Streamline QueryVTableUnerased into QueryVTableGetter

[Zalathar]

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 (in the form of a SemiDynamicQueryDispatcher).

Along the way, this PR also uses the newly-renamed QueryVTableGetter trait to slightly simplify two of the functions in DepKindVTable.

[Zalathar]

@bors try @rust-timer queue

[Zalathar]

• 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]

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.)

[Zalathar]

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.

[petrochenkov]

If the trait method produces a query dispatcher, then I'd call the trait GetQueryDispatcher, so both say what it does and follow trait naming conventions.

Or keep "vtable", but rename SemiDynamicQueryDispatcher to SemiDynamicQueryVTable because that's also what it is.

[petrochenkov]

... or SemiStatic, because it's still mostly dynamic :)

[Zalathar]

The name SemiDynamicQueryDispatcher is a relic from before the QueryDispatcher trait was removed, so for consistent naming I'd prefer to move away from “dispatcher” and towards ”vtable”.

I have some ideas for how to rename SemiDynamicQueryDispatcher, and possibly reduce the number of explicit FLAGS parameters everywhere, but that needs some more experimentation while I figure out what things should look like.

[Zalathar]

One of the candidates I have in mind for SemiDynamicQueryDispatcher is QueryVTableWithFlags, but I was also thinking about using that name for a trait that hides the flags in an associated constant, so that callees would accept impl QueryVTableWithFlags and not have to have a const FLAGS generic at all.

[rust-bors]

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

[rust-timer]

Finished benchmarking commit (b46d288): comparison URL.

Overall result: ❌ regressions - please read the text below

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.

Next Steps: If you can justify the regressions found in this try perf run, please do so in sufficient writing along with @rustbot label: +perf-regression-triaged. If not, please fix the regressions and do another perf run. If its results are neutral or positive, the label will be automatically removed.

@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)	0.3%	[0.2%, 0.4%]	9
Regressions ❌ (secondary)	0.3%	[0.1%, 0.5%]	35
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	0.3%	[0.2%, 0.4%]	9

Max RSS (memory usage)

Results (secondary 1.0%)

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)	5.5%	[5.5%, 5.5%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-3.5%	[-3.5%, -3.5%]	1
All ❌✅ (primary)	-	-	0

Cycles

Results (primary 2.2%, secondary 9.1%)

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

	mean	range	count
Regressions ❌ (primary)	2.2%	[2.2%, 2.2%]	1
Regressions ❌ (secondary)	9.1%	[9.1%, 9.1%]	1
Improvements ✅ (primary)	-	-	0
Improvements ✅ (secondary)	-	-	0
All ❌✅ (primary)	2.2%	[2.2%, 2.2%]	1

Binary size

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

Bootstrap: 481.318s -> 480.997s (-0.07%)
Artifact size: 397.87 MiB -> 397.77 MiB (-0.03%)