-
Notifications
You must be signed in to change notification settings - Fork 12.8k
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
Add regression test for #89436 #99257
Conversation
(rust-highfive has picked a reviewer for you, use r? to override) |
You could make it a dependency of But more seriously, how silly does adding it as a perf benchmark sound? |
@bors r+ rollup |
Rollup of 6 pull requests Successful merges: - rust-lang#99113 (Simplify [a]rc code a little) - rust-lang#99131 (Add label for generic arg (+ APIT) and RPIT callables in `label_fn_like`) - rust-lang#99237 (removed unused CSS and unused HTML IDs) - rust-lang#99239 (Add myself to the set of people notified when MIR changes.) - rust-lang#99241 (Remove comment referring to constness.rs) - rust-lang#99257 (Add regression test for rust-lang#89436) Failed merges: r? `@ghost` `@rustbot` modify labels: rollup
@eddyb seems a bit silly but i'm open to it if the rust team is okay with perf benchmarks that are Actually Compiler Regression Tests |
We have a number of crates in cargotest that we fetch a pinned commit of and run cargo test in; that seems like a better fit here? |
Yeah we could do that! I don't know if the yoke crate has full test coverage of the bugs, but I can make sure we do and go fix that. (One of the reasons I'm wary of leaving this up to crater is that a lot of these things only show up when using the crate, i.e when building tests, which crater does do, but folks do tend to look at test failures less) |
I never got around to adding such a test.
In general I think the
yoke
crate has a bunch of interesting testcases that exercise various edges of the algorithms here, it would be nice if we could simply depend on the crate and add some tests that exercise it, but I don't think that's possible. Do you or @eddyb think there's any use trying to upstream a bunch of common yoke minimal working example code to the testsuite and having a ton of yoke tests?