I tried this:
src/test/ui/fn/issue-80844-no-gate.rs
fn returns_generator() -> _ {
//~^ ERROR the type placeholder `_` is not allowed within types on item signatures [E0121]
//~| NOTE not allowed in type signatures
|| yield 0
//~^ ERROR yield syntax is experimental
- //~^ NOTE see issue
+ //~| NOTE see issue
}$ ./x.py test src/test/ui/fn
I expected to see this happen: the ui/fn test suite is run, possibly building the testsuite binary using the existing stage 1 rustc.
Instead, this happened: the stage 1 rustc is rebuilt before running the test suite.
This seems to happen randomly; I had rerun the test suite quite a few times using the same command with no issues. This is rather inconvenient since even with 16 logical cores, building rustc takes several minutes and grinds my machine to a halt.
This doesn't just seem to happen when running the test suite, but I'm less confident about saying that rebuilds after modifying rustc's codebase are spurious.
@rustbot modify labels +A-contributor-roadblock +A-rustbuild
I tried this:
src/test/ui/fn/issue-80844-no-gate.rs
fn returns_generator() -> _ { //~^ ERROR the type placeholder `_` is not allowed within types on item signatures [E0121] //~| NOTE not allowed in type signatures || yield 0 //~^ ERROR yield syntax is experimental - //~^ NOTE see issue + //~| NOTE see issue }$ ./x.py test src/test/ui/fnI expected to see this happen: the
ui/fntest suite is run, possibly building the testsuite binary using the existing stage 1rustc.Instead, this happened: the stage 1
rustcis rebuilt before running the test suite.This seems to happen randomly; I had rerun the test suite quite a few times using the same command with no issues. This is rather inconvenient since even with 16 logical cores, building
rustctakes several minutes and grinds my machine to a halt.This doesn't just seem to happen when running the test suite, but I'm less confident about saying that rebuilds after modifying rustc's codebase are spurious.
@rustbot modify labels +A-contributor-roadblock +A-rustbuild