Store VariantIdx to distinguish enum variants

[Mark-Simulacrum]

This saves ~24% of the instructions on the match-stress-enum benchmark, but I'm not 100% sure that this is OK - if we ever compare two constructors across enums (e.g., a Result and an Option), then this is obviously insufficient; I can experiment with continuing to store the DefId for comparison purposes in that case.

[rust-highfive]

r? @petrochenkov

(rust-highfive has picked a reviewer for you, use r? to override)

[Mark-Simulacrum]

@bors try @rust-timer queue

The only PartialEq user is here, but it's not quite clear if that compares across enums ever.

error[E0369]: binary operation `==` cannot be applied to type `&Constructor<'_>`
   --> compiler/rustc_mir_build/src/thir/pattern/deconstruct_pat.rs:821:55
    |
821 |             Variant(_) => used_ctors.iter().any(|c| c == self),
    |                                                     - ^^ ---- &Constructor<'_>
    |                                                     |
    |                                                     &Constructor<'_>
    |
    = note: an implementation of `std::cmp::PartialEq` might be missing for `&Constructor<'_>`

[rust-timer]

Awaiting bors try build completion.

@rustbot label: +S-waiting-on-perf

[bors]

⌛ Trying commit e400595 with merge 8dc9aa2668caece29926c3444fa73fdeda459f27...

[bors]

☀️ Try build successful - checks-actions
Build commit: 8dc9aa2668caece29926c3444fa73fdeda459f27 (8dc9aa2668caece29926c3444fa73fdeda459f27)

[rust-timer]

Queued 8dc9aa2668caece29926c3444fa73fdeda459f27 with parent ba8d7e2, future comparison URL.

[rust-timer]

Finished benchmarking try commit (8dc9aa2668caece29926c3444fa73fdeda459f27): comparison url.

Benchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. Please note that if the perf results are neutral, you should likely undo the rollup=never given below by specifying rollup- to bors.

Importantly, though, if the results of this run are non-neutral do not roll this PR up -- it will mask other regressions or improvements in the roll up.

@bors rollup=never
@rustbot label: +S-waiting-on-review -S-waiting-on-perf

[Mark-Simulacrum]

Cc @Nadrieril

[Nadrieril]

That's fine, typechecking ensures we only ever compare Constructors for the same type. The rest of the code relies on this in various places already.
LGTM

[petrochenkov]

@bors r+

[bors]

📌 Commit e400595 has been approved by petrochenkov

[bors]

⌛ Testing commit e400595 with merge 754d171...

[bors]

☀️ Test successful - checks-actions
Approved by: petrochenkov
Pushing 754d171 to master...