A couple different related issues here.

In tests like src/test/ui/regions/region-lifetime-bounds-on-fns-where-clause.rs, the errors for mismatched types are actually duplicates (exactly the same error on the same line for the same span and the same notes.

Other similar tests:

• src/test/ui/hr-subtype/hr-subtype.rs
• src/test/ui/higher-rank-trait-bounds/normalize-under-binder/issue-71955.rs

In src/test/ui/rfc1623.rs, the errors are very similar, but not exactly duplicates.

@rustbot label +A-NLL +NLL-diagnostics