2, 2016

    @nikomatsakis
    nikomatsakis
    on Jun 2, 2016
    Contributor

    triage: P-high

  7. rust-highfive
    added
    P-highHigh priority
    on Jun 2, 2016

  8. nikomatsakis commented on Jun 2, 2016

    @nikomatsakis
    nikomatsakis
    on Jun 2, 2016
    Contributor

    It'd be great to do some sort of bisection to try to narrow down what is causing this.

  9. pnkfelix
    self-assigned this
    on Jun 2, 2016
  10. sanxiyn
    added
    I-compiletimeIssue: Problems and improvements with respect to compile times.
    and removed
    I-slowIssue: Problems and improvements with respect to performance of generated code.
    on Jun 3, 2016

  11. eddyb commented on Jun 4, 2016

    @eddyb
    eddyb
    on Jun 4, 2016
    Contributor

    @nikomatsakis Doesn't look like it, the regression is in translation for jld-day15-parser.

  12. arielb1 commented on Jun 4, 2016

    @arielb1
    arielb1
    on Jun 4, 2016
    Contributor

    The problematic PR here is 75e23e1 (#33137). Seems logical - that probably broke the caching

  13. eddyb commented on Jun 4, 2016

    @eddyb
    eddyb
    on Jun 4, 2016
    Contributor

    @arielb1 So the solution is to move forward with #33816?

  14. arielb1 commented on Jun 4, 2016

    @arielb1
    arielb1
    on Jun 4, 2016
    Contributor

    That sounds like it.

  15. arielb1 commented on Jun 5, 2016

    @arielb1
    arielb1
    on Jun 5, 2016
    Contributor

    This is fixed as #33816 landed.

  16. arielb1
    closed this as completedon Jun 5, 2016
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

A-type-systemArea: Type systemI-compiletimeIssue: Problems and improvements with respect to compile times.P-highHigh priority

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions