Make push_outlives_components into a TypeVisitor
#127438
Conversation
rustbot
added
S-waiting-on-review
| // OutlivesProjectionComponents. Continue walking | ||
| // through and constrain Pi. | ||
| let mut subcomponents = smallvec![]; | ||
| compute_alias_components_recursive(self.tcx, ty, &mut subcomponents); |
There was a problem hiding this comment.
can you inline this function? It's very subtle that it uses a separate visitor/cache, so I would like to make this very explicit
There was a problem hiding this comment.
It's being called in two places -- here, and in rustc_infer. That would require us to expose the outlives visitor there to replicate this behavior, or are you asking to just inline this one callsite? that seems weird imo
|
@bors try @rust-timer queue |
This comment has been minimized.
This comment has been minimized.
rustbot
added
the
S-waiting-on-perf
…itor, r=<try> Make `push_outlives_components` into a `TypeVisitor` This involves removing the `visited: &mut SsoHashSet<GenericArg<'tcx>>` that is being passed around the `VerifyBoundCx`. The fact that we were using it when decomposing different type tests seems sketchy, so I don't think, though it may technically result in us registering more redundant outlives components 🤷 I did end up deleting some of the comments that referred back to RFC 1214 during this refactor. I can add them back if you think they were useful. r? lcnr
|
☀️ Try build successful - checks-actions |
This comment has been minimized.
This comment has been minimized.
|
Finished benchmarking commit (a1a2f19): comparison URL. Overall result: ❌✅ regressions and improvements - no action neededBenchmarking this pull request likely means that it is perf-sensitive, so we're automatically marking it as not fit for rolling up. While you can manually mark this PR as fit for rollup, we strongly recommend not doing so since this PR may lead to changes in compiler perf. @bors rollup=never Instruction countThis is a highly reliable metric that was used to determine the overall result at the top of this comment.
Max RSS (memory usage)Results (primary -2.8%, secondary -1.2%)This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.
CyclesResults (secondary 0.4%)This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.
Binary sizeThis benchmark run did not return any relevant results for this metric. Bootstrap: 699.372s -> 699.027s (-0.05%) |
rustbot
removed
the
S-waiting-on-perf
|
r=me after handling my comments in some way :3 |
compiler-errors
force-pushed
the
compute-outlives-visitor
branch
from
447dc44 to
c895985
Compare
|
@bors r+ rollup=never |
bors
added
S-waiting-on-bors
|
☀️ Test successful - checks-actions |
|
Finished benchmarking commit (59a4f02): 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. CyclesResults (secondary -2.0%)This is a less reliable metric that may be of interest but was not used to determine the overall result at the top of this comment.
Binary sizeThis benchmark run did not return any relevant results for this metric. Bootstrap: 700.455s -> 701.456s (0.14%) |
This involves removing the
visited: &mut SsoHashSet<GenericArg<'tcx>>that is being passed around theVerifyBoundCx. The fact that we were using it when decomposing different type tests seems sketchy, so I don't think, though it may technically result in us registering more redundant outlives components 🤷I did end up deleting some of the comments that referred back to RFC 1214 during this refactor. I can add them back if you think they were useful.
r? lcnr