Disable download-rustc for library profile #143577
Conversation
The feature currently completely breaks `x test`, core functionality of working on the standard library. Therefore it should be disabled by default until that problem is fixed. Having to wait a bit longer for a check build is nothing compared to completely mysterious build errors when testing.
|
r? @clubby789 rustbot has assigned @clubby789. Use |
rustbot
added
S-waiting-on-review
|
This PR modifies If appropriate, please update |
|
You made a PR to document that this doesn't work (rust-lang/rustc-dev-guide#2482), which confused me a bit. if it doesn't work, why not just turn it off? |
|
Can you also include a warn-level change tracker? Yes, I think we should disable it for now. |
I think we generally expected (or rather, hoped) that the cleanup after the stage0 redesign will go faster, but it's such a deep rabbit hole of subtle invariants and nested layers of scary hotfixes that at this point I think it's fine to just say that we give up and disable download-ci-rustc for the library profile, yeah. |
|
@Noratrieb said they were on their phone, so I pushed the change tracker entry. |
|
Sent a heads-up in #t-libs > Temporarily disabling download-rustc for library profile. |
|
Thank you. @bors r+ rollup |
bors
added
S-waiting-on-bors
|
@bors p=1 (contributor roadblock; but feel free to include in rollups) |
Rollup of 11 pull requests Successful merges: - #143130 (doc(std): clarify `NonZero<T>` usage limitation in doc comment) - #143415 (Get rid of build-powerpc64le-toolchain.sh) - #143464 (Make tests/ui/abi/debug.rs cross-compile) - #143482 (Fix short linker error output) - #143524 (Move `stable_mir` back to its own crate) - #143528 (interpret: rename StackPopCleanup) - #143551 (Dont resolve instance of root in `mir_callgraph_cyclic`) - #143558 (mbe: Refactors and function extractions in `compile_declarative_macro`) - #143563 (std: fix typo in `std::path`) - #143564 (compiler: Deduplicate `must_emit_unwind_tables()` comments) - #143577 (Disable download-rustc for library profile) r? `@ghost` `@rustbot` modify labels: rollup
Rollup merge of #143577 - Noratrieb:Noratrieb-patch-4, r=Kobzol Disable download-rustc for library profile The feature currently completely breaks `x test` (#142505), core functionality of working on the standard library. Therefore it should be disabled by default until that problem is fixed. Having to wait a bit longer for a check build is nothing compared to completely mysterious build errors when testing.
|
bors what |
|
it likely got confused by #143577 (comment) |
Rollup of 11 pull requests Successful merges: - rust-lang/rust#143130 (doc(std): clarify `NonZero<T>` usage limitation in doc comment) - rust-lang/rust#143415 (Get rid of build-powerpc64le-toolchain.sh) - rust-lang/rust#143464 (Make tests/ui/abi/debug.rs cross-compile) - rust-lang/rust#143482 (Fix short linker error output) - rust-lang/rust#143524 (Move `stable_mir` back to its own crate) - rust-lang/rust#143528 (interpret: rename StackPopCleanup) - rust-lang/rust#143551 (Dont resolve instance of root in `mir_callgraph_cyclic`) - rust-lang/rust#143558 (mbe: Refactors and function extractions in `compile_declarative_macro`) - rust-lang/rust#143563 (std: fix typo in `std::path`) - rust-lang/rust#143564 (compiler: Deduplicate `must_emit_unwind_tables()` comments) - rust-lang/rust#143577 (Disable download-rustc for library profile) r? `@ghost` `@rustbot` modify labels: rollup
The feature currently completely breaks
x test(#142505), core functionality of working on the standard library. Therefore it should be disabled by default until that problem is fixed. Having to wait a bit longer for a check build is nothing compared to completely mysterious build errors when testing.