Streamline `QueryVTableUnerased` into `GetQueryVTable` by Zalathar · Pull Request #152841 · rust-lang/rust

Zalathar · 2026-02-19T11:22:23Z

QueryDispatcherUnerased is an awkward name for an awkward trait.

We can make it a bit more straightforward by removing its responsibility for erasing query values, and by observing that its only real responsibility beyond that is to know how to obtain the vtable for a particlar query from tcx.

Zalathar · 2026-02-19T11:22:33Z

@bors try @rust-timer queue

Streamline `QueryVTableUnerased` into `QueryVTableGetter`

Zalathar · 2026-02-19T11:25:28Z

The NAME associated constant will be removed by Bring back enum DepKind. #152747; I've left it alone in this PR to make rebasing easier.

petrochenkov · 2026-02-19T11:56:28Z

compiler/rustc_query_impl/src/plumbing.rs

+            /// Marker type that implements [`QueryVTableGetter`] for this query.
+            pub(crate) enum VTableGetter {}
+
+            impl<'tcx> QueryVTableGetter<'tcx, FLAGS> for VTableGetter {


This is a trait with a single implementer (per query).
Would it be hard to refactor it into a free (per query) function somehow?
(Similarly to how restore_val was removed from the trait's interface.)

Using a free function seems tricky, because the generics needed per callee would get more complicated, and that's something I'm trying to cut down on.

compiler/rustc_query_impl/src/lib.rs

rust-bors · 2026-02-19T13:39:03Z

☀️ Try build successful (CI)
Build commit: b46d288 (b46d288105b795d2d5f97fe4d4e7edc8076a0796, parent: fbd6934114e905d3cc61adbbd7e4a355eac5d0d7)

Zalathar · 2026-02-20T01:13:14Z

Looks like I need to try some permutations to track down what’s triggering the perf changes.

Zalathar · 2026-02-20T02:23:31Z

The current iteration only changes the trait itself and encode_query_results, without touching the DepKindVTable functions. I'm hoping that this will be perf-neutral.

Zalathar · 2026-02-20T02:23:46Z

@bors try @rust-timer queue

Streamline `QueryVTableUnerased` into `GetQueryVTable`

rust-bors · 2026-02-20T04:35:50Z

☀️ Try build successful (CI)
Build commit: f16cfc0 (f16cfc0ab2d20a448e430f462e86b78b6b8c0928, parent: ef70767064ab87b0a41400f69e1dc0b55c8d5284)

rust-timer · 2026-02-20T05:17:06Z

Finished benchmarking commit (f16cfc0): comparison URL.

Overall result: ❌✅ regressions and improvements - no action needed

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

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

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	1.5%	[1.5%, 1.5%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.7%	[-0.7%, -0.7%]	1
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary 3.7%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	10.0%	[10.0%, 10.0%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-2.6%	[-2.6%, -2.6%]	1
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	3.7%	[-2.6%, 10.0%]	2

Cycles

Results (primary 2.5%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	2.5%	[2.5%, 2.5%]	1
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	2.5%	[2.5%, 2.5%]	1

Binary size

Results (primary 0.0%, secondary 0.0%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	0.0%	[0.0%, 0.0%]	6
Regressions ❌ (secondary)	0.0%	[0.0%, 0.1%]	4
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.0%	[0.0%, 0.0%]	6

Bootstrap: 496.246s -> 482.423s (-2.79%)
Artifact size: 397.86 MiB -> 397.86 MiB (-0.00%)

Zalathar · 2026-02-24T01:19:05Z

Let's see what perf looks like after removing const FLAGS.

@bors try @rust-timer queue

Streamline `QueryVTableUnerased` into `GetQueryVTable`

Zalathar · 2026-02-24T01:19:31Z

r? nnethercote (or compiler)

rust-bors · 2026-02-24T03:29:15Z

☀️ Try build successful (CI)
Build commit: f8e0d28 (f8e0d280371623a57b0d3cfead4cb65d3f9e15e1, parent: b3869b94cd1ed4bfa2eb28f301535d5e9599c713)

compiler/rustc_query_impl/src/plumbing.rs

rust-timer · 2026-02-24T04:20:09Z

Finished benchmarking commit (f8e0d28): comparison URL.

Overall result: ✅ improvements - no action needed

Benchmarking this pull request means it may be perf-sensitive – we'll automatically label it not fit for rolling up. You can override this, but we strongly advise not to, due to possible changes in compiler perf.

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

Instruction count

Our most reliable metric. Used to determine the overall result above. However, even this metric can be noisy.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-0.3%	[-0.3%, -0.3%]	1
All ❌✅ (primary)	-	-	0

Max RSS (memory usage)

Results (primary 1.1%, secondary 4.5%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	1.1%	[1.1%, 1.1%]	1
Regressions ❌ (secondary)	4.5%	[4.5%, 4.5%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	1.1%	[1.1%, 1.1%]	1

Cycles

Results (secondary -3.0%)

A less reliable metric. May be of interest, but not used to determine the overall result above.

	mean	range	count
Regressions ❌ (primary)	-	-	0
Regressions ❌ (secondary)	-	-	0
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.0%	[-3.0%, -3.0%]	1
All ❌✅ (primary)	-	-	0

Binary size

This benchmark run did not return any relevant results for this metric.

Bootstrap: 479.41s -> 479.13s (-0.06%)
Artifact size: 395.68 MiB -> 397.73 MiB (0.52%)

nnethercote · 2026-02-24T04:47:33Z

@bors r+

rust-bors · 2026-02-24T04:47:36Z

📌 Commit 16fbd29 has been approved by nnethercote

It is now in the queue for this repository.

Zalathar · 2026-02-24T04:51:56Z

These changes should have no perf effects, and the most recent perf run only shows noise, so this PR seems safe for rollup.

@bors rollup=maybe

Streamline `QueryVTableUnerased` into `GetQueryVTable` *[View all comments](https://triagebot.infra.rust-lang.org/gh-comments/rust-lang/rust/pull/152841)* `QueryDispatcherUnerased` is an awkward name for an awkward trait. We can make it a bit more straightforward by removing its responsibility for erasing query values, and by observing that its only real responsibility beyond that is to know how to obtain the vtable for a particlar query from `tcx`.

…uwer Rollup of 15 pull requests Successful merges: - #152176 (Neon fast path for str::contains) - #152657 (std: move `exit` out of PAL) - #152841 (Streamline `QueryVTableUnerased` into `GetQueryVTable`) - #152845 (Skip `tidy` in PR CI jobs not dedicated to running `tidy`) - #152897 (Add optional json logging) - #153009 (Remove `rustc_feedable_queries` and `define_feedable` macros.) - #151558 (Port diagnostic attributes) - #152492 (mGCA: Enforce WF element types for array valtrees) - #152888 (Fix async drop glue MIR bug) - #152988 (Port `#[register_tool]` to the new attribute system) - #153018 (`unused_must_use` lint improvements) - #153023 (Update books) - #153033 (Clarify how "ensure" queries check whether they can skip execution) - #153043 (fix error on missing value for -C flags) - #153045 (rustc-dev-guide subtree update) Failed merges: - #153032 (Fix attribute parser and kind names.)