Don't ICE for kind mismatches during error rendering #123673
Conversation
rustbot
added
A-testsuite
|
Some changes occurred in src/tools/compiletest cc @jieyouxu |
| @@ -0,0 +1,23 @@ | |||
| //! THis test used to ICE in typeck because of the type/const mismatch, | |||
There was a problem hiding this comment.
| //! THis test used to ICE in typeck because of the type/const mismatch, | |
| //! This test used to ICE in typeck because of the type/const mismatch, |
| @@ -0,0 +1,23 @@ | |||
| //! THis test used to ICE in typeck because of the type/const mismatch, | |||
| //! even though wfcheck already errored. | |||
There was a problem hiding this comment.
Please add an issue number for ICE test:
| //! even though wfcheck already errored. | |
| //! even though wfcheck already errored. | |
| //! issue: rust-lang/rust#123457 |
There was a problem hiding this comment.
As nice as it is to have the issue in the test itself (and helps with traceability in the future even when files move around), I find that 99% of the time I'm able to get to the ticket by finding the PR that introduced the file. It's good practice to add as much info as possible inline, but we should avoid holding up a PR when that's the only comment to address (some new contributors will just not have the bandwidth to address comments and might get put off by such a request, where accepting their PRs directly increases the likelihood of them coming back).
I'm wondering if we could try and automate this, make .x,py tidy check for test files that are introduced in the current branch that aren't in master, check against github to see if there's an open PR for the current branch, and if that PR says Fixes #XXXXXX. If so, complain about the lack of a link :)
There was a problem hiding this comment.
Yeah, probably a good idea to make this somewhat automated
|
☔ The latest upstream changes (presumably #123676) made this pull request unmergeable. Please resolve the merge conflicts. |
If you want to override an env var, don't unset it, just set it
oli-obk
force-pushed
the
sig_wfcheck_ice
branch
from
dc70e28
to
0a88339
Compare
|
@bors r=jieyouxu,estebank |
bors
added
S-waiting-on-bors
…llaumeGomez Rollup of 7 pull requests Successful merges: - rust-lang#123673 (Don't ICE for kind mismatches during error rendering) - rust-lang#123675 (Taint const qualifs if a static is referenced that didn't pass wfcheck) - rust-lang#123975 (Port the 2 `rust-lld` run-make tests to `rmake`) - rust-lang#124000 (Use `/* value */` as a placeholder) - rust-lang#124013 (Box::into_raw: make Miri understand that this is a box-to-raw cast) - rust-lang#124027 (Prefer identity equality over equating types during coercion.) - rust-lang#124036 (Remove `default_hidden_visibility: false` from wasm targets) r? `@ghost` `@rustbot` modify labels: rollup
Rollup merge of rust-lang#123673 - oli-obk:sig_wfcheck_ice, r=jieyouxu,estebank Don't ICE for kind mismatches during error rendering fixes rust-lang#123457 also some test suite cleanups to make backtraces easier to read
fixes #123457
also some test suite cleanups to make backtraces easier to read