Sort out body-less function argument names wrt incremental recompilation. #38501
Labels
A-incr-comp
Area: Incremental compilation
C-cleanup
Category: PRs that clean code up or issues documenting cleanup.
T-compiler
Relevant to the compiler team, which will review and decide on the PR/issue.
Comments
steveklabnik
added
A-compiler
T-compiler
Mark-Simulacrum
added
C-enhancement
|
Triage: no idea what the status of this ticket is, honestly. |
|
Incremental queries made this sort of thing a non-problem a long time ago, I think. |
Aaron1011
added a commit
to Aaron1011/rust
that referenced
this issue
Aug 16, 2020
The call to `with_ignore` was added years ago, and should no longer be necessary with the modern incremental query infrastructure. I also removed a related FIXME comment for issue rust-lang#38501, which was closed months ago.
bors
added a commit
to rust-lang-ci/rust
that referenced
this issue
Aug 18, 2020
…eddyb Don't use `dep_graph.with_ignore` when encoding fn param names The call to `with_ignore` was added years ago, and should no longer be necessary with the modern incremental query infrastructure. I also removed a related FIXME comment for issue rust-lang#38501, which was closed months ago.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
The
xandyintrait Foo { fn bar(x: T, y: U); }, for example, are stored outside of any isolation, as such function declarations don't have bodies holding the arguments (see #38449), for this purpose.Right now they're primarily used by
rustdoc, which even loads them across crates. #38449 asserts that the cross-crate usage always happens without incremental recompilation and doesn't track theMetaDataedge for them in neither the original crate or the one observing these "argument names".OTOH, the same-crate usage is virtually impossible to prevent, accidentally, which leads to the implementation in #38449 hashing "argument names" as part of the definition's signature, for now.
Also, ideally, tools like
rustdocwould have their own infrastructure (which save-analysis aims to be), potentially quite different (divergent even?) fromrustc's HIR, but that will take some effort to reach.The text was updated successfully, but these errors were encountered: