Add projection obligations when comparing impl too #115039
Conversation
|
r? @davidtwco (rustbot has picked a reviewer for you, use r? to override) |
rustbot
added
S-waiting-on-review
Ideally I would prefer to use the unnormalized obligations but unfortunately there is #109628, which allows normalized obligations to have more implied bounds than the unnormalized ones. |
|
bors failed to recognize command? |
bors
added
S-waiting-on-bors
|
I think this just missed the version bump, so going to preemptively nominate for beta backport. |
jackh726
added
the
beta-nominated
|
☀️ Test successful - checks-actions |
|
Finished benchmarking commit (fe5f591): comparison URL. Overall result: no relevant changes - no action needed@rustbot label: -perf-regression Instruction countThis benchmark run did not return any relevant results for this metric. Max RSS (memory usage)This benchmark run did not return any relevant results for this metric. CyclesThis benchmark run did not return any relevant results for this metric. Binary sizeThis benchmark run did not return any relevant results for this metric. Bootstrap: 636.268s -> 633.645s (-0.41%) |
|
Fixes #115051 |
rustbot
added
the
beta-accepted
cuviper
removed
the
beta-nominated
[beta] backport and stage0 bump - Bump to released stable compiler - Add projection obligations when comparing impl too rust-lang#115039 r? cuviper
Fixes #115033
In the test, when we ask for WF obligations of
DatasetIter<'a, ArrayBase<D>>, we get back two important obligations:[<D as Data>::Elem -> ?1, ?1: 'a]. If we don't add the projection obligation,?1remains unconstrained.An alternative solution would be to use unnormalized obligations, where we only have one relevant obligation:
<D as Data>::Elem: 'a. This would leave no inference vars unconstrained.