NLL Liveness: Skip regionless types when visiting free regions #52037
Conversation
|
(rust_highfive has picked a reviewer for you, use r? to override) |
rust-highfive
added
the
S-waiting-on-review
|
@bors try |
|
⌛ Trying commit d84c9bc5bc629fe7b228ee1646c5fe6a9ded2db6 with merge 329e5d1d5d4c35180844125567dc932e9bf492fe... |
src/librustc/ty/fold.rs
Outdated
| fn visit_ty(&mut self, t: Ty<'tcx>) -> bool { | ||
| t.super_visit_with(self) | ||
| self.super_ty(t) |
There was a problem hiding this comment.
oh, my bad, I should have written super_visit_with I suppose. Let's revert this part of the diff and just have the code below invoke super_visit_with — although I do prefer (these days) to make matching super_ty methods and so forth, it doesn't seem to be the standard in TypeVisitor.
|
@bors try- Actually, we can probably just land this and see the perf afterwards. It can't be worse |
|
💥 Test timed out |
The tuple-stress benchmark exercises the liveness constraint generation code for types which do not have regions
lqd
force-pushed
the
skipping-regionless-types
branch
from
d84c9bc to
0190286
Compare
|
@bors r+ |
|
📌 Commit 0190286 has been approved by |
bors
added
S-waiting-on-bors
NLL Liveness: Skip regionless types when visiting free regions The tuple-stress benchmark exercises the liveness constraint generation code for types which do not have regions Closes #52027
|
☀️ Test successful - status-appveyor, status-travis |
|
Looks like a nice improvement! 16,241.58% > 2,243.30% (over 7 times less) for tuple-stress |
The tuple-stress benchmark exercises the liveness constraint generation code for types which do not have regions
Closes #52027