-
Notifications
You must be signed in to change notification settings - Fork 13.3k
interpret: fix vtable check debug assertion #99607
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Conversation
r? @cjgillot (rust-highfive has picked a reviewer for you, use r? to override) |
Some changes occurred to the CTFE / Miri engine cc @rust-lang/miri |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM/r=me - only one thing I wanted to ask before then though, is this testable on this side, or only in the miri repo?
For the record, I believe a minimal test would look something like this:
(Box::new(|| {}) as Box<dyn FnOnce()>)()
(just checked and there is indeed a ::{shim:vtable#0}
generated for such code)
Yeah it can be tested in a rustc checkout (with debug assertions enabled) via |
@bors r=eddyb |
Sorry, I meant whether it's possible to add a test in |
Ah I see. Yeah I don't think CTFE can reach this... well, it could with @bors r- |
No that won't work, it still dynamically checks that we only call @bors r=eddyb |
Rollup of 3 pull requests Successful merges: - rust-lang#99588 (Update books) - rust-lang#99602 (cargotest: do not run quickcheck tests in xsv) - rust-lang#99607 (interpret: fix vtable check debug assertion) Failed merges: r? `@ghost` `@rustbot` modify labels: rollup
Fixes #99605
Thanks to @eddyb for suggesting the fix!