Reinstate #93800 #94702
Reinstate #93800 #94702
Conversation
rustbot
added
the
T-compiler
rust-highfive
added
the
S-waiting-on-review
|
@bors try |
|
⌛ Trying commit bdf58b78f4a59ac0d8fee64c431d2d52f3ad929c with merge 508e803495f131ee3574ac1e64dc30b13cafa12a... |
|
☀️ Try build successful - checks-actions |
|
The original PR also had some mir-opt test updates, but it looks like they're not included here. Is this change different? Did something change on master? |
I was surprised by that as well, not sure whether anything else changed that caused the changes from the previous PR to be omitted. Maybe #94209 is the cause for this? @lcnr and @oli-obk Do you have any idea why there aren't any diffs for the |
|
we changed mir dumps in #94209, so it's expected that there are no changes to that anymore here |
|
In that case I think we're likely prepared to move ahead here -- it would be nice to add a little more detail to the PR description reflecting the reasoning for the revert and the seemingly fixed regression (though through no actual changes). r? @lcnr or @oli-obk on the actual code, though I presume that it's the same and likely still fine. |
| @@ -123,7 +123,7 @@ pub fn walk_expr<'a, 'tcx: 'a, V: Visitor<'a, 'tcx>>(visitor: &mut V, expr: &Exp | |||
| } | |||
| Closure { closure_id: _, substs: _, upvars: _, movability: _, fake_reads: _ } => {} | |||
| Literal { literal, user_ty: _, const_id: _ } => visitor.visit_const(literal), | |||
| StaticRef { literal, def_id: _ } => visitor.visit_const(literal), | |||
| StaticRef { .. } => {} | |||
There was a problem hiding this comment.
please explicitly mention the ignored fields here
|
@bors r+ rollup=never i guess we want to be safe here |
|
📌 Commit 9fd2c80 has been approved by |
bors
added
S-waiting-on-bors
|
☀️ Test successful - checks-actions |
|
Finished benchmarking commit (1eb7258): comparison url. Summary: This benchmark run did not return any relevant results. 8 results were found to be statistically significant but too small to be relevant. If you disagree with this performance assessment, please file an issue in rust-lang/rustc-perf. @rustbot label: -perf-regression |
#93800 caused a regression in an alt builder with parallel enabled. #94205 reverted that PR because of the regression. For an unknown reason the regression has disappeared, so we reinstate the changes in #93800 here.
r? @Mark-Simulacrum