diff options
| author | Matthias Krüger <matthias.krueger@famsik.de> | 2023-03-03 20:06:28 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2023-03-03 20:06:28 +0100 |
| commit | db4b3dd608e342bc6b3988035e30e871c69c0a39 (patch) | |
| tree | 4ed044aae9ff8ef2f66d3765b782d14ce5586828 /compiler/rustc_mir_transform/src/coverage/debug.rs | |
| parent | 1fab0fc4a2ce8fd4b98cddb7af59e190f5a198f4 (diff) | |
| parent | 4b01a1a07a4cd1a1dee16ffae384905989f1a60a (diff) | |
| download | rust-db4b3dd608e342bc6b3988035e30e871c69c0a39.tar.gz rust-db4b3dd608e342bc6b3988035e30e871c69c0a39.zip | |
Rollup merge of #108667 - compiler-errors:issue-108664, r=estebank
Fix another ICE in `point_at_expr_source_of_inferred_type` Types coming from method probes must only be investigated *structurally*, since they often contain escaping infer variables from generalization and autoderef. We already have a hack in this PR that erases variables from types, so just use that. Fixes #108664 The note attached to this error is pretty bad: ``` here the type of `primes` is inferred to be `[_]` ``` But that's unrelated to the PR. --- Side-note: This is a pretty easy to trigger beta regression, so I've nominated it. Alternatively, I'm slightly inclined to remove this code altogether until it can be reformulated to be more accurate and less ICEy.
Diffstat (limited to 'compiler/rustc_mir_transform/src/coverage/debug.rs')
0 files changed, 0 insertions, 0 deletions